From db20f3a8bdb2e9347b36a9325ca935dae4f27f20 Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Mon, 22 Jun 2020 09:35:38 -0400
Subject: [PATCH 1/7] Add skyline selection

---
 .../accuracy_vs_di-skyline-order.png          |  Bin 0 -> 43040 bytes
 .../accuracy_vs_di-skyline_formula.png        |  Bin 0 -> 42840 bytes
 .../accuracy_vs_fnr-skyline-order.png         |  Bin 0 -> 43584 bytes
 .../accuracy_vs_fnr-skyline_formula.png       |  Bin 0 -> 43356 bytes
 .../accuracy_vs_fpr-skyline-order.png         |  Bin 0 -> 43619 bytes
 .../accuracy_vs_fpr-skyline_formula.png       |  Bin 0 -> 43487 bytes
 examples_skyline/fair_prep_results.py         |   92 +
 examples_skyline/helper.py                    |   53 +
 examples_skyline/missing_data_results.py      |   99 +
 .../results_play_skyline_formula.ipynb        | 1778 +++++++++++++++++
 .../results_play_skyline_order.ipynb          | 1767 ++++++++++++++++
 fp/dataset_experiments.py                     |  126 +-
 fp/dataset_experiments_old.py                 |  341 ++++
 fp/experiments.py                             |   90 +-
 fp/experiments_old.py                         |  383 ++++
 fp/utils.py                                   |   43 +
 16 files changed, 4691 insertions(+), 81 deletions(-)
 create mode 100644 examples_skyline/accuracy_vs_di-skyline-order.png
 create mode 100644 examples_skyline/accuracy_vs_di-skyline_formula.png
 create mode 100644 examples_skyline/accuracy_vs_fnr-skyline-order.png
 create mode 100644 examples_skyline/accuracy_vs_fnr-skyline_formula.png
 create mode 100644 examples_skyline/accuracy_vs_fpr-skyline-order.png
 create mode 100644 examples_skyline/accuracy_vs_fpr-skyline_formula.png
 create mode 100644 examples_skyline/fair_prep_results.py
 create mode 100644 examples_skyline/helper.py
 create mode 100644 examples_skyline/missing_data_results.py
 create mode 100644 examples_skyline/results_play_skyline_formula.ipynb
 create mode 100644 examples_skyline/results_play_skyline_order.ipynb
 create mode 100644 fp/dataset_experiments_old.py
 create mode 100644 fp/experiments_old.py
 create mode 100644 fp/utils.py

diff --git a/examples_skyline/accuracy_vs_di-skyline-order.png b/examples_skyline/accuracy_vs_di-skyline-order.png
new file mode 100644
index 0000000000000000000000000000000000000000..8673899d43952f6cd3a8a90887cb93ee847696aa
GIT binary patch
literal 43040
zcmeFZcUV>1mOZ!$3W9(MQ6#A#QBcW95)}j_CqaUWWC=>n2L&-9dJs^u<RC$U<e(@>
zM$(ZCN=Aa@OwWb)y?Ryk>Q(n2zy7+uuKV44kDjo@+H1`@#+YNS$EwP5R21|S7=}^F
zU%8}?VZ_ZCwkw8wH@xHA;?N2&r1moMn&j}02l?$s@b^7-S9I+$jPd~bkEo^BYXjaC
zakzZVLBrP6!TIJr6U_RigWVllhdUOxSe#7m*<0A!@bjMIJ<G#l?%-f2#>e+>pYz(@
zGvkY!6YIh-7EJ!q1x=TTU%jr^f*rTEe-kboy1egmy`SJ0Re4rgjq3uTJaVTQg~i`=
zu>Q2oPE3>-Id`2Sk|#WbJML`c1i7+_O@fA)njSBINPt3E2rkxGA%%zO!pg?hz+_4s
zu7{i<Wb|iI$-q`mw0-olXj?elav@##*(!3leCY%i*CKeuVkD1ClfdiYeKADvLP-w8
zf#G(Mn!`&4F|h!=+#n*NhnFlIMg=d|q;Y7hcv${F@v~&(6DLmO+VqOax9~TK$DN-k
zX`K@~%xCz{Dx<J!?q{M>c%szyY7;)5>X;A*7gz0^kmb_`)7CQ2b#a5;IN7h=t-CO+
zcKN$;pzE)~%+e*jsm`2g<0cx+k=8wisLiVCNbsvyE=7CWV))G(+s3De%spU)Qolrq
z+P4+jo2{;|r@1druMIq)(Vy&d`AOwC+0l?Uot9aKXJr(zlQyn?uGRg)x;;_pSVM%U
z2428@=}xcPWO_!m+3nk3O>+wGH^p3d@apyJWcZ0*Sy|cIpg(i2`L`1UdxwcO`ONaI
zqU)aA+V52F+<B*$XPM9#dERQOE6;7Y*SWpe**4#muVnVVP1)+mJ)_Z()7RU|Jl$#O
z=!&Z(R|ZJFeExg||G}m=Rnl`USu0K3l8_O(wU8ubFk9-rK9x_)#FQaq-NnJlS!3N>
zIP;|tjhb?#SWSPwQK*5_8wP5*H}gy$vlk5P8wA(Z`?n3p2@xS7i7#I4Biy@m$#<k7
zJn8l8BfCgR&3}A;d`~PgV=+x9QyVUnn1n>|Mg_LHT)v$?>sbD~>e!QK&oot4U&sbB
zrC;=?F6`*MBUHLJkv7qi@U+Nb{Di#Qbm3^*+lyq5zrL&4maQgse^8E=OebS<k-mBJ
zCc19Z`qytYl9UTepBLJXrh2X~6>TncbHAg+c)w!Of-BWhTQ{?Fa+L1geJ`)1Bnaa;
z(){+~2=n%e|M>U!ah*kuX`Gy#XW>Z+9zA+=r2Y;6sZ*z_dfaD9Px0|JMTj}p-ua&O
z@rZH#&p;`gUrxnV-RsLU!uYFdYA?@Lz4jp?3u|sx&9NU<*3i_vO+iW7FwvgY0CVW7
zy!>N%MMeJ8dbwxdG4od)$uX%r+?uMzJ^#C!IsC_uTlTX97v`t(`{|QfE%fqjtXo=J
z4Vq|=9kc3#lT#R3B`HVNS2KJ6(qj^~iEv%&Gz&OewJ=F&Dp~BfUUkIyp1pko%wevc
zjtqn1t92o#8_c3TO(T?(3TMjJuLhk`s%MY(P#<ysZRP6fTGij*|EsUG8}8%nMgMx3
zD{wF#z1*+*o{Q~?1qI)GTt_2E9v-tfv%0!U=g|NID){*EV>pfE=7PpS28Lf@bWBV(
zKYskU>g4pHrl#hsL9vq{EQ&A~gv_$F+Fa}IjyruNoiO>UcnnLv`}+GIOKR0Moth+M
z3R`#Oo<(DzoYYz$A=;|%K9!fTh_8p|prow)oRfp&j8(_Gv!RA=DV^_){cGy#_<#TY
zEsO6gcFu&u!)Oqm(F@iN4Gp~xVpj=!c~UM+j>+}&qF9)K#VsbcvEx4-#FQ>vxZo#s
z5G&ZXBu$ch_-A#XPW5xf;o;$g#|PO{6+?Nu;6ZOK<(H4Y^=DGo*N?a(6n^65Ni@^b
zujfy|rR+?!Y!gL6+rFL_dE3@RB@G9M+>+AL_FM~%%@N^pwb8~%lblWgy8-E5yGp9A
zteX#5Ee1dN75#jEUf|C6Gce4YA|lsdV1fe!15YVOBs=%H%HeaZ2)rh@Z<`N&eMu|n
z?BP+OuCC51UszbUHWn9nh?$vB-m)t<WofEA2_iyGQ<KW(+G3jB^_kM;DB0%>DU%t+
z%94_jZWAd<L&I9D9lzj~BF{Tdw54dkRX>Q0i!+#f^5h8|C^On?O+HOK9iN$L8*DVM
zyWC?RDCYDlIVOfBfL>6cJKr{xm5rT!64qX6M$rV9phe3ic+Tizg>3uuN1Pa_{ih3l
zQZG)lE21;PjLWlVJq6cZFd_~YOF==Q{`Bzag+T9(i~0Rt#Tf-ZDDv)fs^^$BIW_QC
zafpiQ&W|?fz+&Fm++2jP$tft<KIQ55<1zct&!2IJPwQzeP3^*++aE)u`82=Jt#0wR
z=ZzaNPSZUpaOdL{J|q)8h4ZfSGz=m!v$J-t?mAQYBe1yghXt)Vl8wr}&;`H`B-h4H
zGM;y;U(`2gXOV|HXvecD-hTG2+qOJ?&%sk)tE-bfjYf&PW<%6VY)w+J>dZEQMH%|3
zvXU+M;UcUKWq3wfFbA|V4GjiAP&0}RE$QifxRU}g@0F-?)w82>^YcN`(a}+^BS+C4
z!t`mszl#!uAv6)bP4^ZX%rC<eqvh5fgVn5)W7@zwAsF2bF_y3lA*QyjE{N5{)O2EX
zG}?J{gM@^{aiD^TQOw~AwST47vXXG`kXx@=KfgSjNF!Xx`c-7)6#e?*`n~D(g~@6r
z2{+AMyLL_V7UxeE+Lrv5Wt4Dh6HK!1{$SZx;+kN164NIqqOZ7IS6|<@Fy6XY%bm%=
z&MsFS$V{Ns(LZkdOy=Un#m(jZ=;hZid?qF)!LBko5FoTvRR{F8`nR{7`xwF-M%K3A
z$7@rz-f1YB$~|ZJ_WNhG;H}T(M~(HHw6v-4ZJ9<=F!WcC3EgRiAr>&HeMMQ;2RF|x
zF0KzN_8WTBb?!Eyh}WQqgN^O7)YeKIlh?}4fMdeS5J!e>(jcUZ*$v>V3S(lVMMXtT
z8V#~;_{`1Cy$TH-EfTAq9jsieVUHez*NLvYv_zHY>9GFv>%tH;uL$=#1a-Fyq|Sx$
zu?h%iLNpTd+9;Sbx^Thg$IqW8`P~MNnYuZ$__p`Paxm0IOz^-mBO(}fk&zWwd2RAb
zE_>G2*532<u3-yJzWnrXSDsbqWT8rg$RNe?s12lt!#iWx)D+rassYi7O&%t93q(dX
zd04keD2Ks_*Bpt1l)}l+uK_Pw{?v>MUrz>dNJ$wfCutaY{Pttz6cW;o_TDPY&dzrD
z@p+HL`mY>XN$;CC-=Z}2!MeM)yZbsCtDQ)`UEIH^j83z*IH{$s9{cOpoqUPB{9dPU
z)5eGti00aBXYe&a?30<^n`x5U8}<RrQhkD3jpt{^#=gbDxS{#as^xN}s^IR=T(?C_
z>z)ENKR>@PK`ZTqOs;!FU-ze9doKfHKll6h%joES*CtvB3}#WTvhXAv*H6ZBWfq=*
zH9>&soD2ErsPQWpH5#c6RR~BXjSvb5v6r9bdagS~N_y%AN-Vubk>c{DORq$|Kl$#_
zD{*-bX$mG%411Kzr@Y3xIz@Ce&*k1EX73F(JWS+t@2&Z0y3Haih6j)$T{>jeqV28q
z_Hu9UJS&3od8c2^?K(!MQx?9Deg1q&DN0;-Yip}$IOO_5tBO>Le9$QXGr=evW@Vl6
z^Yt|!Z%M$%->iIS0Z`(}<Hu?<eWiRZ9vS83<%VwKY)Y@sTz~-Hfu^F{f;nLgQm^H=
z*vmz*%n$9`7Xt~pXlgU!oXHc2GrEKAyp3o;nk3h+J|Ls%U<=j%L2p#-ln!BZcyu&L
zYU_8Pf|64D?Cb8aF>{1N!XS=_IgG`0q+c&KSdBfw#>VD0=trkt&FT&D(QPV^&=@Z4
zIHB1PCa4HUbYe8t4<_ip1Y9*1Al%msaH74)(dx3StmW??Uxpi*#T=5O+$K0TArtM)
zCRnD$j5!_|;WkC`%YE*dyJ?u-ef@fcib?#t^=40@J?+t>DPEgvDFD*w#T+;YS-Vw@
z?~MF1i@=(%M2a~oprl#0Ho;9`Yo#x$S_vbHDas<FmyTQRvdZM-<;_srRrBXvJ8oMn
zKo7v#v$J|-9);a5klea&GMSZbtxaZkPlZHP4%&@DYS?uBIXF0o7E9j!v<t48M=@1V
zhEE9pU#mFI05AV@2*Cdcmh!(B`9G6g{&{W_=@zYt<4xXMIYM{7$0R3nsDzSWOpdKe
z!s9dE+r>$B$@q8gcpx690LE@jz=CCPvZc-tH7uK=B%*lZsyx@mlZs#o(n73y^ym>~
zBjB9UBBv=XBO@aN@6E;9`ucaPCx~3?A;b$fPP9Gks`}NL<2ad~udQ);`01lZ%|(ur
z`gOpC0CQ*n>bKk|^qG*x$v$uzsXwX`s>){4-V3O<_ii;KLb$vuD*(1EetmzB5amwD
zRf(1i6S;d!=@msxNE^Xxx4<P*^R{C~-kW)DixX_#-rh<PB5IKC@UQ-&cqNY>kvBx1
z*ExIktYv3*LSuwzTtm39RRxZSV((r}xUt}nkdZJc^JbRK)h6!_91+P7#8$O9Sx!JH
zPQ?c4dU*(epF4L>f4+QUPVvi^FPtna(s%FPU7M@pW0MECr4=pd*<%?c<y{8E@U*;`
z=UOh4%K#Cix)i`}VP}mW0v#P*q919DOoLo8xzuedHM0niOh+T}sv0mS-W9{s*FRkS
z_U#+JlvhcZux&<Xj@e|mZ5aY@+{>mU$p;&!%#I1$yvqvR1Po+<x_osce4-;g9H?0O
zV5NVtCk3P{UeE@v0<}Oy6QRL$gJSI_@%d*VyarrSQc{cXEIxGsfQ<hZC&$gfapmR9
zmu^1;g!Ly)o1)V1{;WbtYi@a2C)2QWp+UI+3O@1bn^>3;2#|Yud9BqMZN)x&wiie*
zfq4&I*OKp^gAU!+`G{!4rLV!nvR9kozww4>!UytfIHbh+Pn3MGKYfy!pDv!Ex3#qm
zdi~n`?;PzIa5T52UlssH;HrH@U1nqUP*9v>_F8!h$v_BF^n3Jogwy^auYh;L0ol9(
z!W3@PTl5By+-nud)z61$Y2R>aB)*2}ebo3FB78nRJ|pS*edjJ;zHC~9XFGM~Ofv(Z
zaCo%GU?k>yo%;(d$6J#E<>lqiRvobi{t^a{{0$&7`?(=`{(D1L<^eh#F@ClRvrY-|
z(~C3`w}Gz$|KnILT^R~ex$5HL;x|K(Sne?fRzE0on32&Mk>T3%aZG%gu=(y>=`YBt
zXMjt4tlnz{ktHZ9$_Cg^y@rZ}$=dRa?X$BZLp4E(a7U&9l}CUip}(8Xk2C<4W%Ay*
zYn<QBt<{}JFth?7yzag8jmgY%p-}yMk9A?`S)K3e>l=Za5rnxmQaBpLzkmP!#H_65
z#Ujbop@W1EOEdkptZ!mtV->8stE;O$RtEi%XfX0pj_It9fn)oZvUVKXFCW>`EZ2Ec
zT3Xt1xB)F7lY%xqA~4Z|&u>S$Ppg@mC+p>!zeNP`r0lciNBd6>0j~p;lLD&|5kviX
z+wx5(N5CvZL_|g7iQ!06qGA-&gv&!Sx@>FBO1{OP*(({In0f6h&bg%}Z6KErI#K~E
z)DQ?d!u=jOJ6sbGSUb#mLBL0V_aY+Rt<I04$qKm3?brM2S7BijsUHE!sfhpf_3-d$
z0|fZgtsKxM0H^_>!w{KzdU~FA@#4CiKSXOKhDPSy_(&`PSBaJj5M;#lEY_E%AXc>j
z(LyY!*L!;t$VwqFLC1D29o<w!@78`i<r8q8)(3{I|Ni~^qWQ-2AZ6GVjx0>~UjJDY
zpaH|Qm?X6+55XoJR&0p15$H?wXK(oe(T|1MSg>S6lIMp7xD5hdr=X?|X)v`9@bmW%
zJuf97(6;DlYMR*Rxoog&_wJ%CgpB=Zn4KCdVChb-1N)4~j{w$@`m<-0iiSo9WDHh|
zxuKd2tn_xxf!=#{r{@+I+hG|aaNAevUTj$a>XhTcm>R*U>w|R~@Bu|U3IPzz5liPc
ztxw)Kw&=aFvEcyoJO4dM6SbhS%zeOIIy$|y-xxY+!}V-|097;sYz~M7pnW9#0ede?
z4nwT7>!@8vru$&#WQc<EK3`woF}SVv>?sIn#bwKji@~kqLr3nmaS;nl_`%mX&YuU}
zhj2|>Tf1=Hn=$v|#fyR_wTBp`ym}V-fRO}>&pjFz1IF*t*UkeG;1B}?k37B5o!8Hq
zBvPlk^SObc!9|9!8r-X41B~qFG}SenI{5kV!HgSTHVp{u|6yiHyC@G|%itFi>#`2K
z>g82dw!P(kSx)YH-(jpttozuzg>0ae2shf;t3-+=5RvTW0D>m9F?-Rk$?-T)^EMT!
z&7q|fxVN{OMeM8Z@c09a;x4N2Ny)&xrfD}WX@86w-@;745zJ<{@q|!KP0e_dx)6)W
zj`U!+Ul3)`93GF}-bi}#^y#pJ7$lTLK-YR1#-AVUUxUb(3(qW4)Ls^)3OBbR5K1_A
zWR$8v=1ekX4^^qng`kI2$4;d<%+vwPb?o<C9tQ^Yiq*u-Y!Z%ZmJnSWAAboZ8iccR
zRbR&)QnRwM>I;plo_&LmxVW`GQ#bA~0o+vc>eXkks4Q0nNs`g6ALTb4>okjQ2hojH
zo?SVjxp=0GTb^F@UOY9k<U2Ye549wfXl)ytj1PA@Poylg7C?9homlglBCPq{dymA(
zF?(vldAB!byz}e^W%3mVfMzNIJ#T%(Z?*tSq_(<R4m2Qp2^H7lW#_tgiwD4rf2*5g
z3KGQ|q{M=tbK0Ot6Y_X^n=#CSyjx!=B<Ab+ExSLYBkJE{-()aLcS*M^*CMVxO=l8B
z2ptd^7DCH6E`tuc!+OH_fw%*4Qii}EnxCI<<uTn;h_u*pugx|PDIwIfwJfGd0SC=9
z|0V-LN>y7s1jVdlHU%H^B*nz^sQoXhj}NlUly8?EICv1z9P7^PFH<g7*kWL?p^$0)
z5s8i2N?5vhBu{f^6vzM%e#r{D_+*fZ<TrT4NRxIPZ@CP@RSK)c#>#KJg1Z5vO0IY8
zFjqm2odAUv<d`@hn=**NrlTaC=P<6GXVa?(0rCTqcz`x5;F&xZ)Fd_+S^<HymAKCH
z*dCh|2wz<^`vGMP!=mY;Nr%x$=ZO#9HuDagVG`~i5W$4#TMd{@@V*2THrniAks^uR
zq0=PPwdhcc8jE!3O%T)}y{O;59S=d?zHhC*{wmecbC&@5&<mKW20u^Og{j051oxth
z+Ldpc33PKJ%qk<L!#cW@zPWH4P@!si6G@ve!>7B?LGv~6S{+`4On2zOf!Klqk%aj8
zWQfigqVN<5UtgXS1SYvTX|&w}A$S2WjeC|Qmi5q(4AZJ+8neICTcia51EjwoK!b@u
zl`XzJp|!CGMlR?wdqciukznNA=Y}OujyYngj-+Gux6RCo4K^fLSwBzxn8LT^n5iI8
zgkShb#>^ostOJTno*p}<zt@cs(~Mm9OB&FJOxX(QEYIclEd%FXM+maokg6NQLm_Pp
z4c(S+fq8%E@L}!lgxJF{t-sbiEtfx51W6t#a;+w96mAAnZ@o3CX0A6_TrJJuB5xig
zr>1r)c5nIkfUK!_T@6%O19uY$Gz+lCxX+xKEwN@}W#u@2`~n^@hsk(Yd9rs)ON#)Y
zN0m??`i>+OW<as0#~d9gH{&c_`Yn|&jynqaNp&~wr=v?MbzipJ-rD5i;nB3TOa-hs
zB{cf|y>SL5K?;Nzq|Az|abcQAOR2Hc=a?ro0H)HpT5fttj{?BWi69qpz)S}DuBA4Z
zn=N?j%Q0HoggU-*eUQV{ZrzIYPuq=I?Tn^=RYu6{d(+BlxyJz)0i7;D6sd+A)?Fw{
z{p*I`611Kf#}HA%jBx;MUo^6Dz5{9?V<1c}N^2$W0g$Y7<;tTlA?t~8b)24#T4tK!
zSTS9op^hM_dBuGQaF9TZ0B>Y>-m#ekeSdgnX69|JYUH!CwKX;0U;qmR&(SA5aWgQG
z=`R(56aPG7UO}AHiiFFyaU+fb7ol(^5ZhDc-TQH2GMQEkV~$#KxZONI=oEfdG#w3K
z+@=wr-iSzhh*7oi!*`eaz00j!0V$y+F+2sV;36dECjkMF1YhtL|2p@js>%|wQv0<j
zKJ-uk#L2h(`0`ZHex&|s-^e@N9H_4q-6QZBmFbU<j|WH5#7xH!Ht!qiEZglBP;FXS
zSuqFo)otOMoFm9^ny%+1+!mT)?Xk&2u5U#%-lnHuf-6T`e03;@Hp2*%2_$+0TG6qT
z#1%a-T&8q0=&fWjkJ&BCpnl00B=EKN!GaZQH+Oe#m@^<aOW7L8%05K836z}B;?&jE
z#VdGQKopdqKXU^o<~R|Ll$D+-8hU}Rj^nWqH#tw7kWn9xdiCD8dZOx>t!5~%!9w#z
znnEU2n@RKD-a>jY$|WTw;_@vZhP3Bd>3}X?=O|bFq-6FJWzml(+zwx8b;iEX894+<
zEw*XELvIlz=*O+BxWIF6S}bZT3xO)drQTv^9Y`UFy?Cw;hw2sHOF)$~7}$yRnR4gw
zZQzO5Np|m51>V%2Z)-F>GGdq|iJ8o6W={5(OO-x19NquRXSd~}%HwRTS9a}6sC)A!
zp$>mL->B!qOwlASl=2D$o#czwr;AbaP>K}OLS-_bdQ7L`so)(}W^W=aRO~#{0UBaj
zE}1DnwY9a$%wZ7afwA_2c!$tHl-HUCN(c}k+CX?(1fDSlwUQR3FoPVa2$2WTM8EF7
z57yxTr{dv@UysD4WfUopkWsgx<OM4lQ7kp6jRAE`KqyK=2sj_J$E=UjOyBA7hC~Q-
zu;^9e<N|r^Z8bn9MRk${#ZWF@US)QbC`}MYdCHkF@$Sj0=ZtC2{hpzQa+zt4Xaug@
z`oUMO+Q`Mh!B~C`IAn`;es3b^Up3YqFqx20$7%sUxU1l9oK^4X{GJnhe0+{GeFm_U
zQ(ES!VP*i7MxeCFc_sq_xpiNOE=<}YQ2C(hmIl(-3_`Fl_Z%SwqyV&lbyx%lmAhxD
zD&UwPBs5eE0NO@>-sv6U`+Q|9G6-b@z2pU+45V(Nq&g~n+_`)`5;D^%BcpzbXe)TI
zs8R#9GuGg;=)S&RDp=}$xGSq?<jvAltK}+(xFu}bOPBl7{3z*wRhFyIdwO{2_v<)p
zEDM9+_QwlAq7JbOBJPzI*HEFgMIi)<q3#S%0BNJ}NW02Biy;#rOu@}J9P8@tZUu0P
z9%_eB=14zgTk>8zn%TU5%rVd3P{$dRQThXiO|5%U0Zo92`Z3*y=Exc8l#~=atAFKN
ztIX1lIj56QKez4{l24so`IrNebrQN^I8_iUh^3PNd^89d2Ei|1z6A_mh<}C0s}DM5
zK%pE>JjkZCO-%#<V=-_rZ5<=eJneKnHV9U*rqT^d*Gj5KhKI)ieIkHA(iEKu(^eKg
zG&5r~^4klhkeJ6x))s&yMXu7Ll`y+_SM-P?0{=d4@(TFIO@{_TzC%e-53miuy|1lc
zff;o`?mTqxU>uaFu6si=J#9PhL&nq$7m>)FS<+(NJ!}Cu3E?PMs<3*JVYzpey5}Q}
z05aT8r4KNO1HjTE2^&ytDfaBSyIH)@q9|zHC9nt*1JJzS>5_S61J~hTz^ARStZ3nc
zNWz8irYbFs>;3@ITM(ip9DfY(z&A?15*B#`g@F8tqTgj0+2GsW>GC5W2eUjMSbGXQ
z$h#+rVxEY1P6C%z$vjxvcyh1h7r3YC@dK(<@Nnx4oPn~;R{?N&J!@xt=R-^8Oo@hl
zp5See?}6b$IY|lh@dl_nSPr|^t+_74Y}Sq{E{i2-H6%rQO^hGt(QKQn{gGm*qurn0
zH0HQSCWj{x{AyFalG6+UhLuA?!T``IRMoU#5X21*;EW)x(YP6GUcl{E9X+j^{kfG5
zlMaX@v@<i1{43Lkv;;mVDDM0BYY^&CSstK-(`h<@4wLp!QXG52o$-P-AtAv6>Y{X|
z4L~`@0QYk_d+O<hm2J$`DKwEapZPYaO1Mi)9m51NM8F5zeqMQ$sDN@URFKfP6r|RM
zpEZ{Cn`~`qK?)NP5lzS%4<8%)7d@ANwRP{89mkI^<gOBf36JPb4t*a?UK)KZWF6LR
zZN9m&!5S<ycHzNpi=KkJsQ3i6{by(K@c_mx1}c0|@$DfEb4H~<O>vbwJe(*A_X((5
zln&S@0cev8WI`!mVYU)JrsLI5MwkI^vK^u&B*&}r3FYLY4dLwah~2c6x;w)}FmH*M
zul@S98A>k#-di3hwN^f)ve_vlLQy`|tSJhmo$g*>kgqF`U%qt3&@jP>aY4+@pt1@w
z;6o87kylw<7j|9Vg^#a4_(tJcFpGTla&qK(YLCrD0+Q@exfi)bP-O}gZcuF3dw<Ih
z5m*#t&z{}!P}qR9fM5{9#S7q1(Jq(*bUhB>w*zKQ8z6a9;6`PFwWX=Fq$oE9BhSTP
z;PMD&g61_}Qr+~1Uk;2Ap-?e;&McKtQ6b&%`fR1ON-Km3gZU|^VkXz2{Q*b$WnnTw
z&7^(Zlqi|r|DwW`7ZeKak>~goLlg9hiz!N}^8x^iy!gEMIAhjB`+}O!0Rnr69&Wlj
zKT}3VhE*Qm2#lHPUv*+wA@X=wa_vxw1>UQVN}i4c19tf^DDZT=)76;%fOwTvbIb)q
zhT>Ziw}Jd9;&)Rspfb38%gXPciy-bqF0TMs=$vvwydTw`xI7qP1=Th-cDz1o%qkBV
zJbz9IOO}89xZHO%hk`<(+KN-UbyBq(h3p!!mnIe^si1rqxN|5dDD)_2fnVXSXMeo>
zK}1TcjsRP>f}CX2E$X%5lGovQ94T<9o(E}50R*xxs9B~<O5rT=TFNTiwk1!-9FrAA
zcGD*xze#j6q$-UI@ClS8gILq`3)O+4(l!PHyZ~{+pj;Dx2C@`@$iW4P6co%?tfIVj
zdyV_cJQsnOf~mzAlBu?W!V`4YKr%d`P_+ib;0S_nDOBP{B|ZT4C8K6&2ELF9rkM!{
z!l+^mB{c|JP2qq#Gr`L92R^JVSkOikS5xB5e1ltCYma4~aATY}KrEno5&w|2CO;*!
z`?alx5gV0!YRBjsNzWq5^<VrjMNm~+9Y`&AL_b71NN(<Kpp>-(1A%#8#G8bSOug{v
z0Bbr%`r3EEw4pt4;Ky1hWD`;Qkukt0ItUSgld0EZz5#+qhjE}pB2@P#Cekvsp^imN
zOx*JUs`&4&+m>n}SQ)HinBqQNC<j6%y`beQ<Z~iC7A#)^9yOT|YN!q<4DoL?LE2}h
zi3-f+<?)&BnL(Mh<1T`+gtxr!-`H>0T?lvtLk+7POklh%kVMh|;RYA-0$l>V%*OaR
zBQ;xwie|@S9=~0_oEBHq(;2sa$%FR;VlQIw0ss?`*q&j$HW8HV)#tgk@C_=W%rsh>
z%iktbM}7sfPCNk!|Aygti*Zy=w!*SFJ6n70?<SMQ-mXiTRwZZ}Tv%dSlHLEhNf10?
z2$R7g67I39+@%shq}tRumFM*7h5^9)6681Q?VNl=EPYY-&rbm_AT!>z&lV&BBvNJq
zP8(9qBD3}x^uz);E&#a;4hp(hX)&<FC<agMHqRY%3?M1Zn1Hipv9PjUBBNnygL&3;
zebhnBLWb%J{KUXnh=e};FV^7#tOHzuhXt&=G{Fj{`(_e&fx~dHPRG5=nCd3igrnmg
zPr9g2|63~jf5`_K&<24Ybmj!88aKS*A1zq~{(~w?j_^>pPMwlv_FQ;LmP(2%@&}J*
zfGYUlAnNm`gLMMHF9ahs(9gOf037HYJax4h6omA)HqFgJ8t+-vJWMqos&g^>k$DQ-
zQtc@k?8p!Tia^T~TJD~a?{|bE#a-3$$oT*Uy>AB-h@~TcPK^^|SRNn9m2cm^%@~GI
zf`lRKMpU?}q!JzjK%)(^B9zp1x=r8*m%ugpiW3wc#0jAwo~aZrw1eae?UPDMdZbFO
zDl2Q>yO+&6J{s+<4fud8HN77c1eg9)+%i>^3=Z6y9N@fykY;z98HC1<A3xgIABL<9
z6&wXTU@1O;oycR<qR8hM9{S=%jIS@Lyt=05tKC%enmJ#Dr7bKhN^G?%)_1E69x=zv
zeF3cJI?otR^_Lq#9RMr=?Evz6M|}PKEI^vW$Ak8y4T}=M!s4YuOf_j|lt6;-0Eih2
z)uXiQ`8Hu^Zr%r90Sp)q3gh^NjrpcYC|@9^hfD}xp3>Es=C~Hz!)C>)2Qi$iC5W2{
z40nGAeQbQK43r@xnOhd~V9E2}|0US^XLb3ay;g(_Ex1W|a)M@!5%V2pCMM$$SP3Xn
zLjLmvTF#UV)@3Y88Ye0D|7>PpSh6)JO#hW81@=1SC~;1CvAaKf0hNqBRtF2g08vlv
zIzwk?Cj}Li<7s~|cR+@)u}8HsP!%umm2YH26a{UssEhX?To#iIF5~{ISBGFe>_`<L
zbHqa#z#NQtb949l-$wsUlxmXNewUjoK%@Dz3m+O<-E;&-6Nh;|9~ig+ihD(#2gotV
z7u29zVD}dMlHBqW{pICQdwR}zK2edcw9VR3M&|y30|!{;fq5Wlui-FFpn{^}PqV$4
z^vQ~|&7+S1=mVk6^WOHNJ$#r;P*4jvtW)^G!-o?WCkcx9vdyLIqLIpZ;DbCk4D&xG
zWPeQ4;M7#Qj-l&iK#@B6HoXEjKanD607wPt0IcJXl?lMi-a@+Jl9V)jO2-E!jd{Ca
zxc6i#hEHk__n(wChgo4`3^lp<$#p>UlTZ1|lBcra8>(dB^nlm@Gv@2$a0Y756Ebsi
zT$>YiE=4EcMDqCjBucFDr`?r3d(<yn7{6y~I-bqiITe*g39fOdc^SNVy`tzMnHCe{
z08WvLO7iAp_w4L<0bn9;I{{UUsH`(qKjA=mkFbCuc24=ZMMviO#nX)CSLB|HfXmXi
zrKJSDJZBgjx&-EODk6Hvz+PT9cE!O#e1rCypy+C>3XAx?&u=B%4>+tjB(lb}Ch_N)
zsSqEBfw%^>_1@e0ibs#C%gc`wEDP4EXrx$E<3Dc>ea7R7H)l!M*@-b+?MdmlZz7&W
zzk>GF9V5dq4oV{WfQNx9o$SX$gsEXz4-*T}2PjA4p(5vXFE%tZQ_QhwCFmd#CjIjQ
zPB!jaMx=q7TFu<=J%pl@E%B$71jbs44TA9;)zr$Yq!<=_5x&#{Uy@p3v)EcL6Z^pv
z^757I35`8z2gtv{#84-1Hg^D+n;P?V(~<$uyn=pHy9kVz-~O0yW49wY)>DrSqcPi{
zhbvRQCf%vi%V7%xIzT|b`~$nz+eJLW*SIXmb<lv)$-+5}XKw1uCExZV^Q}7-{)yyM
zusE0Q5)++w{zVLHo^^FE!YU5_lK4ex3ik*V%75zAH-GVOPd)j0_}-4MvT)y?mGZvk
zcJRsZ3tTJZ;Z&xkA-szC($rC^zdrxWWG)~ASb3;t2P;o@$P!5>c*o3q|Cn@LS^}QS
zgCMO&#=^Q|%RRD00&Zqfy<9x6o~;hIo|X|%g<{aR0pDPIx#_toiF`FIfGOIBXz>6M
zJ(aRLerVDw$!2ve<T)WdfY^Lt#ep(86h^`BmlLO`E(J{}qQEsZbgmgj0e22b$5q4m
zN<V9fG_!P03BItlx*dnn=f%exWsKP$hNC(%p>_oo(RI016{%paBHF{+^iiwEjLEI2
zx4{M9%bSoO&&C#&K`PC5ckkaC90J0O{V(q4ll?lmhwpO9k>l0W8Ko2z_RksanztsB
zCnUH6J0)inl~Ey{I6*xq9XI@aQ1w$PDea+P!v})%ml6+X2S?Q#&wTvtk6994o}`}Y
z42r?yTcXemUi!s3B|p{mEeKcSE75G83-hAr)4cAVSwXcIYb#^+y5itBnxmjVu~12~
zz#5Zr9OiGWnA%bRMeQ*%To@dsvoxvXE*6C*9_?Y(zkM+DQ^RS7)0P1@x-aINIUsbr
zt;TY<{kqViR?9w{8xWyHkffSx9loagRik)OsaM<Gbv0Fk-R*!X0sW-Y{X3dNgG7Vb
zHHou0JGMZD5mFf}uX2#1l=7x&B5O*`M~2H1wi?yBLijtdAapb}&#~_#cHE}}qhLC7
zll=_I?6c38QYtp%FJx-?6?wdm=8g8Ozj&D_c`w+sv4K6h!x0SE`b2PI4*;Ct*kKmj
zyW#p1o7juGOQK%9VqGnY{>o$2-*Ojogw+7gE~`6X=r*T#iGwKo*V*WLRNA0g-WML;
zlU~C$zSByzUuPWzF|Nm`rt@EW)igg3zBqA$l`pd;lUlA&<Ae3Oi2@rFEQtxuj88<b
zss~>jIN(*-Zi6D+P*J@mFw26A=VZCKxR!6i5@s@{5m1f#{MiWqTNfnRpR`vwc{bR@
zL+8)H+6vqs<1=^L#K5kkgi~7xBFqFG(50xXEXwDh!e44fAXaX|lGs%H!ax5X=a700
z{~8|8YGkzhK2a&5L|lhT{-jWpaDr}5i$(d!<n<y^IKehlPo*LxuJiHdEbJ}KF&i;p
zxx_5`qd7`qkC>;J2L9W4>zl3|_P6(rciDvEgA!)#YD|N=ayd7{f_|oc+*%$~h$>PW
z9893uLlfHPp`n*M7V7WkA8PaPAUP3cegvX6GuK@iBVDhR<7#Yd(}CQaoXs>+nsd)2
ziw$bOzEDtLl`3!HCv8gkxb-a^-f&yyjkL@!miqPU2=3vR_myvv*s|ywdcyG83u7`u
z8Yw*`1b=^ph-RIT)}%T^qdpg<HmptSKf;rK5W-OH0s2IE%Qx-;!*z?PEvkV7`fBZ|
z6@STjGVYJ;A^m7K{hq%a@-TCBHX&5l@Sq^(sP<oA7%)6o{Uwo3m#y0bFrM$Y*xn)W
zc=Di@>d@BsAKk&eR3YXpi*%e9$E0tWLc|+FiUuCwC-dm=aGG*ts-3J6tra#A3w(_0
z+&OhHU#dghNWh|n6)d;0J*&_~@+2@Y^9#k~hj#}}yyrAwT0aD{!rEZZ!8I;dQ^R5>
zr^}bw>Z;lG@k)}CgHN6)XoC%1gro2Yx^nO%K<(C`?)dTJAVcul!_-RzuSq6UeF)mF
zj)ss1qQ!%Ssbg*qYWF2eHAY36sTDgdrS2NBd`Fbrq^_ZX946rDkFQ?@yN-ggatP}t
zAWi@=X<sH%J5{9QJ3ABD_X=B`+C`>#;likr`|*igCQ)BCA9LOfaP2HP#I(;afe;-U
zs#D%COf%^D;18%p%p0nj`gOnzHSO%|-q>qsXe<H;4=GLu6b+K$6+F0Mpk5banV#1X
zQK~<ZnYr?F^7=Hp(tqU2_^PGkuDd?~Y%eZ*&a?ChyHLQIrhf|}#S~P0Ku^-Kv&({8
z$-7;HpC6g(E4>Z{=o|p>Gd^5ax|c)icuq=pyMH&C`N*y=ZT$IcGuBKFLiZ#9+Hc??
zZwKo*G6aKl3JRnl+8O%eVWDARsZc4c0i>PPrS~a;i%9bTv~R_{L`Ylq*^xn^C8)<1
zod!@|e3*L<TmUV_5g~%qSotR=^EL}=|3mwaySfH@onCJ24UW1NbQwz~6lk*H6mHQ7
zOuVyi8m+14F$ifCWq99m&?=i8TkPxHPC9r{k*1rRlAEe&qnvtx0iaQ&*0)u0t#I^v
z^L>!KPG02V<lIzn*{!-8Ch|Ll=q8OI_ANX0{m?#02Czrlb#y91*l_%@pe|qva~}FT
zI79G7gWAL^zTZ(hp=mUw=AgGo3i%?oNsU?RJfLbuUWmU6Sr(xCM)8tq-N+B^!+6B?
zvpW^aLtHOrDX6@wIe|63?I)6+vdq#Mbb+Eca<4!g+8o>hsNMn1Q$SAmpq!x+>aUR9
z&HYA;TjKqxr<nKyFET7I?c=mhNIWR{NH#@+HgY}CNUlhix-GVV2>NTh1&h)9HiN}1
z34F&&?+q&;5%9roDs?Cdp^h`4Lu1H>H{T>RGd_pKbTz|wodmyM6zTKc_5^Lt6RDCt
zRi2)nsFXpYwF_$x{B!2rW8mRKEd?3FpihrO_3#P!A7325D1PlIrpdR)h_z+xb|<d}
zk#Bf-v98XlCH^rK@n$|g!D3ARIt}ItDi~ZteO?dfQh?;$vQA*hM;Fdx$p%zH()rfi
z+T1!B4yQB0MvpvUWZcBKunXl@B@E{op17QqF;sEk1W8%cNd&a<W)6Yx#xZm}*I0^s
zGeG;PJrakUF;Fs!;-$n=88<WTA_@AY)_sa66s+=Z*v|!nev7|7^D%GIC#>xTi32d1
zJn)G@pOQQtWNmT%{aD-HHG`k83M9YLMm2G9aH;-Sy9DW@5+*=MJTmGRJ5AwR#~lK|
zabEX;mPooGre9D}mpz*&fFxOxYU#)rI8=~n-VKT&CeOY>l_m{}T`Hk4j?cXaBA{C7
z(vzur!tu7iZHWw6DbShdw(+|fb^Cx|?ELqOSD?qU_9vFv7?&w|1`Dv(T%;umkb{{9
z@@S~08+!&kiWI;;s8ZvfMvR+~E4PXOgsH$3{0`tk=WPua=>ohQoBamADkc4|v(0g`
z0WHvDvIvrMDp(?6x`lyam{doMvxJ|P?10k>68hUXUQW($@LUfaIg;EDO<Txa17r;p
zKXC(k`XklU{s5D>CNG6-bxmc~_PuQ2<8R|akH2iUE4hDYXed%R2-%=bCm5aj@3V}y
zVsElh;~%W{>>0&VN_a{>$BxEm8{8=;!LAWlNOD<GVHr7Ho}CS0(gD%2c)CatcpMT>
zz+?9+Dr#EDY-?i`72%*xkpZKA9r<$_ThvrE9jLTpgtYSd-@ddHiGM_1Q<t`o_CRXR
z=V|*5k_9jyX26)#f$0atVLWKC;FfXxSZ)1J%^t&zDxlbaNzXClx!ZlKtk{fX%kqmS
z2X-i#Y2xjx;{;z)83DVO@OT>=8nmDZ5qc6GVTnUsc>=7EbRPQuBGc$x?@g8}*v&oh
zBz5V$0P*9;>OSl2CaB^{rfNbp$0{Sg1}UYEp&SM{bxQPKCY=wXvQsPL?40MX1J(}{
z47&dyAtt89k?&Gz3}kVNdOkI~49zjxSnu8S8aLNJY`jxTvyAJ18&%6Sdhq>R6_KU_
z%(Nzn<%mglP;VgXraI5`tr5z%v<83PfA9b`R1xMNJJK>R=#TV(W`)`{hHa25V9%aC
z_;@I7Q$fN16E`j?a&u^&aNgF#6QE<wmN#A3cac$=2|+6#kEhRC!0-}7#w6l>K(#Wp
zYo?jp*DsjqgIb3Oj>{^G$gMq9%cc%#<w(N*SPc$N3n;L4m3x=NH3UO$VN(o=5VnPQ
zizIeva@n>_JL;%-Iocztp^Ct>a)s+0tqQT_HLZ0<dY}0C>}68cyZJH$w@~0{0>wKP
zl%%Em^T<UBZCvInvxCS~dFtxx7-&F2O(iHAH7J4r6)zujJEzk!`;Loh0$5Rhs<mAR
zo^m!P{k7Dh^7&IGTX^~cck~M)x^#%pW#9&Y&Nh(wyYsibm!>7eQYWUjSIlHs_8r)>
zhvkK@?-4527ru(vg+dpSlb3!y&}Z1E$d(W*X<ho^-R1q4v(zs=2^4$OJaH)biGHs7
z{-d-{Vl{bW_G{u_;O?!~u=&1t5z)_VEW4X{$hUrTQ!mZx-cm$Fr+v$6V|9b1nXML#
z$K*mWDycLyI9i3nD(_y(wP>Y}kDfJhTnkm%ay{fVpgzGUgKJl~v77!{^X@UwK#IJm
zX=n&gfkB;T%uD1xiOSn^HBy0}XjEsgisJiD6xHCA6>$1>4f>)CwbQgfTvzMtJg7HF
z>*GeV9p=4S!q=D~p`ac7X~8q+wyuekb5cC(sL-}J1#?WFAVXkV8Z(V${<UBR!*^$%
z-nFQvZi}TLeHlZAWY1F2tH!$|UFYI$3I=yGFL4O$T9epF&$jBT@nB0+y?O4CV?l`U
zjXZ^2yVL`j8s$FmmppSLoDo;g^b$`l&B~lJ(SSg|S9ikumo#6h^b7JX%)rnv1EwLt
zNmKLC$#atXW|wDaR)le?(xp~Mm{tq6U79)F)J-hC#;X%g5qTXqCvgC|eGv>H-Fc<Z
zoQ2YqzXXL6k$2D&Tw+wS+uytZ-X!F^+KL|zz6@LsEZ2|k;^PtWykmJ_fbRa%iBG7Y
z10@d~z`G_#9>PpG4@Ql8LNA5r&T;<nw`5Z>Cuq4=TLP7oh&N>58Gw&#s;5u`Rr^>+
z6>!kUw{$mCMx~-Z{=`5(CPTECanoh!=hn{)X0J94SM>&j0^$t+y7nYHacDS2av!v9
z$AT$2IX72xd)a$?bkfE(qTC>2>b?7l`(!{iIT5jRa!2vQ5p?xrd8alNg@(CP)ZX))
zoABPb37&^HkI?m>o(1z7axmCHBy?;A<##%0D?TBiP}W~OVdTfJ9I{jQGd`@1A9EC8
zpck==X)<K)`v&DDRP8F8et66~)wwU(@#CX4pi06`zpM2|Rqi{H7Nw>tm~G`>(9#M9
zF6Z(6WE16^r#Tl#0*D<CK3Sy3JUGoqrD-76qg2J`y{ahRg7h{3ES=S*bMIahik!dQ
z$Du0x_8LscP&nXPN>Wq{)UlH23|%guj(hNmPJC#VJ|otD`TGUuyG~zA)M9L8-uL;j
zdKZ7)MSk<bv6#wFpW2|V1--ad@Vwo&H<#PN+->As7iU^;r@Tv2L)9>X@tp0U<Z#t@
z&d2>}$%)DjIB1A~T{i)ujX9X_E!$IrWdrEh<n_R^9_T*eDoAKjdh;Wq0emL>8oS$f
z`KptbsttUSDeUu$pfp&&Fhx?rJ4WmPjS&ved@=!*c6IQ}v5vnni~4;xv!vRR>ZKk%
zoKmD*HPiXHpOqB0bIK4tqYOB$1=Lf)`x6U_eG*Hl1~YD-Tq)&o{dJ<1;26qg5H|@e
z35t2p0}lcBiv4I~I}nD2G^0%gXu`|LEGEOK9=g%V6fBX<a$lf}u@V6Pw{y>3f8<L6
zPb8oZB^8OqSDD4TFskQnbi~(u_Q4mA#Wb}+;|e%UtY89^O;cblhZcDs>87h+@X<7w
z@?aGwkvs@05zu^djz4=+kp9_|f_Uf6Gd??eQoR1NC&kr2ds3YG(`<AQ$vqfu&)<IL
zfc=pbc<NWp^Vf^y;KR31v8ho2>QY9KzV=qGzK&LTv>SKBwBJ-`uIIL&#NouD!kTQ`
zxG@45iR$bj;@s%)bC<ce?QOZa?LUxU_X9n{4OY|UwOn#lAMT}PI<2`i;USYNzgPNW
z5P7p#M)Ftv^B7B(lCJP&li=x->?Xk`VcE}L9^^lQyU)HqW>epvdZdd8^GP2wpE}7p
zY9%-I*9Se%j;h~|cfWo+K8JDfbG!)V_3q-+F1ag>gy&snv0W-Bq}N5$w6Z(n<5_EN
z-IeUR)9BPj67&OGDeQT4RDD?YBzy2nc<U%!G%GROw_d`1;*_GU)hjM#q*tGi42*95
zex;>n67-x^{^6I^g4&Tga*^n`6w8ieSmrHt224svleH#Gp}Qx@@Z}fb%{{lZxqXgC
zeIyr<p+x5qtvxibP2Nm?dgo{y8o2F~(kDeCXOxfq{S9Mkf%_^AkwTWdhu;;pK3r(^
z6)Ag+t(=8l_8;+)ogGv@26E^FIEZAA1qh)o`-td2`YV1X^5qKRPR~BWttv%oP8zMp
zdy_BcQmvlI13q%>W8IKK8c#h%8#8=&^qJy!DqT&mtM)!r@(v3A^Z6pPJ}>dk=i}9j
zP3KQ(ao2D(AJ}vdJ#VxRyN|D#lq%3CO_m}SxK9sn$UOYdZ?Nsv{k`uWzwy@_+9xhm
ztZlg$mu)&dQI6zNzjTQ#A-=wOIJ8)2D~#_xr$$7pkeGwrL$fj#jFc@%@ksWM<2&~d
zKU1DRnU*=HWiVCZyU~^+1CiIhaJoxI0ImRwd3(l>glN3mHmTr~eZa)TInVXe`2?7p
zQI~PD(qfLOOWo!XnT!>MP*8}2`bR=yq9yz>Z+adYISrt^hwLVd&OL*}OuKotS4i$3
z(-7GQx|i5kTgoTBZX$hsh=Zq8qTWHBj0<oR_<>Ou8DLtgXz(psf#(^y=}>DtSo~J9
zs(s>2{mR2GD_9=3!@pMEO-d;nv;T#<x+w=w{pJ$=&6_V3<Oi4Nr43*w2wkv}q3R;k
z3Q=P&RKQxSqrv544hAOFHO86(Rh_ai%Zi(RQY<U`;W)hIJ*l|@;U`YuS#=NXe;J=E
z&@D>0k2u)?Kr9%{4Y~o=0cpP5Z3eD0<aR-hG-$L%FsX0aRQeVXJIl&h7_iB?x^z<0
ziU+6z8*Za+&<<=6*EZLe3E*x4YuL{Z(G8|S@Gt8o_;>D!Fi4K2Z@Vn%ro{f}SvsW@
zKOY$eMeP?-wu>KKA7nzrU?KfdU2O%pFJlVg75uq)DhP^*P!39i9xr54M`jx&SHK9`
ztV77+-maD}1Dyk9rt0;XCNPPNN-H+*f^8?pJ(8*h?R+;}(`XuxdH8WMVEgXhyLT_w
zec1*!n3020u3nKt3Y0f{mX3l=*~(!Cx<b(H0?+Vh?ymSm;VnIDOiGW*sJgfmAmbc>
z;55i<P`*_KCu|~URj8iC$EO00;ts_B78V=?E!(a@S4J)rHa&V5aTUbM(sn1ME4MqR
z-hH8Rk%oQ+uzvJ*p$`1D&Rg%zp{Xy7-|TsQ@ovlu96H-oR9ZwBF8U_5z{W9QP1qR6
z3p~Vsd-u3>Wx1nBQ%ldN708{lXNUzhx}P0wJeC~Ad9I}sYAORHq~<HaHp#HJ(#Brr
zLEQGwhpDi_1J4j62iboL(fhOh$orJ`l3i0})1Zlv@6x~Y2=|$yAAY3+Awv3=pDi4g
z8^Qpq_6q2>WuU13%Qt&qVmzfSgPY*{hqG41OB;UWe6dcwLYZg>6(<=MzM(^3J;^rq
zoNS&w_Us8pPU=(gLW5<QahCm=bS2*zE@Bo{(m5GrpC0FmXgE-LMKU=~djD5S(OviY
ziM(ZS4~=eUl1j&r5BWuy*1w<^cCxkV5jnF)y>bvng*fg6MtnJjZ|L@WeV1buU-rig
z(h<8}B9k6_|F<}Ngk#XP_E1Xfr_%J;`{Va;x-fEn(#+q8aMBeHqA@q^=%`nyLPmCD
zO%u%@T|?u#x`^5sPeM9#k+)th&hJjJ?iP2Z+2A+I$zAAqO8$nI?6n+jH&#JuIqht8
zjpOLU`QM{bp{K6eHD1iHsB5w<*yp)q#U%MF;uIZSwS(w@I)oey<4d0MS`H2#oz}6Z
z*?35vmse_f+3yf78*E`QUQH%LyK_RHH<XL?6S0!K^rYfD`aRw2Le<Q-yc0i})^TPi
z)6U-Sns)9lc;6QP*!}uN!?tRPC88#ji3fJ&_0D*GJ25^S8hO0yAl>5KK+g*Pf=3Ya
z_(c0Rrx~e7Qy=zuC}E4$<U3`Uzvpy=rvQsNPM=nS_=2`QNMGLoLyr!`Ea)e7sAhC_
zDfEXqHA_1p9i0HH0y>^LVD$vUdIfX~LJOyaL~mXbR8o;o4>_Ze;TZsn$9H4+oy)<K
z&1Rh?t~tPMgYo#-Q%Ygx3uE6<xwc!9@-R;`aOie`CGzTEaPs)31wUEAh}L+(<EC)+
zs)IE<D{FFm{AcTEXj6sGGlL02X7t*^IC25F6o;b=+n;FHRtb%gd9dI`8K_}TCS*7V
zhhF%tTeo_aaysvTE`pB-$JhnfEQGAoFCY4D=Y)4-ETsElnht<*0Gb|(K0JN;6m+_o
z4=PZ&c3YlK2ZwSRC=Nvvsc9&b3=a(zOohVU4jUUD^6KE305jOw?=Alm#Rg%3Q@HN&
zs+>1|XglpXVJEV3j0D?9{QA?~5QP6Bih<uz?eb+mXmU*jvu?g87$K+WEyNwhULXt#
z1$pL$!%oGoqWNeKD?A+ZsBdxx#DlpI!#Kd+0b$U!i|qK&M`Z=(PLzY74>QlQolCxD
z;Nt_h;Uox12rB+9Aoa{Q3z$67)E?>6)*iWoLI^1g(h0C!0fV%$hR#>#%{%X}379o9
zfW@V{+n`bxdOq{uocdglfttk}&Jd_say#=ogT;k)ZwC5ty4qy_>V7@5t3sSufVk@)
zAQRSj8xR~|F@R*Rj$ibXqYN$=i{3|7Apl<tYv<~`qpps>{a%FGTBsjF2F^P?+vZ6v
zS1(@}M;cG-D@T0+q6&r)>5~JxkYX@89f#bdrzaZ;)YxbxOjH(D2XnRcFZ&@5;^xuj
zX)>RlO`36JzhA$6@MXYB!JlrtVb~37z^D5TyfaA8rpn<=xs2z@=79g-sl1czWI}xe
zvwsa%5AvfO8QR4BIK`@enEy2YhEhkJTBVzBP{3GG)!T&5%_n3t+nzVLHr5kTFm-C*
z-Xl=njqy=>GpFH_p@znE-wJ}f0$H*1?;7rvIblA2fR3d}Chel~bm8(@ve0rhzrk;M
z<>j>12M!drS!Q9U9(YeX?RE}$7V%$xc;t2#;P#OphJnE_nk>CN_0n0lAQNHOKZ0$R
z7g?W=57^hkC9tTXKhmMA`RkA9Do80pL`XN#ipEKQJm8NrLJ?sf2k*_t7c(SC>Cy1u
zq}7_(sI4+2cmRC=OtPZc^G$jaT_QQFcABOti<ZuL=1IF{D)fEW#|6W&ROk6iSDruz
z8Y)=YXMPPq`};j?`tIa0hj;$0FOexhYRAMg4EIgGbL|&z{=>DGJZRV7$NC!WbRnK;
zDi|7eOr7v<<iy#0QX>ETaI68>cO&;%z-t8PtJ;#6`~*7aKYNp04e5fg;*dH9f3%M>
zf?ZQkS2^^5qV_}ZBF<gl+b?jfVb=xUVmr8+)YQ}#h%0$cQ+mjP4ILBUCsa^TaqhNl
zOIAbcDbVe4OeLK4z%ywQY4War8lMn<B0{9+I7FaJuwPEb6clKnWgsG_ADK)*1r9iJ
zRt8)}i>7o$6}AACFx52FJy8GpEK<@D(*-GV5-eC!W>Vn|r{G_C)DkyjCsNYWTA`z-
zSL>QahENMS3x}j+pY^3efL7pswORlC<x8O#5}p<TR1w_XLEstQE{N6t6MMeTGE(28
z_J1(1O%ox!LZKl7;S6GwIS55vIc9-C#&e;q55jZdR1R2d2u-k&l{%>7rM*hntm4jH
zQ-C(Ku{w`D$XZ%j7O)-7U!euLO5v9l(8q-Kj)8riP(KIQtJaIUu>m@ma{FQd-2{h+
zPj(1-t~n5z(1v}`<yP8z8k=>$ItUXO`_}~V-{+uz3R+;|VE>U<{_SVxHy~U_@c8+O
zGEX@fEt8Y)^KijPAK4i>W8|lHB5822Xg(!nWH<hH-dctcuK#Bx2@4MJg+L6ahS%Cw
z=FTFvBSJshM=a2c+lhZG`<a<AB3}%?Ma>BD@E`tult2FKmE{j#ub7?3Q4XMkLC44L
z+j$B?;jdn?>hd(og$G#0;GX_K9|5L_@zG897$sT!<WDAzyYUa};y<Ia|9GPV{Qv&u
z3Ph%nSIy(opSW~x)amA1`|CeK24-#HIbvA!>vO|Nv-)xG>N}}d_z`8FeH?qjehOE<
z!_lM1nm_8FAK*igqJkYT)QOY5-6u{)PT{0Z?)*GUk1bqDK>^Qmcr>n|YM1Y<_f9-o
zc?MFV?7%;+>Hq%PfzanG*3_t9dp-ZnX4x^2m}waqW5~7t-9tGbJWxiubM2&Q@Hi-%
z1Z!&}rjmuxwM%Dlah)SP7u(q?lQa!Ihd5<)?bt`ls=a`JOHV^EeeSk%?WElQj+*OB
zyF4zSgcvS;CI^Bk)iV@mu!?^zV-Riu-+`*G1_(H&4PlOr^I#j(MxE2J_Y#DdQs`8X
z9Akq6dH8gHkNht&kaan1&tL&H3(1L{FZ=Bu%lYTLjnE5zHe56d0IL5Rfd!)C#jAtK
z{{LNIK|PyLbS>@GazU+)sDIeVnS|*B$Vf)cA2T6tsZLJ#{E_cqi-=iAI`BiS@t|09
zvKDlBD7HIyv-9#cE~33B(qSuvf(~Ouy`U9-r_~iHflOm(H+FgdY?c-&$$}k2ev2lP
zO~`H9^$l=e9TZW7%?F_^EnUKWDLFP)2Gvgu4Wqgxp$_f5Sps+kwqNM%?uGg*4w#B(
z2y9B%5G`c{tV$d4X3*%Nw;MIw!2vYkmuSl{=v(K6-O-?J0qvEQS&Tvj-*IWezXvno
z06k&T&(^lHR*CmmTz(5vQ*A`@J)75%ipXCZEcvK|G|;kXI5Xr$1P69VlFbbs5~dz>
zUNAx?!{z2l+8%j0UB2xF#GebcDBIpzE-&r5DB$~aGm??oAG);%*BGKjzi?}+PS(eH
zwb4Y&t+0_$jX;x%Ec6Hbh8;<cfFyYC96x^mY!Fa9l>@GPV<6~~6UTYmp>^xV9Wd^G
z{rWX|aTT2K-AB&dexzn#V11`6w`L4lRYKh#J+ymYJ`Y4Z;JQuWm%7q9=C-1>w|~xb
zQa0w;zB(LP`PSw-w~)|MNmC}+VA0-cJcpkmOng%Ac|q^eKA*u`-lM8=>1oXHXl7k~
zrA|l-vxE@I8Ei}8D)P||gkW183AgQlzERI!#?06b*WHV8TQaNj<oztt?wwgTD>Y%y
zhV6|}sUq02K@eIljGXu2<Wc7(G#Hkx{Gdc-Wz_SVSv=&N>^OT^GH`Ymu>(o!<hcQ8
zx5Y=);cVeX(>#66+K0DE$UV>|r(kMe6`eU|q^p|(b<WiSoxY3d^LybEeb+|!mtDZw
zNb`_92Y9>*-I2jJkF2fy3~Z0RObJp-BcWG@1QOxQ?F|>oO(>S86&2~h?lzLs!ON<X
z;tya>zuj{6waq$kljPRX*#Ti_82~1Lwg&`>18vEnsjJJm(=G;SYZ2|Y2xJGE^U=;6
z9zDiw)nB$6(f+Sd-dnDI{{CrbZvfB?U|0j)rzHO2322`{Q|RgrPQ-5fokRW8{N>|j
z*_Qkgwn{|}eXvE%=D3Qtx}#$rYt5jYIrIQaP9Mkg#Vi@^cp<#EX@K`7*Ktx?z6Gcg
zqFrOqcY-e6qlX_e7ikk4(1(CY6$2s<D)Rz_DaaGU-ME}i5&<nVB!DN?U=2|-hzv|-
zV7O}4N$ihy3x*JLmVU4GbZFs-hiSf2#ElJ{Z4-l@27yPg@+_f=st8JPChxC#^fF?z
z<d%#BpAAHCZemclfs5CI#|7&Y?Lb%#`}%l7hBsG5S9AD5ojGJKoR~2J&HvqC{Kp*k
z-vwnzYuJCGE&mMzz&HL8CI25r9RGW^zl;0-`vCq+T>Af?S#n<=8X=HJA20^aAGMUB
zy|;c$+O#QxF$rS5rk2()gd)dVXZKdlI<iBgm_cm^Xd5&`&qb}kGG*vMK|4;O)<`B#
z#tIFPA+LzS#$-B%{%VIXs)p-}lsk%^BiQdyn<`{gln~HXU%*cA@gRP4K-&Q7XsoM4
z+l=$Ni9h`k>pDMzdS6gOJ8Xy*KQUpsv&BL?l)!g-kzuDa*x@7*Ii6tK4o%pC)B&75
z;BAtJkT<l;2=;A6yJw&V0kqEwv@k<oBVisk?+Eo;Of%vH6$Mfd0dn+W|Mr#+2>EDB
z8xYHK0pSp~py|Gigho6D>P2Xy7lOWhKpG9&C=9iJLX!w2Uq}E2UbfH)jr4rjHWKx7
zOjM#yA?Ub8TdV$6jvRLKLfca5NWn4R0Uo#vK0MS%3ww|sBb1?9Ajs;F!zg?I+L{4|
zGvVnL73LxfV~e)Mgne97VLV`K<x1DdO4u^$|7!2c<GEhfc7GbwT9Q=83>8h1A!N#s
z43VVFGE@=~k$G0Fq)aVKg~*gKv{Et?LQ#f{5yHwWWXSAXkG0pk*V^yiXP@_T-gEvq
zpYzY&OMdtc&-2{(eO=dm-R;s(Nmq!3)7Mlq)LV>wE=JmDFO>y@(st*b0H{S9(M^<2
zoUT!7*P(AF89qY$cz9IM1;sSnhNldx*+g_96K<sV`r7n9@-RtT1uE-1fq~j6K&N2h
za4Z-AT4HSMPGb0m9+0^n&gG=**H`KBGB$g08w|G~S!sxL>PUtxgS0O{uaVAsJh374
zBE{%_NXP|vQ!@hsEJy1PS_dRp(2$b<4A%2p$WL1`O#Mz78Cerg@p3qixz+V39|Rm{
z&L-{NvX_@vI6eQHiV3a_IbS&<yg!&-77)WiJp1<tWMmxdf#xMs5C+pdkT{mZvci9L
zX}LFr^1CkpI)_<<U>Ua7*aV$Pir}sNI~RII6<u6%x@^S!(G_NbGMfBq$l?O3`10Ph
zz_=R0*Oa)oWuuANAA+%!G!XQznY@BR&U8`VBR7A%TV|x$<$x*!5KdXKh7}>NrH4Uz
zK170w2;bw#aaggQ%aH^4fB0~k3zkt1{%C_{G4`B|KgaAN2<E5f_MS%qKsfw?#IllR
zw6@ugXt#e7MSc)3?!2b9HY=fn1uYu(mI`d>0Y<2ZRfbgV3iG{t_sU^L5rB~1oU`V$
z2l56*_z<1a)U+?2JBvSX_7$bM&jB_#<*+_J2LqGRq|{Vl;`G)7jq0Hdr2=U2xk%i}
zL$M`oA%yBdqEZeosLRYg=N8cBCXlLa89C<&TAa0LuJP&0WfDVQ7!>U{0rQ;RL%ht1
z)j5{Mxx?@nBC8eu-Q{WwZJ7f*F?jIdj)UWtT1B9*uRw?Ayj2fU?DDuEX$x4J&7VFg
zau#`w(E}`$e6(GskKE#LeKYfg-|LbpoEMy7vtclTR%THk7Xhdx491*;{fRkI<|0f?
zSegx;BF`~bQzAn{`(4v>b6b7={e@RHT3gZ56hD`^`0Vwkr?x~NKYrZM3Tb+tATW+n
zfrw&zNHo*?q~Xie1ErrR-hqJ?nwAbNjop%A285!Be!Bn&s~$~AAe%b~!;PUPM_5;~
zprAa23NrvMAIv1rJOx<xWeqHc{b8og3IHi>Y9WFp+H2g;IQtP97!UyRk-R!6Cj9E)
z*wbv+*x0BjCx?u4ECwfw32f4R&KYD&GyAxCbD)YCM<-s+G+3*H!0<0_-MW?8Yrd`1
ze|UIU-}q@dRc=z!oAop_LU?z*7H()$3ji*a7if1eDGfgYtlX0L?xlZsD5MxZFu@r;
zmEH&0oi(~rCu2RRSvPBlecpoRJcmU#_~&9}qPl@_PC0!VavUt`<j0SHi>){gz<x0z
zB;>b4Q$6qcHgX(J!fu=DuR98RCpZlg1tNTWiv0=|(Wcf2S1E~tYI2}pBm;~Xvlfr5
zsMz<{?zs2h!4TXG#QH=awVQ(PZq6Jd>>zNEn8TWg7>*E^Dxwd8n^ZlFNn+surxHw@
z;VK0-5Vv(WT@kfr&AhaQ<ae{_9G|lDr{v^H6l!QFH7T<OMoH2UX}|uezC2&FJk-h$
z!)c#$1mYtvB6BJ9*_BRdG*$Cp^EGN~^i+S%;zrNnz8yPa)at(tULAfur0e1$Zuh!N
z3sh17im1Hzb&}BO5alimdi<8Yg2*(KVParN#@0=@@+<3}%RwCrDme@fH615kKeV<I
zu9Hdo<&EAzgBrbMk6Ock12McJTtx5r=3!3o5>N#bIjFT=kR|u--4<l^6|#$8l_xG~
z*b6+$tcRU{MfcqkP_a)C&xoXiyYInU;zf;nD46N?%f-oKF!5+81@sBJ*d495`E<Rc
zeI;sPQB^ir8U`A83LTNTBW~Ep#EJtbXJ0|)moSH6IdV12-<Ox7;79P);z_ctNOfuz
zmLd5yjqhj*)C>b@XV639X#%(trSXaynM76Mhff$3oZ)yDFriK(+zTFVBM$i?NBZ`#
zl9Cc(+iv+4?{^$^+-MQnHP<Jztd}UggcC#^TnRCyC&FK%pSEm&3aKKg*Dz&5NlRP1
z5)~BQGpm&A3-YVb1xWm#;gdLztRo(2#sCNsVk$!vF~mL_TlF*!a2WE#?0{IgTYjm$
zN3>O#Xabk4FRWdjIIj%OWg*yQ=<-^}bvoEq9mq0=X8^H#AjCHwq#9iBB9l!hV-Yls
zx1OH}7gi>?=ZgCh{3{yah_YK5rqW5I&48%Ja1q2}o%mcTe;{EM@h@w6=IIF-VUcVA
zyqptg<z~{(e!w7G9dwh!2$^a@oC4r_*F^lP>FHy@$!8n&x$h|CtDtrVd0b>mx_;rT
z)V#b<vz&xiJBM-}c<%>^#jnKd4~F-|+%W6YRX<OhDG^Sc#B%~Q?;!j{i47IWxuEWR
z$~VV~U9Cm7Bdp`}(zFZU#YC<<ZT=9Z84z<jjCqN}LL7k6N}ciYoKEu+<`ZZWpPJv@
zNTj~K;YvwQ*Ty}eKOmW84mbkIm&5=`OGk$k>#(;v1^pa>k%_jQM04%|tPCw^07$Pj
z%6?+>j(jM&KIaLE6a~|*mc-$Juv@S~OM;mPAawx;Hp(O0m^m95&*hyU290FS34*Yy
zVXl20E|p6eISAn=;k9dl-cfCIi^exSHRe}NKXMIi5%4csyuPTHap5+G9}v5|cE>g=
zQf7bpI^+%$U??d}iJHR5<qHFO3kcL``}pcqRKMlq;TWWE?;lCEO+KSCHrc_jhGd%n
zMfPJXgUM4Q2M|)W+ztPrZfHn4Igm9Hy=-$kl?^@7N)S+u%*<&v9B;kO>-`j9^s6s0
zJxdH^6OSeP2Ihz%=a|#IuCD!D1gHH;j50~Vl7Lcc)8_THnYzy}1|Z*qyJ!mViO30e
zU@_@MORmWAEMTy*Lxg95hNJ_?y)71vkGg()yT{Ta07|=@BgPd`a4}QXF|fT!Qd)Qz
zS|Y-e^9URVT};#3TgYOQuo8&BS9(rwIJMODti>pt6&s?Tu3p2MU=u-LPvX-<jvL$@
zp&Fo@Bb>^J9Rijm01aZJN_q~+uz{!f-|o=4(NcN=L_{w!Uz4r^m`CKjAUnyN41~=l
zIK;J2lK~~^d0#P82*Y`*T{tKU?3o)WgFzIJ`*69;dRPPkIO6$TB_t^j&+#d{PP`X&
zSeUV8N3Qo|agF!!BWbI>=V&0NT2)8-;bD_YoZ(43GNu|dCY_42`-1|G;1W0U)WeE=
zSa_0u)oQP{&yMi&0!oB;G<L@)$-$x6EWujrBvXaBg;<Q{(BKVR1U!wh8h^p5QKPq*
zM2_=QVF-T(WYEs-+v0t?a4&@|(*tf|(Gle_9X2Emw<nxwuAkXoHP5tBXF)nLmrFxi
zawapKZ{}SCZQnq*m=f(CU5m)Xv8YRCsBMUYa;8}g2U^QXp8<*)yS6@PRMNLK*I<wl
zWGG__I#kh6UsBY)*7QF|`eX^1nH>HkkH`2H9xkq<SV7QMKj9W%rV$03Pu>;r89fnp
zu&knhnqs%|)4-7x{vu6r`CE&87SRgc(O*ZII_gcWij+?isVT0cChsI1#3oVbH2W1?
z-$Ps~0I4Aq0g7GaY-nf^DHFRCPOwJ!VaVZU(4c=WB7)bi?DUW{QZOAPo)6r7d@eBi
zknjk#&wWM3V=2eeCX^bCaZOVdZ04_iLi0J2h_ww&KfZge%ZrCilU*ko9iu)Zw>mR@
zX??A?vU|)=)JxUt9o(tqR(ti$fgGSJKmV*BMpQOdNEgo_^AhrizzW`na}N#ir0osP
zhy=n%CwvT*+x!;17v%OT!NjQwmBaVoAaVG+5gB@)hbOiJ@sNr9zV5S3l7mc4(=2B%
z{>E<j##BAvZyxphE(yXtrK4g1&PHgv;Ze@TAQ@q#E8clMC1FnfN$+9cfUni#6BC;+
zc~O>b=zrg05xETdWNdq$p^Do5l9GrZ$N6|pA;8&KHLks3X4Nj~(b2LbA4P5J0MDT(
zNlC|oCyw{P_)Q1?Gej_fvT!lKf)lzY%UwADPWNFfOt2mI7c-gpFrN)9u`ozDjTGY7
zAtz$hYl!hfgB=BV#E}RluPVVr$%X<d8R0jbVTA7jbVz?21r>`BmD`HS)E2H$1H@Dd
zLUA%Qt9d@fg&!qQ5)88Lg@q*(|JP+S3Xu9%3bua4)dwzZJujDGa*=|di@NGKX;(r{
z$=@nN$((qgFN!uk3--+F{h+E3zy1qx6`b01Tyh%pEPnTZcFV2-aGdq1r3lVk0HZ20
z+#(YWtjZsNb7!6_r6vk$@V8T}*T`@w;`)R68AP#y)|x1!y9vb26fg7<aU3Io4$-Er
zlujlU$@L&&y$*4*3}AIAHUTJ&>iWP31gV{+0z@-R3y(GuX$hZ8%-B#Lk;msc+21vt
z2fO1YoS(@^FDtRwNPiUOWf&P5rNx~<)FP%hBol?m0})CUERv9iG(~+V$sy)8z4@!7
z397RLH1UdhVb2#&8#;b8CH~B?YCfDbO)``~$**_039O1~Fg8-6?lU$d$%I>2Cyvkp
zqwmD92iMww{(UT|CrDQbh=2$agYgw)m3z%r8Efk5!mi4Rgo#`fLI80|45Hz1gArEl
z$jdfZx7xw(Xn-shl2H+P4kiUXfn{b23U8#;!x5cn1NHo#<1{4P!IO~5Jq`1xRFGP0
zfX?CWo83TXc9<%O4sVlhD+CO5v*X5Fhdx<1!Nbs!WG%=>lj@fM`=>|ImWqvfLa~B;
z;QnJfYb*O+K>|BNZ&D?&f#3_o_k=X9;dz;(gt^#Jxr+uYG-87Y=H!rkj}^s?jxpzY
z(*}uA>_WO#!k>4rON3gAg7MHs=4`+JGqDAOK6;ljtnX3%rw#P%hVmOZOt5|N><tmZ
zgy6Xjq7mpmcchrD92&e%gOwqi#)>5#ypDruIQmdnG-A)9T24Uv-4E`j0bs%tJl#vk
zEs>Y$T)LFpfdO)Mo>`FIBg50e(McSuJ;uwJf*<$y_v`n~6UQ}FC3Em#-zCmuEQSsW
zh0guncBjZP@TJf?zwu}J@Nqepk!9fWmA!O_$GlgJs*#f}r(!LVK{zB<KJu8>tFLe$
zD_%E+h9Kg0VqGB}1pvVk!HiYtqCYtxrW$oDfjyA%R<HxH&iTSf>=+T*ZRZQX&ZASO
zc&Z2(e?J21X}Hai1x_5xNCylQkp#GbW`SUJ#Gj3@S&pgjk0i017&Q{NO;|-83nn=g
z@vlL#Zs(Dsd>WF|7neUCB%}}Wk`J(4c_X(<R*Kx>j<?=AkaLw<xdALG99Yly79}QM
zZVgZs1DT?^lyls>SbQLKai7wvnK}?kytM^3l!_tqhA9%A30|6!OMB~5hnd&?u5{_X
zTdumI9fwakBs-XWd!X-hagkQOA@^0W>eW{_aB39&$PTy2Lb3-CF-Xe_vajzB4=54W
z^qU2e9Iin$_iB_(ADD1pqP*%M{~}J9<G%~j{{g#Ra~ER!{lA9{VsM6(!LSWmn|4m0
z42;6eyg%5Q2k!4k#YqehYv%JyyJ`UyJe3%w+b(u$=<_H@=S{N2<7S1yL1PYzig=ux
zwG{;^dFYlk9CJbVL9L7$D@$#D^?J1ii5W@7XsVMLEA96yI(gBV<Dby^U|m(KsGy*H
zm9Zab1hKFo{=});Pd1d^<%aK%0`WEt;*b)CYy);WosEy~-HQjk1e1R|CXW#`G$a8<
zA(u`Z<8<arZqg9P_+<14lKSv#!FMC&YW@;*ejX1d1S_OtB!k3DL|a-dd=;a~bR&dN
z;+p?ydKu>sF&Wy6GagUd>5LVPchIbI+E88Gy3CD6P^o|3_ZMzIxLKQR25UGIkYeVQ
zP9v;~dO(3I#$#AOMEFOvJH(wF=AxubpTb@+Ad}9pkEY-nYdF?|TA`jR=x{8ggh}`$
z@#}wqkU(ntAep6XVq>`<KEot(0w$`ST|`EF33&if-5?(#c9nSQIRU~bMXEE48Z8=8
zFQ<^20jL{JC+n79)<%0SBsCHDm^XKKG)OqJ0a^(m+7P*9q7C`};eiUaP{`f8+m#XU
zv@!hwa>7Yeam3&@gd1=VnVcbU^*3LVzavx~UqL(pGxg<HEa5m1!CX6Hs~!Bj-wTCI
z@M8dgdvS;b+mn151ytXHCwkKHW{EE%fKKBv4>ox4p%x}bH@7mPgoqt0Tr@ZvbFob|
zhCVaE))>aP?=qV;AXT>fDHh<>$f}1JAJSX{So{N^MYJdpC>ktpRu3E6Jdeyf2%;tK
zB|)v=5~>fI{aUED^XKL$tm`wsUPnvMZSh*~MEm};=>?DAFqw(^Q~3ijNpPkv#fz|l
zB;L!-4^&J|;|Y3!8i;r=BEST5&nxr_OyivpKUy4Xk5f{3U}z~%Rw!ZzF^nUEa0y`g
znAaJAa}=c@!2yu!Pc~?r9DoND&w&G4$Ps}tcvdd~-&Q$xtQMy-w=z!b(;q)_8ByGC
z=qshs93i*<C-Dj`nS@DZfreoD<4Gk0=4Sd-gtzXBJZg#z6~l_fGEQ&XM+y>vrP^rJ
z9NHPd3TcN1M4m%l*y2bZ?!&X{F*XnZsz_|a88Agga3Z5Q4fA#~-vPK?Han4oBhG5!
z&5^!%GLeCRT7cFkQGMs~rHa=h7h9npexGjEh~?f)F{696E3WC(klfEyPjSDu+NQxA
zc+}Gy1S1F<*hk@4F1{EaeB$$S$OUFbzlrfi9jPW!udAqWfB*so{lE-J;hFpFRnG58
zo)+_NQHy#$d2<n*gXh|vCk{8pH|31~qTi?ORp719lnZTxb8zEerl0x3TP%V}49kE&
ztN~`f)hqBvP;L60<j|122|0f4XN#|9KT?YQfBzvIl0-GGdaye8^P1w@MM84yQP*l;
zb>?^?`~CsXw-F0;pV*OVfeom<LKQY_%0PU(rmy1#S)%H(W5<XO8W2C{4i8if`xK8v
z1Gq}pV!NiU@j$;XD5<FT)X*Rd(?V;rZB5{8Hhz6xjBO)2mEIqhTIKDd^ZSR!7>Q{7
z|1IWshs8h2i;$4P5$i7TOGh|f1TivpgrDlw@=9X_QQ|=`;tC^y7bX=O$#%koH=DLT
zp#HVUK&Q|ZYl$k60DJBYXDDO*&`qvCg7F^ONORHB&;ZlS01hdA(&yS%fwUFyeDGt`
z+r%@ru%CxA6G>!pl+PZ2l34eH=s_O{EDOj{J4v880y3gZ>Kn0M<!wxKpcXKZ*i0-Y
zT|Zx2M-CX0yb@lbq~z1`_=k(nVF|HfCi=xjOtnA_$<sjR4+;^9em!i(XNYF-I(Akj
zOyHUwu~zWk?b9pnVX1mt%EU7oTA`H&cr3QxTR#5K9nrFh)oL08yZh7cBO0C^MA%7e
zxKUzjo;npn3|TQ^)+!1|kvVdg;KxM2LlQacc#k?m${5-k@T~rgq&jM8a?b(XUu8jf
zk;Dt-BC6T$o=2;_M~T8yPV+AXv^X+$c!o|oW>H>^3jXu<i~m;f17z7Fr>|ygW$bl-
z0r?IW#gp$-GdomnX1_Pr?YFaLVwA|*_10p_u!u4vf+T|yV_6vzg^s6;gJzPH2-smz
zHs!7e9=-8iyw#gxZsa&p&=W@4RQabPmSV$!=Ox>p@eley3ZE}jwiLOCq}BNBkEC_s
zUuRf<J6Q045QP3`sog){3s{{r06GBX{+IRwB=g&#X!|eh1!(%b@2#)Fwu!~;#AZi?
z1%lRQc@R>Q_)in?*a2pY1SaVCx*NCQ{u9NJ?)ZJ)I*fCOCxbrGx9|Z}4oMAoc;9)*
zcK&527}TH)&$Z%m9?M!w*`5s#yZ_X2P=gdL^Wu9}<qz;fPD5r(gi9o=ZzEG^02Ks3
z#tJ7GUwR*QLEEA0>%7h^EKh%xd0G7}f7^^{1$SBePu#m0?7Fj)mG1w<$#O!G{^=&+
zhjIl!KX9Ph9C%1V{=jsCs)_#Wp3{n4rc~o{m$qq(+4DT<Ewhi)DAAAAj_J@;WD4Lj
zG?HVqJQ2|nudaCfAxrq@nZ=$4YDu-L))jq+`zrdlcFemhv=q9(%bA$6`#PUXm1NZ`
z8}$5q*`5B8kpPH^2$>q2mnR9>U8*;{si~<2HvjVS`A|@f!xy9MCX-R9LJoxz5=N0y
zgY6B>4n=@1Gk_lpM4>H?1_@SdCVxamdHL5j=%ksE?KvA@#s|3HTxSfvjg5)f(Njr9
z*}G0dY<F#49X&fcyGb=3)}_fsq`t2~gzJ2TnO_p+D~>GL6@?bSQwrrSX<PMPQG#+4
z-TU^HZq^k|%P}I8{Wl*qt8HYcYhtWx(RFTXyyu`mk-Z<D!AIfI)6tPPGkg8{Y*+Tv
zr#5W#KUbORFj4NZJ4JU1AG++k=Wq}+8*juq%HS^)+C<wKn^qSXx%^OT!<v?MsgKIL
zb&qW6rOO>`b)ODT8`#-(9j2&9tuYr1FyPx66{P#?cHHjRp{OkHO_aefJ~kWboQPK&
zDEX!AkhT8xss;3>XQ;h%FVOE$EgzVkvA*;+osSaz+ox+jZ-1>2r8FF%V%=TJ(Fy<2
zHRhok+}t%*V|*tI@P#|m)6?0I5~$71puS^ALFR{qy9^9I8<T;}5o)c2gHz3EdEP&w
zC+z~`>eilzJfgzFTQP+?2zA14kRMDSN&FC70&;qvM=f3k2)k!`_~RDnC3M^udN^y6
z^kbhq`L!wbDgSxrnMemFIpGU(-U)>Pyws|TN^dzjHFHo`D7m@0@fu_cpkIyyxiL{s
zMtz$Do>zu2S*g3j<Z{a<+wRvbc1?z|7uT2%wteKKe)nbPMs}WsILH3GW2#y#%*?*9
zr3&Jxc__+HO-;Q!|K-!C{76ZdIT!nts<b$I^qi;|6t+nYKJ*It?4O&LXM(^ih^6y`
z%%hq<uBaL%d3m_>EErsI_(KaYIXTJgv545Y+e@u|_ETSX|EHX~39seRmWldACFfzq
zS+Cny>7$dcq@~@wv~5e|E!L(FA2wq?0*lnw(b1s$8j}K}U%v)oY{hRO#!sF<7n1P$
zxXxqt+ZS7~O}iU`ul=$h?qGIyyQ}9+Y4&u@(r~2R`}a4KO*rySEg8oesCGpc>TNg{
z2@XnOa(cQPvR=Zr*R{8^gPM^Z$OW%ZankeKFWKl)I+zrED#zel+D!*0b$UFahh^Rk
z$q5MwEf4}@pgj))85>|bJ6^+F;X~G+n2}a67o=mi0@B|`OH0e5eMT~O+EJmc)T(YK
ztlZo!`#|pWZu`1WskU62q)iDnvV47<I|olIDJj|8+xv?{Ntz&7fphT+%l;HAL$|!x
ziGgb7!$!K<pS~*^+e;gU&IRW+Ft2HRjpe%c25sIUf1f`38x`wJO8L_Iqg$~n)v0fF
zhGys0Bv-G#cifxW`=EKGxUEx^M&vr42b2?wq3`auQ#3A2RKBlsuWm_~dEX~o(R(j<
zNj4MDobH$QjBRJ{{KofQVadmxH#BEEYwQljcLtkgw|!&WB**aK=8ZX<+yi+IpDC5x
zPM6fq`mlc9C_Gh+U-=3+Rx({;Yj_*JkljfYK6kdtnsWBe`r3wdXB=05QNP{p$@6S?
z(p5(K(=wMo(uRCEmKF8$+A6Mg&z%|D+{m}6aJ;<I6f?|<O4U4nLLKdxl+YjVwTvWg
zw!F@Y^W>Fg(ras(d%9a1C<v<wa=RSolV68NQ{zwUeo}qy>XSddYbcf^r5(9iFZuU>
z&=hxyPA30A*~KlZ;g%|~j-qnRJ7HSaVIw8vF6G3?K>uFe`z~3hQzf$0=_!N33O_a6
zPfzzPS{m~0>T;?U5tjet*w(-oE8Z?wcK?3zyfj6&%ywOP>%qEJC0$*<SKQ+!J1gX1
zx&;fl%y<Ku^#?RzWxPvP<4k3ycdeX4G+ReO)A`Yk{yCXW@w|AZChJ!ro1&8sd})2z
zG@UJ*9FUsXKjg>E5!f6Q$rNe4F!gh3T|AS5*=Jj0{zV!J*BWz+<2f(WGNt&`Eqa>t
zH^?8~Q0`V?5d31Yq?I=GV6Kh2TV<wI+KBD7)2vwu-MXs$y&17ZMsrLQ#cbqs_sl<q
z-x;t?4Xvu4naVwH_@b=X+&{+L??#=-yoQ;E>C5t(0nf#$)cvufCqVsnd0M9V?Q7<N
z%Pe-vA3pfKe;79uaZ=vwFo#DSV`&{T{x?#_D@h{p$VuVEyr(r0)|JHFSVh=6t!#Ke
ziECMDS$Z~}DaYWl*fPzYnmFku&B*%bJ>m5p-9L%9hoCLzml%ev>Kk(mUdJ50dDeL#
z{^$9sWf52Be)YM7{-sZ&M8{_>BB-s@L|HaR@3^O}{z|WM@XA=Y#W`aQnGKDK-aeZ8
zN2ttpR23ClrrcbVmr*q_xwj;sxs8hTw9Cg$Vl;Y7v_={ShWE^!W8_Wx!0BVbC3}>I
zf7>dvE=@1X?~$DhR&-ekJ!Y(NBV8~1ThD}_`7ruO+atxL!O*Qrqe^eva_PHAkq!zC
zKj0TOHdsc>=A4+xnvOTD8!M8G8x1IA4(#f8i{qDw`udJL+i_&%PIcn`IL0K+Rl-{P
z>UT4XOtJ1&x8AZy+cEZV+xhHE!})>5JvW}6)0u77zTA?zgE~k@@ql{r<u4~1S@gVx
z4&Ji4K&2+nB9v;Ce@&T{MdHxqy{diYRssDF#mDmlORt1@S(POhnlWa|#$R0*o)Y&1
zwfac;$#z2lyKa%KX^CRSDv5KC^cq&(Q}Iq1G`~*O5dMC@sM3VENuPsoR6W~6o%>!^
z3cDKr^!-oqId3}e2CWrhl0Pc^`g$16POqcVS0CK>O3@{st7J1zsqTGm(kH8Qp?*!=
z6U$+?^7ju<_)+iXce<2*Ue1Di2pecn^#vvIQ`eL|4a*t*N}Y8Sb<YGF^?P?du8$7k
zy|Bi6ZkKE+x9%I*%^mmlQU3KVw$=l@{IjB0C`ra5+oCa1Z2+jHe3PmAkEX2^^Y)#+
zA$cxJI<GqlSVjcZ3J!Kqh3LGe?gz7Q&~r-1O9b<R9<>yCc{P0z&{AwQo}5zHKvz`1
zbpFl_3&GCb$2a}Qt4>NdrasOrb}pCn{+9BQvMA~`Av!*~yfiu<L+7zqdi?gu*IGJ;
z472Xip2$MY3|Y<MR7d!CPg%X_-Og1m$)gu$B>Ynbev+Nuc4@Y+0$9d*!1m+NCRNBR
zl*+t@f{WgGri3YjFW@bY#=M`h$F?-_H+RPaRWuWrXk&&g4#|$TJmd2G?&KwkF*B`V
zU|@zj3M7GXhs~gL;sjTJw&{zYo~ebGVDgnXk##JB9jB7>1(edr^#Z{uBRK;Qoq&^G
zbM=BwUqAW!np;~37R@~=s?3(yokp&A?ogCi;{K)1x|bS78^!DPD9E!(4LZ{7GR==W
z?Q$+qU+ei9mrhigwsi_j0UTr?-FW5xr}AgcoJlpvIapdLwjU4NZ6Ry6VL08nzb~?G
zvUAIDyVTY3H*L*E+vdxb=8H}iGA5AQYHZbK=G!Ad*O+>7RJPN;{zP5NOm&bbVth8r
z+JmXSFuV51?_9K!_UiC4m(YGabT`MKYvYu&(euYY`8#RyNKGVb;cY`NKN1SnMgKTt
zXC&gXXv=FIcVYjgxg6#4K6o>E3ThvPI{j=lbHZtB0SptRoXMm!aLaK#IGNHrNf!HP
z)^svlxHHFUb05Dt<6wdLdQ+|ERC9GooB>5Menq~>k3(g=Jjt+~aZLO)HDqk$`Qq^A
zKI2op)7s^d<@IhXDo-vl(R=nE<-7w1GS3zut*@`xX#T{(bl;Tw8P2xT!zGp>yNvqR
zbNM^vw;db(Qe2<vLS3zfmFVjWf@WwO>Yj28bP#d1#z0VV8&zNKEI04@q93}dTv9P2
z{d>l_9}*Pbt950Ao$7%E!Jpr~i$?J9VKy%;3nq^(Qy-|3u9i!7oESCTTY#Hg{>Hz&
z*?6{6=RByz@jkDmzHy+v6~L-9_-M{nvWz;5K8%$cXuRz0S>x8Rh5lw_Hj*38a`C>1
zEs&mGE!Uz_mFi(1P56E=#OQtXwZbl+ZQU4BddYpRJ}MAX4R^A#vUdLxwQ!-yluA_`
zpVUsEJrk4O_#1mac$(Kx4fI^6;q&dBFX;3uEG#6O4E;07Nj04?{o&PrvA4CYtp}W<
z<UkJqI49mSFL}YzF=CRWt$|JPT8%|6^(&JR=Yy*q^WzQ;6l{ps`h_a@*)u_?L}XkK
zV_;>dxZTB5QIh|qEH~=IT(S_CRN~y+PM=$y1LC{nHGX=|GFol_eEZIBIQLla>Sb_~
z@&*XwP)NfD^%0AMvP$w~g+r&C%*?@|HkbeJ^5E*127@05T-A=y?-uG{uzMbzl_iR)
zn44khW{?lD*Eqn|lMVwEt>lZVCo$cF5wM>I#93U`aQ$@dobO<Q>(dg8%$2iex5MWG
zcYCEHpP89CbgUH*QVfKG`(zwvWJYfCQCBtF`@jE@&(IkK!I8ZJ!|8VJJddfCyXILa
zDa!Bpb}nEVwb8dxl8!<wd>i@TiJ5VCX-=G2J9o~_&&MSunoXL3IAga{9XYnt?)lY@
zz9KNv-E(eo{c@1{PjCI5wzTukG<z*J3pV?|D9Hl~WMXc<7aT=LHYP6Ar0Xg=AZ8(T
zeBC9wQEBL1Q%n9;yWjXk^jK1kV2xw(c|96i0C8p4T4Q<X(rplg{v6-Ge^)wwyy~ZB
zM<=HWun1QxpEiE{$P!ufNyg!Cxp$yvDz5FFU?$hBlca!vV`sasTurjObjdQ;)^iy>
z?O{ZbTi4dMO-f3tuCtR9f8fM$d*1iQH|rDg^wj-b^JB*59zLyjawVVD$>VdnFWuJa
zg1PP|<C#lO%ZBaDPR@u=&4|T+V?M=)<Al@s^R>F~ouCq1FuBMbyCKjFT9Fq~S${8`
z6`?k@$MtqDjdVV;ZwH^yvqxrD*#O*)hLNO$V$Htl4;RYw&Ra9SLwy5<bEQ~cgkARy
z&V{oE?&D?TTK8+GGTXHeSTk;o<G8Z-h14fTk1G-RXMRJ_?5>Ss3m>x^9Jf{ZA{c9c
zgC7H3>rDcK*u*z`ecR)8wY<jC!r~6tl5!yb=kX38Aeh$8wZC-vQ??+4Hyo%p0)0!`
z`_PW1h!LsEgNfC@U0o&g%b~FG^w5WEMLV2LF1Nirurx-uEMVj|k|_rE0|lK;<<L_l
zB>Bz(r6kFZ$HH`m6x$DD=8}^vB_$-F7cqI4GcKhC0>2}fhxV)r!^+w|3o*<1`zxM8
z_j&a;V`2FLYS|9|9F2si*q52rmm|kE+7G>(Tzr|AoA15$i#85cu%4_1lJ4LITF}63
zU|Yt(qO(?DYPV!Z<iN!vxah+WFN}j=I$>4(y7%70xRrWksk-%6uU)vwMCFUcmrAW~
zf_P(?FL811kUzQ)XP8l;`)vRGi4!LX0FjOLDU`ZCTG-#I(d=<s`A5etF~e;xRIDnA
zxG#j?ZmCh18gsuKjOk!zWApFr?e*Uu$DQw^%r4ZSm55a4x%HXEfWsH!E{dMk%eiWr
zk6;&90aE(nLhqHOi#`me4!$f&sAd@*)EU~Bpc*xJSU?2tstCNeEqEmuxr1I36kBrO
zv$&D=YqDK=-dv`OU5wR2o`@(@_3C&HJcWW$k1Ot@Zk^ktJiBf(bwlv{{@0x<<(8oh
z;ys3om0x~z<2#_csjgPt`C-k&V(NJ9=XTN1XLgYY*{$RGMd!S;vzRB&fihsiVGepB
za+a3oTFd48Vm3054n92eP2E$(ZJjFwGtXeNG1JXd#0;&7`3lAZ?WWjJ29FUykuzuH
zO-#D1J|@ziOz=Kf<P$b)vZjZ5lv4Qwj|!V|^YVnu@)kJ4DjQ&WGKp+6RYeO`sZ}ZF
zVNc5*UZyg2rxlv0Sik#h{_4F?5qDOQAe`TE70_1Sf!_RZ=^Y;9Ud!?71aAYr?%EjN
zEqi{va(vSDVs8+=>MQ4U>PFalw|snXKhE#QflM7tP5Ne9$jCpjsyg*YmB$-dMpL4a
z&#4>Y2@}OL5q!SJG}iMVc*xtHX{nID@1VzUV8SytjLGJrmikVIH_kfks<bx-$%BSQ
zCnyLHY8*T%k*2d^O%H?yP>bpHF3Jn>e}0z|EkAg7JUuszmAVlN#NNcj!~m1(>YADu
zfm)Lz;1Fu^h=27b>^!3B*u*vwbcNs`prn?vF<EQp&hEC(CCBo6TDRWK-@a({DopFh
z?S>zE46SM_IZ{r(Z@;#bX*{3ljAmUh3*($UlAW_^gjXg|M{3oo(5vNnt!FKmHKW35
zYK`Yxo4aHE`0hZ=#3w!b_GCF8KzXX#x+I1_eTr;uayF4e39=_Yv<8czho^BSPG&A{
zrdPBm%)y1CQSPiGS=mro_|=^5SaO4bH^+F#_d-!?h&r6-1`^|niex+&o4usaUb~c9
z#Tv)>sKs=d#iHxTjeXj*J?tS5r@cc?<4kvY2ooB6M@Moc8GNP<kijf($s4}$OWmgG
zq|FtQZ?2OM8m{SA5f)HiJUN^}r>1z6a_6zs_IPetv)?rHhPCVe_W5Y@De0c~)z3nL
zDUm-9)bEitKJmBDd&h>3C*UV`$(q3eaW1OGlrzHLhoT_ZbHYtl<+ykCo8ga*%dh$7
zKVJ}Dv-b|mm9?#_PxFXm-E$I2zIh;~YDuV}<Oa3+`IqZ%eYlFDAr}5DSJd^<aoNBu
z)Disx4hY_?!kh1lUf%o8gF{btuYBppEzwlb)B!ItwhfdEeyHS=^YwXdiq6S~?P<f&
z_J9wPj_7|pGC+#j8)}Ui*3-jt9o^d!uEQ8Y#5`8p5qQ(Wu-Q~WnW*YoAPN?>@@d0(
zzP>#b6-H)8!jHE^pUcS2&ENQ==8N$&dPPffubab|XfFZn^Z$8C-Dus~lw$tn&t7iX
zf*Tdjr20KGX$+nHV7>7b%MY~zEOSSx8oLd>#zf0)UMLtJ=7@~j;rfoT(|V)qDFPQR
z#&}x@7IQ9)1{BZuwb*?7{X4_*>wl?h)z20Zs&nyHxq3%txQ%sVMa<DM!=YY^|Cf!z
zs^#*FG3B@<^&M(cw-)95G}3QM?3O+Hb9w5Mk9Yk>KOIV!X$WmTLd7Fr$+__RAKP0Z
zKV+f3=)jkIh|@JuCS+`bX6$WNxu`+9r4Y-Qz_8V{3S#ZKtjp;X2L1>8J%3fnVH^yR
ztW{q#WI^$!UnRV?IG2@j77h*yzot`Dn0Lq?rB8iAMN!n1U$2mXfAO^;BVL040})oS
zc&C<bb<-s2BIJIQT|ybZ486Ii>EUs+XZ?^$k`V!sciH7e8pP=TP)m`0+w%6FJ~#O{
zlRY{XKjYnm_(vKfE36U@doHmp+VLa4{g`H5SO&hBCSUYy6eeF#k`Cb`|MWkiRdmN`
z<JOo5GYI)GN><-L@9<KP<LsS))S0gYX7MyD%c8n35l^#D`OZN2&#ihUJUV7WgH-CP
z63>iV*~D+mS$!h-eD1{yAN7vv(0OnEwR>CnrfB8U@0i2Ch*|pI>V|ku2=w*Vk&zKw
zB43{zhgxnYA7<|AzPzkQlH<b3cz#p$T?{+Um1SMBZ&*zg&RymoqvzOE+qn9U&_B<D
zBkkMKcd;41Mz_&l^BR_TKOc$L{As6-`$L`+1%JB$Q!`pasc12B)&XSJ&9*tG8YL}7
z(8ji`D>b>iJX$~RFFX+!XSBYSd*8(k2NcW#l2hq?k&ylc$W_bz3Qy2G5P9>W;@ARo
zb)VU7&G5gU`Fogk$#3Z|Ig)`Vp=xRtpsDA1C21EXyd6zFlf`%OT_}B!)U+;uF{|<6
z=QWS)4W*-kkf+kqFD@?D0fe3O2;`6oUl5;4V&zDz@!;!jdG<>Ukmyr^AD(mx#va0Q
zaY%}fH+{$c1vBf+FI{4>gWG_eVrLNsVs}FsA2w#`;OMB>$3L&YXIyxg6w{COU0q#+
zWfD2GN_&I_1YX`Vmp{j9D9!oD?}Rgs`Yr8LUHXE80$X&&h{1)I9I_Z$Sz$1N8a}A7
z2-*PMd}rI7s>hX~ib@4oyU`m(L`2G<3o}7}B#4)T{v@|Q5M3x$gP`rM0DU`)a#Tqe
z9|asw8pht_UQY2|dAqF9Y7X}H@=8h-p?09ISUtRe`9A~ZfAzsy^>O0e@-JBu4F-tm
z_doI~JmBie+B4hN)_qSWH-qm7N|Huv%c*hl&7}z$$)4A@OPaKcW)zGU`n(t&rRHC)
z6W+F#Qdz}r&eK8X`Rh61!$BB)EZ{kIgO){W4W&VdD*VoOKVvpE2|9|p$|_-oE9L%8
zk0)er4h`NhxAZf=%$x5)FuzsDcFJmuGxB|63kY~`QY~`mP({hjhe9dqLtg*_t*-Vh
zECdKz(pFel;^NJ-1D};1yJo<v9}>hO#3(O+^j+w)?TiO1&&aNCgnq@h_?aaul11{_
z7@BojU8_0Y{P0elW@FU(krE`Un%kYCsrE)SX`~v8HM)5Jh<&dv{ZLd%ILNSmQ$WDY
z35pi5#Ax@cIlk`F=ZmlV`uLQzt)=;~k17S{kF3j=tMs=vGBDC{NNg2!b8KRu;)Z(C
z`$A8f%aHsrsX9=}#}0m|c6-cq&e*t1tmHD=;0;NfF~PT6UFbGi^qnB~Oy1nBE*hhV
z4%=@#b~H%h;6HZol;mjd#gY7ED6T1YFT^@_%;vBCsez7s?@(c6)|f%@BwNJQ))p#C
z1CP|Y<0U}fz4hb!&MK2bFH@zaICxX^o?qcAtNX(I)QhaF3Md3cyF`o^2X#y!>3@bc
zl58ZLRR%jKvh<T|yg%4QSFo_M-XS$$R^e1ysOS}=RNZGgLGrPK2W2Tp_){7BE?i_k
zNewpar>2PAeE+bb`dVPEoz0KgtEH+TbB`-1kir}}W+*~~P)U@7KEh&$(?bPBF(%mF
z^z;Np;#FBo5tG8j%XxE{{X?<JZ1)~mLpOR^ZJ~#(P?G-ka|;UpoT@dHy@8KzXjE?v
zRCb2KDh^kL9T<zegLE{sKnqFAf0Oj8zKNe}`nYzI@8AQ9GLA?-1uGgh7M;NewpEj2
z2M!#th0j`D&wuQ7w5oLjlk&H#7T+U1Q%qM*=rQcuXQ<x<n}6T=ZU_F@`E@~4lav19
z363-Et7NT&Zkj|e?AMmZv}v}(&ZK*ivAw<h8RmHlHh+RDD$a@7`0w?SzzqwKo1#>%
z5&v`M#@E*u9c5)WgwErg*+Q04hNk09P(HuK{5wiz@z%f}z|#C_n-_sO<#6qfA>EOc
zh6sWzb|e)3caWnL%D>`K)42={>8-5jv0btZ4fS-^P}DuZn|goiAH35)><?~~?fVo%
zAlJ^TIm3gi5GKFMHs7Y**6~9P7xg-tK0Jy?6vI6mIn^ZTD6d|x>6fT$4R_?Ac;Eh^
z*5Uj_uWxVfv%c_$|M9`UY(@8uRdN(>&0L#?pAFQ`VmY4qZ=-~h#Qx#7#sU*kX(--K
zII#ZnZ83Lk6s9nkmijoAO!yWT<GfJ;L!^Fc<+jwVv5gtxGet%Y%jO`*&<uJ1kSqLQ
z#Uo2=zKC^6gs^7V^}z2<9~UF9FY5CA_k9Ppui?4H$41eNAos=d0H?<PF1`J)-{3zd
z%|mzo{s8#%)&Jnb_X(RCQW~!Reh}j>ZON&&bLWF95YcA`b5_%8oc;R~phUl1Un{Wo
z_ZO?lPsN%bYiUWiaaW|HYY$xsG`Hq(OV7!%;h-cPSS74_?Qev5v`zlmBV8?cu}M<7
zHm|n6{qp1T)TBI{M2Le3MK+l*NdIyz{_{KeGm%~${bYY^+JV?-&&FOp**|p0n^y$)
zMA6LnXE(K7l2Cw`F1L)SIdce#>59=c?7<nVhNkit6S5;n_dg5kpQ*22op}<BQ+-Rm
zcx;Z#a87S;iA%IJUbGEsj)A;1=d-g%<c}WhvXZ_+wWC4EP>k|<!&UKXEY7$7OK)`x
zIRgKReV#l#E^px6b@Lf4he)dDK5*Q)GFpW~$vC^G#MTFOF_gLEX;(b<Lj}wP?8uj-
z4#B&Qmm4^dN*FnYP|9YCcj~`Bx}Rex3lRnc2?+>nf`FG(>JehYh7B7c?AtX?N->am
zwm=PcLm+#)dNUQ}&E0)$zo7F_^fgHR^X2|51okeo%4%n<!IWd=<}VYoZ~7+a=?So)
zOMsUOPY4Jqpc>JA8NQFRsIB!SzLR=9$sDE)tU^r;rFIEo9Em~`%qJqz8DnZFr?4fu
z0xrQUc1k%o-SzeL%MH+?8BzCrG`lA-8<M%*3;L0fk#!$Gig#5ISZjIu%JQML3>udf
ztS!%2h^306FJ1pzpL6xriXQumUyaI|D3o<_%JN4vIbf2tHPH5dHDg;t+0OM<HmJU#
zVJpP^0pzg8D4S0p7D#|?feMlri6V)iXaVd)qQ_o`kD|3QkmTn>894g3P`R#xyal3I
z!QM|0Bq?@w!YCE5^Gj-BMRJ0p`XVm+VdrSWL?|P<T3liLIe;BvLpv#$8AK#<*bM!f
zpC53xB5kxLz;12d?s}j)GY$#dOpnY`kNK4=K_sy*{9Z76FGsGrJj}BQkf$K&py?aQ
zOLVriC@&QfdKnScw};_eXs`<c>n)6D7h5o1o-!^inknP=;m;e+=pyL)I}zihZx3!Y
zKQ)dzbyKxTX0cv|wg~;CrEu1Y9P2^zX!9*}zt436#@_%$dZM&wd|T0}-XF!cnawSF
zqF<c@;OElFK^?7j|665(D}VLgEB5})83&w4a(lxq&qG*}xG;mPp@(kiGaV5FA^Zou
zUSXd`=#nqv_XJlv>#~_tg}Hd1CO3~CUF4Z|IyvPEij1`F2L*Wyn-E&$qM=%u5QiGA
z$#L7mGW$fRBuq%bz$Rh8cL|S%%X28pbH8=o$jj8@tExYvA063c;R-88lr1E3>pjri
z;(tqZZOzHbBi?*5=d;#E37RO%Z+*Ge3&>lgK3DvyccYI69SAq)(ih1@f3t8LEpbr|
zj7?LooyA1_5194NDvPhL-F4y2*c=*G2WKjE3OARQmX5>ASM*|a1Pha){i!BXL4_4O
zZjAp-Kl)`Q)gr{`R|0)X<;BAhtmXHGCKevE4a@7wy<Gd$5L!pSM~@ykLt%0r29D!l
zHSW8iuI8Yp2X<(NoZz)+;dlUf@qj01zC9?4uq}HmRzLckFpoWc#tY|3*#P!(8`6}z
z+S<FejrZRF{gI;T$Ey0DmI<ZYkZ>L#oC{uj0A^Sn=0IZ$VNd1dp{RL1x==lO!C=R$
zk3vnwZzr$$C`+Jq!364kdV1)fpO&M8nN?=~V@<bE)U0axJt|dp6ZVLA?B=1XkhBJ%
zZo`J6O>4ZSZ<=}o0<VNO(9aRtfaIO2$Eo(Rn6R(Z_wwF8`@R7n-;TY*h#!5QyVW~b
zunEF1jqVDWCEHJUDdWi}J$OJkW$>+w=Jt0`kEU6t`T<fj79xa#=w1mbwS}e8WzU6y
zsHIDUk=logNj=wcP#=>{!(lf5e>?R4Ibi)C&fflg#Qgsk@c*B{zkdQ_%TzaFRf68k
T{S*QLEk#*DO+H=j?DhWv+7o#k

literal 0
HcmV?d00001

diff --git a/examples_skyline/accuracy_vs_di-skyline_formula.png b/examples_skyline/accuracy_vs_di-skyline_formula.png
new file mode 100644
index 0000000000000000000000000000000000000000..271ba877f30856507b9b09d19f5444fe4898ddd7
GIT binary patch
literal 42840
zcmeFZ2UJyCmNk4xf+C0kRH6YzK)?VZNK^p<B`1k0q9n<e94;!NBA_Us1OdrelH?#L
z2#92^WRT1y=ghyh^<H&%Ro7S5<NwAty2t1@-g{SX+-IM)SD16I^-M`YhGrMzE)2tH
zWUpOS!7$PW45Nso+5yivG}$%71G(*GSv4y7$Cc{tQ~3K%t7|uHF^qa2`j4ck+kFF`
zyl5w-Wv6O+&(7htjS=?Xww;x^rJcFSog?-}Hnt{~7Uy}-@t!?>#MsWxN|=xD-(Kgn
zw7Jh0Jty3OVMj38tCDJtVH4d>T7kH&?OB54fva>a-@PVdPI2vwxsw>EGyEkXcEH^x
zQMU43wo!*6Z?sotb@tb^vy3t_mmZ4OcFDzOpW<mgEF*t{M^&ORBl0f2WS)n|%wmhO
zgatWm?Yy;~zLm$1k**Op+O?&x#oD_o&XB|6Bjvhu_23a}cu5yY0+0KtU>7j_PI6;-
zC?hB3hlg9FB#iKI#uKA~hqn@Vv_;pCNRYw9$^RZ7YdM?Tnv|-OD;MzIW6P=Awq7)j
zkm0&&)9pCm8y6vHJA8d}h|l8{`-9#Ry{M?DIKu79{t9pU!-qN9*w~7fI?Qs`w1Z!V
zg{9!}Sa2H_X>H|zBf%Qph^`L4=5?72wJ4DF-ouvGN?0KpY$rus@iy%(am!f~#*~78
zlF~%#pWl5YHa0emm#p<oP*B{*j~`7%mp4}D1$1e!mffWMc&>Ni9wi0#lU(p~i{q{G
z)5Xj02QqSdoaIGbmh429dbhVU1MD_dtbLgT6(&1#H4E)0`S1T=mQ~T!4Vx;OSE!m-
zRaG_0&|>y9`S$YYb6?-qNXbWoLqoCAGXAFRnMQtu(WY3R9@qIB@oR1A2I=3vUCXiP
z*48VuZw=zn$-Y;|xH#EKFzG4xqQ<RH=4!5pIZSrM`mtO}ii!DhcB#i@no~qXHz3<(
z*<z7MP=u}2@TO&17!luI&5Y=-3FM|{W=^r6?BEm=>lw@N5%=-oF(}d1E3kXz@9&?M
zKg`%ZxnwgUI`{2(!E7bdV!g*!R?YP4uPclx8p#Z&wIXL`tX9|7Oq*h4xlW(96Dcui
zJ}D(FT~lALNJdH9LNM^q8(U|+VE$HYb#Q;(yYn(VJw0nnotA;&;i+TI@yTxMOS@=k
z+u**AtWcP&&J8j08^2_AosrBjBPdu~TZahQ#%(Rci;sQkw0NDHo4bRCF=l&n+QYfe
zld>i52Ginb;}w_XDO^$9*E@}`j>|CDmWtgSx)Zn=z5`zaCafk)NYl%4Temm5f1j8b
z|Gp{Qp(jx_Ucsa;MBurPPxEtn?l-SrN7FD}NPsQn;^b^hP>wzTThh{^W_0(i@!Vjw
zx_faTw<g!oqmudg`Dn4Nng*#iLN9P~aLCjK@iaf$BQ{DkUg$90wlLPb*d)(4_Wj-Y
zwTZ0yyE)nd_r4!uxnzI6*LB{v**RV@LcgD=@3FaP+L{z7wzXnd^)pF5nIAW<Cfl^R
z+?&2WU1}gCB(%1*J}qGV>x*2kjGWx_t(7Wq)1jI`VW+v6%*;%?UtgYax$Rh8T`ia{
zS<iNwzc(~IJnq9+qICCeOmkRMlL}m#yn)ANj@ZV`l^gFx+8125a<-SZw<127w|_9K
z@*5+jm$uCf)t-=roonrO=uH*ubydB2^X6Iw+g9XR!^+XFd|Uc7;()(++YQ0)n6Ewu
zx%v1My9*r%@B+GXV4LI3isG%Xd)emg=c;VMfMR*|3PxcEz=Uxv3thJLoafK0va+%=
zp1FlzTNsP}{{6Zyn|O&_>6bV0A4;1sD<N=2>gR5kljEkkZdzDagz%dL6yS()@H@n5
zkL^NMw*_S>DXHo^8K>CUuYUggxue)6SK+-#Vz9omY^runKzw|B;_IG!Kf)3}ems+3
zSeWcIKfE?+>5-hKS6Fi=!+56ebHQSp4!ZlYO--@#xaDpKvoBBg+SuB@E-fu}UKl;P
zXV0D+A^ftIrJJAF$ND}$lDl!E6%2sOZIV+p?$p_{$^maeLzDN2tw^ZGU4O+cVA-Q9
z<HxLg?_OL{aWSrfMo_IF&7sFhCR*CJsZOx>G<=`v`b3tEZL%{FXKUN+Nk(b6G;tF*
z-cmhg_4L=TJ2&2)lM>%tczwa5>l7bfx@qYM^;F@s7}|mdou67q8>3mOcsbbFNB4NF
zJW4CwauzUayMF%O_m|$@-X$AzwOpma0Rj6?oH#M&bx{8$98oxitj?oHQ+4y3&c@r0
z{0O<^kk-}R-I{46FWdAp!>~2i(!j~t*>vJ(#@xa}3*4aAl$+<_qN`1)vk%jwb5eU$
zf-E6`o?GLheee3}>R6d4S<UzFEwckvt#AtEhuOs4bz%E$9MXI8Ydk$Y6>38TE!$`W
zyCOe5=tL)-&2`4}-1+k?>FWA40sMxQyW#H0J-KZRMjLX``h$pw2y7^vI@8IH9ASs4
zu(mY4WHGn3Qx`94sod1s26xaJEwcyiwMlc_jleYGVD)Q-=~(q-4O2M0`L_4#-}6{n
zS~|J75LZVcG?O(_vz-@Bd2VK24Hw2WS$eGNun60IJG=eu)p6}o_u}`?<D5ChjlZ7v
z%u~SGvvZvWZ(39&B5PE2t`xS=FQPOeBEpYNJnc1yqJ~Lvz?*mP65-ZV-RG2r%dGkK
z&FZwlhHY>REsOBQ63h0?p%b+SP5G1mYP7+D5DIQ)-7}f}^|hv^rs>`J`*xE#gqkzE
zDR%4_hcjF{C<NOir>MxceD$M7s&-;zWLd^WDfq1bAqJnXvzz(0BMXJSn+kixH$Eyv
zh^3yt|KpUfu=bPN7xj~i7}>|3Fj*-1Fz{r9<)W>GT`8PcpKMJIe)#B74xIS-3PJWn
zom>mK>(^W0J0h!CU9JyQ`D?nnyL%DGTM}Qx)lF&fo;szZp`np)8+`ia2iSn2p&_BP
zA7^a_uemKW$_V4e-@p-wf(sEEUY=-IMZWAoZuhXAfbV$GNbSzuyVcCiQ+|XA4Xmgt
zDfRt4XF#v1hYJi5v`ov+=C9Inb}rc1++?XO-OAVW4h|+5ZCl6&ZvSCVc_|wkt~XCJ
z<XHCV2OuB8a?wg6z^-0&+4%EAijJ)N{<|w0z+a=Ny0y8U3@+JuW40<VA|k!K{Ic8D
z>PSEgggUpS_WQ7pn)dde^d0-X<P1s^60VK>2v=QSnjD&%N~;SMBv$MZw-eb~@7>l%
zR{$odrmday;>GTGo}A`*MM9X6ZA+2UJQthvIq<W^YvaktH}iwvsaol@EKOze*iZ)V
zGdDl~I!10!^&xThq6-f?V!(0|;J4<7>yqcxrO-`+@X0MMZU8QAZfU7xFemQ%yALA~
z9{RamHa^XKdt06A1K$al1qH!oLnx^}vwLV{ga}?;6TUr=(1=c5*hS@xjEtJPI(2RB
z5C&ddtufg*_aIzCI8aBi2O|2ySbRj#rS!}{J!EievL>JCLc<m0M}_T1BRAKV%^D&l
z18V$^`$D$RlBhZ~Fc5udD(?#ND<J}Aagb_^!&%t1YnO_I1YRRupGTJQ?48FaZ-m4^
zw1#wHIkRd+KW}H%muLH1A%-FoGpu}JEz(o!;ZgO>tE8laQN;0Mv&IiNkH{OuJ?wbH
zsft!+`t#%Mg7Kte<^i2g=E_=HZ;>0^-dyJ9<gB(%v2ZFcFJGu+vNQ%8L}u8LZK669
z>I660w7=ZT&oBU%sJaO*y6|q#i^J#nEqg>D@?K9&Ol)<p4;&+;NAx%oc~F}0pd(xP
z+O?+x<X6r^yu{_0w#q?H(&#JLlK_rn9DHWlE5;LJ5Q<wXeVJgVISRYeU=yOdoPNjt
z+KfXJ>7)(c!bM#)G!5KxkbNY<iAm#)SK$A;yc@%L@ibtJ#ig*C=g*&;sz_#XmfXh@
zxVvhYB%<%-Xd4}R(w3?-!q3XUFl_g>io&vFHO~A(Xd>Zb=Ol`gR@IJHRv&3uMU2z>
z7?m1b(rCuU#>U}h<@?9R#GE>HDxu8`!xvN@Wrk!LzmvHfCE@=qQ|$i{b-@2F@_%Oi
z_K)vd^ylBX{akRVF*x&GG&DGz5Z0%X+if4ZwOa2{Q1HT3R#7n+a(+2EIbFS7N+-yT
z`5gtj?OxNeiA{{3GwtSxaQy%kTMdDrsgb;xIiLHWGmcf%`CDzR{1tCnaIxBx2X~ek
zU9jv)Ieh+}Zlmz8U%%`|!X4U_r0BD>Q#%T9$%aCwPF-_yax!g8xtUseK?$m4<CY&0
z;?v{%l46SrbaG5%3#UtT_=*>rj3Oc+Pp}@S9BaQH(W)rE8RNdWW&(9jjxrZsxJhaK
zaKoX!k&U+|x(hYk))r9tggh=pGg%cIn@mztQs#L`*7RLxrM$eolOeki_Sh<doKBvG
z`Jxg;?&nXRD#05Kw_YB-vJhOlAq5508@POFf7aBLl+*3z7f|7+tgH-4D(?ICFCV@G
z^z<hT3=AM~8iV}*C3~uVvG(TLqHU5>;|D3m^Y>n`k3qujywdMYf8YR0XN;F7+J#+~
zQkq>HXTKh7%`lYd#u4;_i?$&In}M?)Z;Guo5>JM#4~n_#`%hkf2{|_G+5-65v4*HC
z>gyj#TBr{$XF+sS>!{t@S}wg5v_0$ZkpMZGX>**fv>)?W_Q7$;q*27_`1BwdesvVG
zHUbK58}s#MP$Wb6)VO-}YHUuph?5Eg9_7wY59s&p`v6sQ?W`8m#g+bS>9Vg+R##QY
zz)m)T|2`oba^XQpQ**O`^MWx-1k*h>lU4pwIG3{PT6FK{P(ZmY3WZP%Y(jaP|Dpcy
znL3OS4@vQ1NlD2{sE~a>d^lzZ<<XPE!os~wOlGcg)f`Wt6fm>^U%q$m-rF(p@k|sH
z6qV2RbG(2|f^oWZyF^(<<&G=tp)@Shm(6|U^O@Zhhaf&S=9p;?K%A|F{O`qkQ5VMU
z0^E7AwbA3zd3itI!g9|7I#F?E_i=9S+L`(J^BO5yw+|dV$OQT8xs8pD7f=u#7Z>lH
zSz5YKvbm%q!vnYO{J#DBX)4aVpk#F{J0$9y{VFifXmD`w5Y!jvSUonZ;p+xoD~g?k
zFn>>EZ8Tc?_U$NTB_*Z)`?HNQY=<Fz4RqIsGAH7!%aVC~25iLuDtOntY0HM(yX%vh
zYEgiNZ!v_$N<((-`{vC7v|dL@w0ud(b{u+g`w&#1(p7#ez7*69hv2>n*x1;-;49r|
zg6no1PmJFkTBi})VNo!4a=_nXBib738^h_I;!1T*JtOnZPjwlr4h{}zm!*}JU(G^2
zd{`s7^%NJEO;NhP*xI*@G$=t22fP5Y;!5qTsHkv+{K0pAq@KxNd@Ij-bCF;$CPVYK
zoWYo%%D{a!7OM2cGO|4hP~0&>Zua5B2d*%WUbkh=GiQ`|bUwA99By>fSV~5wwyy5R
zap@OIrS6*xpDEcc4)=m_k9XuGySlmIIv?aFni0|phJthkA#vQf|D-W2Iv{4b*TVqz
zqB+~789fhWUxJK3Ye$}SbWu@}9Uv;)s0v(wDj6LBbf|V$C|$n%kby@#29-Zxql#jy
zF#!Pq%mR7&`C}-3gIZ1dlev>rgr>g!d-(D2O88%3XsC-6kwR&icOffjA(I;Ai5%}8
zF9S~W2F&rj>zpj$Ly9IQCU6j63)@d5c(XcRd;k9Z-0W;4Y>H)49HjWf`Fe2z9QIR+
zii(gpj=y)AYJ@7B*>-JdG8rni-!<@j#C8hGqF0B41v00YEQ^#I8ymsJy^FgMI{dSM
zriGrApYbtNKQ+C*22hZkWngC3?cXRXlc-vT-`CE%=L-hP>|sCgGYS%Op<%nxhR?HZ
zkY4lM{83wL+hRRS+(Ajnef;>900dsh;PrD%1E=x^Xu*|^Rk3-fNlJRcf`|$If~Bjz
z?i;f*Rn8r`mVB^i*(SIGzQ2@nX5fz0w<EXw!OMY>g;m1;1_uTVo%JwX;$`t<IljUp
zJUlnx7L^+6!|8*Z?A*C?2u;IHb}T+OMTm%@VHri~lvin*4p*lH9UWaHTvpq+f=l)j
zqbofS3B0_#n2-AjOW~y-H6*HX0H*Wp%x+Y?ID5@|uw;{wkqJEL;13tFt(@h6>>k<R
zm@pKnuGOul!QTu`PxJlGp3z01`}6GCvk+li68l(Ea+(V42yUiNQii6zz1{DYyLBBh
zg}J%8H^IU2C**_Prky{1`n!qGb$9oYlI<-QSj#*8gBYQpf$94~CTw@{>TsxnZT)-E
zWB>}dz@WkOQWtV~Q6awumK3&Rx>;~g&PXSK?-UPDD|n?AC>?7{6x7>j@)$`NY0B{=
zWXNm@`Yt!Ve*LOlXrF37-J=&f(&ID|j*NI~G2J6ECgv*qw48!M8+^lc!d@1ZR6sm5
zz!5dA&X43kF%2oJaZ7@7K+JGms9E{roz8==WEKG7F#)(FpypG_cYu@t1&MTD>8F%c
zk(PdfzE^!UM_yiDT}vwgs^`N`qQK<n>FEP6IRMrmY}JSFaa-1-XJ8nAcmKVzqhlTv
zo440Q$r9M$+S{}469K?z`22|Sge+_xGC6h=SOKa^^__<3!!J1~0NmD*V-_5fS!@jz
z0v8WYYIYE88VZ++BD0ToQByC%xiA}Rid6_M>x7aL;R@hLh=H2=uiy+?<n`~UBB~wu
zm7Hl&)(VlGPCMH;3WA#{;Ebl7pHiVTN`-)~K9&#V`fum}FlJP-!e5=oVuN|t*{(Xl
z65;{v8ygBX&;amufl#`o9(v(H6S%vfx97q7A9NR}qfiNl*l^?PK?60YThLFVXpL&a
zM53k#cvk>(V?RVk2flxw2C;CwO~(=^0>#D2!fxKg39FY5y>6ON8=@czVA_yfkpTWl
z)pb6mOZF+OE)!<~$GZvGZCegOKN*T%`2bf}S197yOa>}_$A4x-Bj5~-?`i-lY-Np#
z(-b4b0^44{eVYJ4CPJ6M>9)cRP1Ui;|2n%encJHNwMsHvn0CHREZR7@^YQ`j9e?c<
zUGB1lYPk6!1@-lGCMyA-S)$u7U=$<^z*=#rf(JM+jaLfLSz(*Or#<fL4>W^GTXuZ{
z-k_ljGL2C{u?mJm%phrNV5f_$%`mEE@Otuu3s5|`JEeoO03Xn>iRt$D_cNZmdo9N=
zmIBj^CJ4}qZ_GSGjt{JY2vuZCBhDa?vEi%7C%7fi>16!;%;nav0RF;(iL<#ayn>tx
zQlQu`U#>z{QG5127%roL>6?4s-$ubX)<SRq0Ni?70@&Lt>16nEL+S-7I3B4lU((^m
znxuHNvl{?v%5mRxju+zO<YWYxn%!iyAu3pe7F%Q~rp1!{F}GI;O7JPABqzV9SELtk
z?AWnv`$=utCYR~r_^vKpBO{|$lW@%H#3&9^a+?Z{c<95U^-;oh_<JKjtho6)!CG*#
z5Dg3O>O<Jrw{IW2Eae_nH3-^z!Ow*OtKbq8bmw?50njc0QDfuh#>tD*rW(YPyz5h!
z_+#}oS;$9yR@T=qb&HNc!5-YTnE>9wFZCSwT4xCLQ2RPm7u>mX=fM8`$7Lap7QodR
z@QYMeSHtE$O|F&TAwpf9ET6D<p<vBxOIg&_A}2m$p`fEY{lUR!&!0De<;8<_Bm&wY
zD=Uj4Tyq%YCRJWU1IK=fZnx#`v{Uq0lCM4`)+&Q9c?x#Uz^9)A;7t-_2MFMUsvni|
z4}N}>RB5*?0v=|(r??PdCO>~_9n_@3h8u`j<XYYfJ(k<K_XeuczQjctl)A40nJ@s?
z%j2xK_{a4&5V@+(HsE)HFB7)@g^oC%(|5@45q?gcbos%|!h#xL8h}AfjjP~=HE*UT
zD~=Xw`sr(mk{g#jhAi+K1Z0h5b%<(X8E|`6S61>8&b6>toOu%%7z?jkncd}=4O}c(
zdtBmTG>U8e{d3?Q5!6_`wTD9U$DIsB1uVJ~nUr)!aW}|{)7GdO@Uk1;P5d*q(yCRZ
zo^hTLwu^L;o;xb<WQ<06v)Y{IgIf;gA&!vcvF#3Vnd{s+6@=#s5Okvf#_G1I<q^=D
z<O_KHTEMjBlx!22hQR$FVchy708M4%h;6&r400{PRa8`xMA#%YCDYd_iqp3G4;e>#
z&d$s<fDc;yvR`o&VouVMv-9#)D)1(<0SNk5j+QprSeX@2zZdQ}{O&%Nn=529=oC0F
zj;qJ6V&i+9yZJ_Oi+gk;J+<%V%4;MK4r?s6T9nS=3JMFkPo9(pps`>u1Kb3HV$his
zbIpUucOq~z-Wu{OlfE)fs0(~hMA=0Uzx$1@D%J<kUZWiC6M+w84+~m^xzC68CIGnN
z3?7N4es!i|kHD=mY;mkv2K=HD<Us5ua2CfKqrJg%4CF=Y83Fc{V3uC20dXr8x%;Y>
zta`EJh6W|X>xG;%dJfc5pQFLSo?0Lw()5d+5s8J+(vp>5)X2{KMdqF!JAQl-@U=^I
z+smcfsVIUTJ9Zi1F#z1AjEH1PVw#HukkU2~M{&HL8H^)?eom@fe*q`&Q&^D3r_-_*
zIs;SJfgn?d;G}9890Q@%wDQH_kh6wQwxFgNEm@yx2EWQK8?O-F1dJMFK}UTLpO4el
zxwSOe8DH$OTvJhT1*{!m?vU^$j=XnKfNg-%E_DIQGlY4eEddDsv>Sk%yLRqujFElK
zrMC>#0>VTA4wrv&dvuMMPJoIDx3%F^ygpS32xYpOu5Jn-<O!C&B?|R*fH5O)14S*6
zA)~-9A@}_DtYJ-ehnkuieC0_KaDV_~sVFG;!9q;d=CQsGeL;(^i>WuWp6JV@C)=8P
z?l6@#pWxe$(goLZ!97+x*8(wcM2IN1jJ^4`@o@^_T))dKNc;L7C;To=y$lTv1v82T
z7*&&q1D=!@s!bphk~Gux<>cj?gALphp?;uY<iGan)vGlqx)Q;)0~mjeplj<3TX)ju
z>xWCzMOjU_>7N-_z{#Mk8JwI%QP%{j{xzrw5o=l7+^i}lCdMf0yl|qvSg9H4AD2|U
zLJfU={lJnoAaq+Hb%QcjmTO(4ySx28rbb!pr%yPr7i@aY!{tD-h(#OO9+Ajc`Cu+_
z@lv^YQ4ta5UQK&46(~gjt(R*$5qx27Q5d145Q1vY?Djo$_A)G>xx{V#O+dg0_+W?=
zx90cmZaqZgt;`PO0Fk*lELfV@OrQ)o=d9NqMQCLs2O}b<c*i7|42K(eK2<Wdi&l}4
z{Ikp27wRE!Vr)@6Rb#R}OD2Fz)P3e5lSQ876us#AKeYf69$Wc-%tFUxfj1%K_PQqn
zU~x)7K)u&vyO^Ex?AeT_#V)(%8*p>N0l$3%aQJeMiz51B#C!s>oAmLc5_o>75t>ib
z*VWdl0yGY31rQ#`+!k9k1K8QvB5eGZscC3n&*B$gCv{ZX>ka{?;nBXlt50P?*jZ2Y
zDjtl%wa9L^M}RLqb2enI;mRJM>zXq*z!uGbb>6jS4*^2wBA8-`{s<uKDF`Snm?|KK
zwduVA>o<cG3Z2vPiVAer(LIM!Mef?Q2E@mpqmNM6=J_B#gEXMtt0I!qOSFOJPy?s|
z2Q?lM6)sSo#U~`3f-@xzey3paQ)f$|gC)?(!l~SXf|^kDs=(2-D4Lb#(alo|;CGt4
z3qXuKS_boWzN`wONFaHw3%{fZ#XBGl<bwOKr0T$8i!waPf^Z{Iw_GB86hcw5yaWvd
zx5Sx@u&^-s${zR5GY=opRNUcQ2QE5CQ^znUG}H`A-@%a)Ch)8RP?xB3r0ew57s3I=
z&2`Yw7pFNR23s?@FD6r8lNCJ6+?vG|>2$M;7Xz3Rq`;L#c+ql#H-$U`u#*-*w_D;B
z*))U6`!q*!0*~|wR=`HLxPblJ;C5n4;FD-5s~}_g$4ZL^{9Pj>P6#A$wor=qn{0tY
zAh4zl>2LvP8&1fgC=U^ZKHauD;DmO@9Vst=4h{|?6o?6Pu9)xsV%kL(&3=sWCFGPQ
zplk?;$;-Ql@RNRTHhXJzbwQV<2{|@J1%)6e!FJFxYeB?d<ZxygzXaJ_&bN|YULv<f
zzUbshcxYelG}{n>m@{U_qzwg}tPzbfNET30j;Peox|0=W*k$1h8^M_&+R>L;NCjM{
z%?p{US5JX!wk(>hG%~mZ^)$H7-<d4%R^QK-*o~>6j1D|EA?0S)Z*_#x_Fi5HsR0We
zX6cFGZIC@esuma$k_2^?F~D2TU%Y4mI}9IY6SBFA(rie~(2mC!J|M9=Td0?*eL^rd
zGu_^qr8v#nkeXjf9vHOe?T4Zt{vwrC(Y6n{bD{EDfO#=-*pCd+;E;2+0@a9o89Jkg
zdPQ~xxsxXN2~;C}t*jhp+g|Yo>*M9oFVX_iJd4$7@Rc0FZEjFjbm^xdqP5^)M(bD>
zs}tK&Rl(L5fomEE<}V%cb$;`oXQZxP)tv(Y$p{g+**OU903LzW#X~`K>46CVW}${9
zb@!T4)XOleLWdpXCj#rYNfKBK^`1X_wX4QCys4=sC|0<za91%c&KIL;3HbRS!*9k_
zQqtP<PVU#XoA{~T(o)Ec*=50U04xMjErOQ$V&vZA($@k2Oa;n_0}5VgNI)79Y5m*j
z14^pfUx1_)W^hwM1ydZ88Mh5kr)<EEJ*+7Jb;t#ry#6*43df~@5vZ<e?`4fajA&qS
z=u#eA{|1~El!cXkEZnldGxN`%#S(60WRyN0q29HtrjmF{qM4{fplXodk*!w{6wue3
znws+CIRF=@!12@3(GiGNz$d^<@l~8Tz`zg(^`CI+2!uKTZNd=Xlk14I%eEen{#sRK
zhJ47{8fJRUnL;%6!>s8CLX;TFm4e7ucf4av=LN|Speok_E|~v}1ei%~#QVvUaeyW4
zHsV2clk$ji4`QtWa_z`4^k%Pie`k1TeR(<^FkQ8G@7@6sr+nwm2XxiooZO%eUjqR1
z&*&U1lAE2aR^`t|061t7qCD57OFH0?0i^*Naq7gpo2%;qfFU5KlMVP0ekuN;Pp%`0
zm2c!OXKks0)~o4P+%6;UyD{VFB(QA5IBz-?yV5@^uqK>!udXRR!aeEfQ>w|Xd{toO
zlI+~U#V!1}G(DP8#j1Ms>SIWGSb|j{#J`&=Ue<?1Q&^!q{}w<ayQrwFHe^A)fW%35
zqe@_BD0wK@T<Qb}BORb}6?DEpTLh|<AtI5Ood$(eH{^8V-G#Yng;T<SaJiwp0*bdq
zNQ#6MQdV<7_c9Lofhu@ma1IH`Is|6MNb>H?&ISbAroRr>Y<HwVyPt9B;C7v4DiuT>
zh}L>~JJK59ctk^<6SS^nVUY&bA6AJDU^heZ!h{{g>JCnKN7rzc?Fl)wcS}oca3lom
zC$vzx;b*wU7%=DF19)h&$JRXCL3Oq5Mq<)Pc_90^PmQx){=`zI;rr(1+D{@?RfOu<
z`k}?{f~)Cf1s8WPCP;ni`E+vcTRw(^2jih&MkfexWulQ{7)e7}yE)Fj{Fd8gcrVZ?
zrd?jdccc=A;Kk8L!4)GV+}kfEWH-pf@AKm}n2Z@$gf#b%FrN7g2?_{g_`{@soY@T}
zpY?t+{+kO0ai02`Zp>sfZxQeW#l<KXS<zqy03jht$A1AKn3r|9p+&8YIKF>vDlnR$
zj=n1jkW5q_*SMV)61qvybI@baB*z5OA+AI;IP`P<FID*eNTAqo^+9{46y)zHF^6-*
z=OiL3ss~wiD@sg&DtDXtOv#(D4;yC#xg@_NsBQ`%8$$vNMDao#gJ7><3z>`%V)h}5
zsH><1AhlY4f;N_504CNzKG{>ObM)v@R6!y1srogC8Q8Ug?P&%dKs(tH_%+^@pT#Xl
zfp@OzZl%DmaeZ*4q-10)wIn1YQ27Nnhe6Qpb|$7eL*j%K5(o&wy~C9F_JXu?2#G+z
zQ7%FbX9Cg)M3jKHKO!hN5+1yB=T6AUUb1gOq7C?;y<`rJj=nx?gpR;W{%7^y3tU}a
zCqlmKH-w7k-rio5A0U;=0mw$S3BW})K=%L)(1YAkVgBEN46Q&?ndYuYpAr&!SD(tb
z($&WW=K#B*q^9-;>ZVIb^ai;KM4KikRKkYIDJV?9O$EdNk1mbMZ3r;jHE_*dpgp~T
zu(N-RDFiIc)$$Qjb9+?410vPMP;GGSos7IZ8_Y_SwjaZv<O1;wK->z3^&;~x2G*fo
z90Qa&otTM5_CV~1ll@<jx{lhI|7&)EVQCWh5=bncA#ePkr|2da1l$faQ0+B`p=uEJ
zw56xmCGgFg81c<9McyO^SPKMgw^yKY;|4$sL=DQ|<56mxYEar^;`jBAG8A#_oDfwY
z2HHy{!UwEKEQWsLP@pdo4k9BZ%{C+O#tQ-9fb?yPkh3FU3RIq{#B(4s=sQL}YI^M-
zS_RG8-&;ijH@M-A7bKW5DOJP4+2v&da0x86913ACfgWoPgZju6O89^nsHYRar9zl2
zfI2mN`0JN1ry%PMIKar5bo9!T7K>S{#KFP4?v)tEPr^**_6mhiK&b%MdkY2X8`uaC
z7+$Q~xocMw<P_|(2-yLcr?te*Me_|cNF{`&{~5;gf3OD(YoVa+KP6*lmkmWxc;$Pi
z;lowS#m<XONU1|Gg58N^z4c$Vh<{x(tt{R?8PLj=pC9O!)2XMo7cPjs8XfE4kDak^
z4}lWQT!<&t{APgp1p+$kM?OhJi#_+<JRxF$%5qw`dw0l<Tg1o4hAUyQ<8&m9i;7Aw
z^R3a&T*XUo?q}n3auV`)atib{h-!|X?8t9U8oVnw*=5o?)*Kc_g<;>1N<?pEw8uOq
z=e($usZ~>Ql3jOp=i_+APTY#siyz_gukV7ulb=P5`N~ree8<s6!lKPpOp^eMT*}#z
zR${$ioFXS5uBdRDho_-2`uCS6`C+~%2fFg-6fdV3gQO)=1lah)tgJ&3kF^>vVwi*z
z;OplK?4AXOhR#<%v#`(~ZyKsmQ~)jd4N4Nm8`9pi7rdUF{8n*XOGi=uwSz$OOT}Gw
zZg)~9)`ectVVD{Xyh#q;#Ioo|_%K~;SMc%blbGxY{;%dQj$S3h@LX_J)z2V(cz9<d
z+4{hNgQx77GzZ{t1k2#1U%!5@x1}z~#Yl-^D*~kaAz?>)=h6$5(aOrwft4Tpvd2GI
z(-uDID6p)uVU>n9m)9Szsj96r?s4PJT}w1UQVbO9JGgJ#Jv1*)->~q8i=4FuwK$&p
zr*kLOclfI|NW$S4|4zlzc<SG7I^oNW9cD(c@b0L7_RIXj_Js%bAE)2wt=rKIY5R$4
zjqdqBUO#$=nx8TJ=T8HhDWIH<1{Td?vB*q=hUMw|(+mxGy2!@S7)>jgsppk?xhmL=
z9<O2@DXD<l8<`G1bTlnyu>Zo6^j)tCKUfn|4@kVO>fwo$JSGv{_{=}JC~Ufg2UYfK
zIq*qVmt6dgXPB>a)xFpI_Qmwws)oWh<fWC@$XaWS90|YEL^mlTds?}tboIStFLNDk
zZi<GOuT|SNQW4ZY3n^M$P|twwP&wZaB*>o}xKAaV7MZUVR)UQs0pL6IhMYo>oV=di
zPCVDILw_EX5dL-d!jyEap{49lcrfa%&(S2w`!UvR52mValNoA1l|8;-;=E6Z&6<ib
z<TSh+Dzb#gya~n>lXjEm?OOKd{8I(1*hmv}{xDrR`C8|ud|MM?*Sr)C+^B>i>G*y)
zKGl#^5~=DB{>3H)&tOW+eyqMYs66E4sqGd{iSA6W1mnzX%Bn3@mXsR6ltO=mbC)iw
z@za*!&<hV;|N4T6NX@h;y)T&+3m8!f<mGdwQ?4!+>BIWZ*?!RM##;8G$}(2erN!gU
zdtr04sfCZBp}_@qQ5+<T$by4Xb8qa7-hW_%>9o&juqlo5QF4VN<Vu2&vKG?C#mWYR
zgnv^Rt!2O|Sa4Jy+v9%=4uuuzQKdcB^tAjYpL#Bkp2rAJeFJxeDKgq!(utX>Ep8R%
zZI+TEN%&2$^z9m>4`YTShj)GikrU5Nu5M*>3odG$2DNmg(SdsqiuiUKZoW0P2oxcP
zX@2~RFUT_&x|nC)Ug>mL!mUd7F^1(muhUrK5V|RM1BVm9lAeMf4oTJiL5GN1rA2Ux
zMs;24L~^%&{~J*k_jf8@@gx|2$Ma+>j6MV0+gk&RoKOU<3h|y-?|D-=9klxj7jN2(
z37iHJRD-_)YY&+HZS8^AIl0zOh-l8hsvGbC<UNmxH?S}Iz<Eyt4lyGGDZD!}ver+N
z|Lga0@GljY@zNE_b?4_5B0iaoC?p5G?#{N2R|jNr_C?8xV|8t<Q;L%<Cttnt71sqn
z`M7Lj<qMO9qN1ALUnd5H_Y>;-#Ms!t@F<7=??ttbN5OdtyIzzY`4KC7&TdLKgqF=O
zQTyqqzdFts`^lP;dqzU0A2)4d_T8IMva)JdefTKWW8;eUr-|Tr3d-QrC;O=&W-@~n
z9ZIsgY~Tcjr*4usKmXhEjOpt~v)W@?)DpAR=9YLbU77j7XwKG^2v1n{W+q+;op0uS
zL1L^6bx~%?oxsj!3-sV7PDmyX*@YI&71p5n-gczMqv*qIM*>MgIuh^$NbdFcvGLZ%
zYOu&w-=jT<YCs4Pz+6Du>na&--tD#2{9bt|)+oOEWz=6?=nZn<i?7ftjB@*X%dGlq
z^6N7RV2jHCAbFAUVL*5hkoFP7Kf;p%ugy>BSmUtQN!j&0eA$BjxBo>8B}EIJQy5Jy
zf9IrE@C7s&t4_bYe=Otu1@){QkzySn+rNDIa!aiIK|m=lBYu8uk;oku0pujeEn04S
zE$TSiue!Vk_uY^OZtsk)>tzG_`tQVs&v6Y6<W7#{@|h@bL`Cf!Z-`=~JfVrc7IfN5
zw{Arm3LQBjarp4zp&vH@kp|2JP?nDcT<Ja3b37jPs&FzG4_XZQJ{c%)cRGn1y_Kwg
zEA2BQ{>f=hPFW<GhN`>9O$3@dASnun0qEx@U}`umJ(~3~4$kAz6Oa2Dc-TEW5+8KR
z$MNjeOMv|hWQ_GYE9evY(Ctz3;@14!9OKf+@G(~lE09${nXt9BCG5VDXSfP%-Q4mr
zpMwQd1wFuP>>|+b-yi+<?IBqeW8;J;yAP{G=d`_}4zAr;r1vN);_7NfcW&a7R_y(F
z?r(zEW-A5+(i|53_5(g=krxq4J^N?Qjuc;@OGB4C2-ST|3_G9&;{i_4+0#@nzMTSD
zT3%$0utxjPx!Ro#;8oh4vt)J|KSxz=$PvKA5q@=C#!n5T2heDO)Bz?vMNS<bZ$C2B
zht-}`e1F>K>#uTCB2{n_xsnu+2Y-`fH{0my>au7I-jS3vauDLNWX9Y<P@6!Qu&*`^
z?CPp1(Qq7@;!;ehJBINa>n81!pwFOSTx|O~G9MEg$RmG{#ZPPKS<X#z{Eq4J(y^aE
zGb=7Ud{|R{e!GoC_c44?F4v`qc<23QSyY(s!@85PadNS-Mg}|Uu^LIdbg}x-yp!qt
z7d8bhQVcr?nL$s|Xi>O)WwJ=d1s`x?H>9Fz8khtv)J#mqhB!X-B+adg<T(A>!$BWK
z64^^4(QTSd0>nn^cw1^HYBNFo1Fb`mjOL#*neY^p@(l9r1RF~#jo(Y`6JEH4CFuAr
zaWSJl5vbbUfaL;i2R$aEfSfdeHX2kb1LJc_QWtdSvrGdy+V1Ven>9>hkro0SJPiRY
z1voS4ebGmzk3e~VR}z7rN9CNJ9wXLrYK;YJ372My6a>Hu?x`sta9q5+ZByNasMtif
zzB6cf^z>M<mP6>hsnRY~RbT+glx(g_2#0LW3l0tr;?*1fkc34JH~!_9F&4l;6vBmt
zf{?-lNSs>J9MhGVe%_Sbm?^_1EjF$ZE&A9HSaTcO^`S}BX##wm)om$E>C&hJrj*P0
z051j_MW|c*wJaJxctHZMtF;$vq5aDmm?ArKEz^NDyck4)&a~vTf*LP+5{b#kUID$O
zP5YY$88wq8g6N^7cWOKUe=@3(m8|$~^*)d?B8~Tx+owSC!mW{Fg}aF<4U9V6c<olI
zcb3UeO-c8NtX!>vDtIy3fK)+38l|Z!+)<!qTM_4knmGahu=e35#dk=SnuP%DDO^5J
zLGhIWashdO*KWeQ0~Vl-iU?RVYD{|pqKW!HS0_5wJ{DI=(h?@^vU_2oJqw|`#anB{
zbGxsYwWTP7Hb1$R6wfFF-~7<^(RE+41R>uG5*{F#1c==WA-wVLo#ie$Otk=mLmY=H
zv$AnIB<?Ti!Vv#(YnMEnT|KqE4R-t;*yPw4XX1o{E{EcKHE1V*ULekcniQCTzdM1y
z-MRD48DYuKC7(0DeYn+o>%<F%zLootwEP{0kw^;!jLGOL8CK;8F}J%JTKqsHY6inc
z{oDz5@KLJocxwI{Ob_}U0ClScOtW;odzv2@W{};0ESAt38}%w<I+q*Vc2idKqG`Qm
zV21dk?e9wQ=>9G<4jpBN<^%+4u+&0dL)_MR;C_5AU%7(hn1IL`%&kCoTLLiH05v87
zzySbZ@~eWzkGgp~%m72*eVB`jQw8pl#a+BGE*4DcmVWf^h5&xo6)Qw#p4(l~G7lJl
z9kfN7!nMoE$|7hLXzsA#)7shB{$Swv9j0`F+DbbqFGHr>E}-REpw{i~eX{(o2}iiW
zoF{jyG1)?>Wl17P_r3w|3Lr=jXl#Kgb88>@&rufoABs=~!|^oDwBx7KeMG+~g<EF{
z!G0s@O-<|-r`8$wLlKmw1m*&2Baj%WgJuz(aOnC1s;E|s4>WF|2J#yM5(~6rZ!QUf
zAPvNk%^z=9n5J0pGynBl@jtZ^%bXP7A-J1FO($C9nNoll>IJI;$YYq%#TUFZasnU)
z;a;A7{FwX%JNvZ&_pR0YCpb6=Kh2^oHWpsLd3BVYUY1Au+SZmz=O-ioKOzns{u*&j
zE*t8GQrCkC35G82Z3(Kl#_0x$Sj~?zDIcsc^(NRIdA{Pol@ch=fy2~*=)E{w#iqF?
z;^E=JBJSRm;ri>#PC$EtF9|~I12RV1i}?K6Gv({oU$ArX^Q&EXvis~TZJ+7ECD$$I
z0pH-@H!qL0Z%tojx-`5k>o&~uN;4yBkb;_ju-}_6oV+D-eCZND3Algn2X6#vBRjxZ
z^nl>3c@B^RB-qqlassXL8gy#dP5e9yzY9GgvOs?o-0cC;_5$EDX28-3Xq+Dpu32)`
z|5ho;yDGtTfh*$Aub!P9vfTM}#KNmA&^WtF!r(LndO!>6j7Ff>&kTC67=>(NoR=mN
zLDYui%3xp3PiUB!=Yftzc*@PwK^>==c-3M>BPLj-MAcfsq{{AbcOGY8K7I4FkwN>F
z9l9CPg%r+HUHN`iI&g@9`a^zM-Pz*b;|K9Wd3kw^iTf8d1D9MgT@Eak1unpqUtsb)
zoNycb!ZMLiKU)k47J{ZEAMMEhc#NN4y<@(5O1XvXkI+Hq2fIQ9ARiHfceC2<pmi{z
z?-!Z2G6&<hzy{yMYFOuwG{elKs}{oDO<bW9!3w)*d3EkBpVGy<YucK|8f^B;O)`Y0
zOmPg}JDX<Q{U7|R&9Z741?)d7C^vq*7bR+mfyUb+e~96rMhpL@$I_J8=^C!t?Unm<
zQhP}y>&VHUU6v$aYiN?2lsz8Xk)9~<E;Tgkfktolq;)XSrta7S+dAEK4a2xXf#e59
zVlBh>+SCeyhVGdaoD9TA8m-k*UM26}+B_>H^t{IR`}{giPp`sl*sA_r=<xUW=KT7d
zO9C6IHpV}0hc>CjQ5}BX-uaE`qjqeQN}_={;duXHF=-FNxuPb6O-B`(_9N|gHE+6Q
z?!Fn~W>Iz0+nn2+tJC1*l_xbz!PTG2rJwF_J$;wV?#2yY03W-m&YnH{;@gU3LwY~k
zf+g0E-%2xb92(dfko8%3o7B3MQm>a?aIF>h-Zv#JS)<W(jlC@R@&$=p(2s3{R=vB5
z>4Lcui1;>sewaD1;G;&q$s}<xvefLb{%YR#O6`#8M&+HX5#HY36~3?}X1v4Wi5riz
z?$@`Wwj<P3dS<s9s1;Bw-t12Kxc_gz!LiALOY>dRkF~DuUj>Qg+tt1k3^EbqD=Oc1
zj6ypc5(=`MIOY4N767T6*$E1G@MKH2$>oX-d@6hY=BkoC_(F%Bt`8Xh$oEM<74*QX
z&X7^Yj%ZA8Ao=@y75W{4pTdDK1+p1=z`0b^)dRPtx3}th>*fk1bY{!V%vRRNeKU8G
zJe0^=a^R`sW?N=cG`+DIJdw$nqwaP39PGQ9W%5Mygq*K@pcF%FWo(UD4X>Fc+Pl-B
z2EU)i;0lsf#J%}vf+l4YlsfbBE}#JFj=s`kGeJAIUq<mV?=;0I6HYjsQP?h_SD&q<
z5n9O*b+%>k#9Lx~+O?73b!^RlxWB+j@yHU+ll`ttB#=<J4G*Rk0wMX+r#63{xss1J
z&m1%FIn}`5|Ff3_FYYzLJPMqpF~~{Rpvj+8)_E#_#I8<&u#&IdLw|u<`f(i-MaW6!
z^-I`H{7V*3kmy$7;NB2Xg9g$Q0rKi%&QJGT0VnA2jpxwb3zi3rL#!sI%T*$~2T31i
zyrgMBx&csdG&MD4gA5SGtnv1&_|niJ^{-#k63eQnt0|6`5Brkd*zdQg_jBjc13gc=
z{Whu>0lSPtH4-q<h@1b-dpbju+vqwseaU`XSNWCMg)7i*cKxu5Af4Rp1!>P_>X9wp
ziLj}jJ!@iRBbRw4mZ5bWTA34}G@yadaxDN!z{PWG98%iOjLBEPkH#}hZR%<D_v|?m
z`MB$VA28peMe{#U=;1eM`rYkuv_ORoe@4dRpmg3c8P%zy2P4ygxJKDMpuw%^jnD^x
zf@A4gYx)8-+O8KmVGVair^~69-7%6AgQN>*LDftE8aM?SfD=A^xQse3fPS8uzDFYA
z6J1J8Ht~EH`5V}(Dq=&_BWQsG)$HVS9ER6y?Daa6tj@&~$jGSc|M-5HW%;q+f(9Oh
z>HjA%89;9T+87F2Qd;jfK_{iI=CQKSdgKD8dBk;ZBw1L+QNSs0zPo!UO(xWrjs}bR
zo>3i@4PcSu#mgpH&4WL$#`iMsmiVkg)j*P#a4jiq>2z|MeIOZj+28Q>l~42+0w?)i
zAH8C<UVB&P_VKHsU$6s3&oGw0n%X*e(?gxc@nh-)C%-T?f_k#Cz;TXyXD|xOV-l2k
z7Tw=}zhQN&Qs$b1x?knlKVJBChf}yC>$Gr(y5I0^i=%jnHlDQEy#cC$(blkzP*}$g
z)4&OPP672yEnR>6!er**v$=E<k=#$Yg2ojN2_^d22S!&P!ao$;8~F&|$`u$SM?;e)
zQ*Jx|`)%u00~#jzdL6-BhZzV1Ht_Kqm9S>crd}G%^;y0bMyltO&tAdxVD7r}5W{vh
z$J?8!<o6sYk2g?pf9D~m8+}_yO%B*cx!f%M*XI_KLY%w>xa{ZYw^W$%Wo2D@M`rd$
z?hR4M*nE?dtG=Xcj}HGaAC&<WBQ~Sp9Kmh>JrcKWP4UvLye0E(ur*&#G~^y->vWe3
zNYS|IqO<cmM#^aG;8|8l#eaEEry|^6zGD){7d~ztz_6KPr0`_!M=~nS_`L0CuOzaS
zy(ElL@vn4#Ci66A2GP$`kQ}jNDcp`uxqj|N#MJghsrdVu!uSEY$i2>fg<jn+rG|pk
zx-EV#3f^3Oh|v{X@7ex5c}!v+t=PS5yU4}#k&3ShPm^M97WyjocSYNeI;HT#$0d%(
z%)%4qZ~yLzuJV^Gdp!T$JN|gWkmE9W>U4?iXsG3LyTNPuA3hXtiWX~)yeGELv#lRH
z#btA&bF{6F{ykbqz;%n;-~7BK=3${gb^VCqE6q-f`i!GqeRbSuZBCwG-|g#VTyg|3
zEBV{IC?7Tn_8!Qum_G~D_}Va25CINB4|F3a3ERZX8u2Tr)Arrv)=X1_kpj@$0L}9+
zq=_5AvukQ-1Z8GrAp!Avp4Vp|mMwfk<pEmq*n|WN_-IO_4R&9{z#ttFiqJg_<emm}
z832<K7#KK|rmCXy0mOY~pfCi;<PnITC#x7J{S7Z|uyn=@{@kIkuelnp^qGqPp^CF}
z#<61}!GxzzpK3S@*P3BBGQbmpLplXyD74`==2;IQv7)~)O0YnL(;9UJdH{fF+-z)#
ztf2Lf7=<lA74DLz{_dT<lvJ!-CN%rNpO;=+93gMWtErI#wlvTG_3PK_H*bz9-v=5H
z>0FS$!iTru8eIGW?);;DOczO_{5hGav`5~Xoj60LZ?`b247}O5F}o!2O2K;e?@(Lz
z%8H}EGf3^=-+%?srNPT<Q8}IiiSaQV9j|LUvtKFTOp=n4%C`yxC<!!MptZoCL97!H
zl6n5zTwL(yL1hjg-l4lr4MaMv#V!td&{~dEAFzQkqM&2y^ndMo{``5Qqe#ycf2pqF
z{KGRym+4~rWK$xqZ}Zshbz&Q{uHXu_>P74bYV%Hp4hyJ-H6J`kMKmVx|Ng=-`lQ&!
z&Js#*bhRL;-E8RSD@WU9NY03rqiG!oS%bnYxuirNh(a|`@}U$GwZlUWU*WxnfO?yt
z^$&GFqmBn?4%_Trh%DQoAYlx9J|o0D@s-9AdfPlfD=1Qfx_h?TVekbsGlJ4FC_Qus
zM&gV4mQ_AX0vB3#hLX|KmqMKclfWO9IQ}ssk_*GA&X60Ef3k@G8M3Z61g5DkjhDW~
z&1=sS>0C5wtSJP5FdG>9Pc0>S7@KrRqhF%WP*NSk=p+t#0(tcJ(}Jk9|JHXuCbKwp
zRq2kmC2(7OFThkXDg?nl93<s`xbKoHDBwI`HUGya;a3~}^7%b4{!KD>hV!kU&9323
zWS$Q>fsuMk_nk?I!nV^>qMVAKbVu|t?4jft%fSQRA5fC+1cylV;yFge@VV%f>X}>?
zDt1`NkuzMRN3_W2r4>GR*&lfcmq~Yso`ml7L#N~MzwRXB%<41SQXfT6N*sX&{_*B@
zX$6MbF@KV<!z7F(N8ifWVUOAV8UsJ?d++l+rbD|vvqL+smgF9|D!PYmS1GZFB+Wzk
zpwC>)RENy#Xt18v)KIg)=*k)rewO0^y02U_Bz9k7Zi>6nqeHg6LP$vxUOjmkzF`<K
zEHj?)YH*<JlYZxo)%w_H`}O8u`Ukd~^Bnx3=vr6oqSrjQZ~r@9$}pNZE7)Q(-&6+!
zO$VD|L$kizs|gP#1kB8Uc{HhipCv9%mxgXz8Rhq`1cHj-qIL6;J{E-cyOw?Tb>_jT
zvYAbg^+>`Ts4_}Mo#_&sS>g?*`plrBM1O%Ll4&#<>i74T-tt^1r%t1r^bK9pGGv>f
zwKztOmB}dN^b|Lbb+k4Fr|*nh{aXWJ%CH%dGrD(=3rcF3XkwiG1BN3xfS3kx-3Y|;
z*MUfndEBO5*f~xl3Lr`xCKCW03u-M*Xmo@Qi7p(V3H)wrc1Rrzs9gg<WSc+lz{ghh
z9dF>!g7F{VMc+V=4LA$qt~_fbm68bu#f=F}@<FbUJrTMmwv55_u?;-t_V8bGPY~Mr
zIIi~_KZ%y{hrvo$72;K))u~`2vleu?Cqa|u(vsCVnb|oR5Ca{ek_z$7ds4m9)zsBz
z1$Tgppo>&|`Cs%w;D%(;6c$jK758je4O(53l2UQXH*@Lub!R)pY1~eDm5B_a+wr#C
z`XSU7;3-iXby8|->#wgq(33T)JOa7^2Y>!4&?1A&RfTaKY}(Y`S2EjUGS5|?(e=L&
znheLLjESm&<0*(Kp;K4@V4FpdzM`JoKb#RrE#w29z=SMlvVmJ=0`ni7e?8*>1ax#H
z$UFYSvBZLotdxQdb;MlLzz$vg$KL44{x`v)+`#T!)Pd8NfFs^ZW0r8VHMcqSz&_*{
ztDjQG1i#RQv%S=KCvZwR|CU?PVpZ_2Z=cHJGB=IR-Ayp}cn5XjV`D51PFM?&QWn7~
zswaHDE*j^q->0X=JX4Kvuabx&mLUnh%Rq||>=vzU<lco84bt=Y^W0G1)$NVvqQtA-
z23JhSQ1Dg2-S$gj4N0FGNMi6Gm0YDVKIpjg@bB#w*8Zw#)l81r{(j-s<by!BH+qTI
zi2&EN<f*mGv>Q#<F`vPP!M9>`7gCBL{lf11?){Sm1&7*ab<M;?<?3T%<7b;@iEp=d
z!3OYr?|*52A#ttBD89Bvw`yOl1!)u>eMnnNxlyry_3gN+rVzeQ1`cARS75{)%OKT2
zGAdqlQnj^mRVV6fl5DWgD9x6j=H?ChAi8X>Y?qzPEHQH&j%Y*qf7lXRvZHidT3+C2
z`95mz{4SAj2QK<QPJKJ__T$&1eGcEjHyqJ|Khn{m!~4e{;mR>`>Q5KRNzu4C{H@On
z-VzeX70{jjamVM%lqk72?e7;Fzw^-gce`k5oIG5qxXilmgXb_z7Q*PRr)U5JW@>CD
zdg1mW4SGKVdOuxNvA0AmDjDd0<t@`0-$a*px$9<^3svOa-+$IUKP^gbwaa`6uJh3E
zwacR$_UGDNA9d@}9eax=YZOiYwBW5`h!Oh~>hq;*A32LOI8OX>Mkw`VD#U?0vK8hZ
zd5lKd{bx@exenu#R6x@LoKgb-Gz!k{(7J&JnEcVp3F48SJ*ii}-4Z<bm`<Y~!=F+6
zN2!;xC6ueEAtXZk*~<zGUaGwivLp_n?!v$KOr@a|m?WCMcRL|7Qw79BQv)ELQP<FT
zuQrWzm!Jef-T#ogtYkjJ=su!rMSYsl&>|qn$;BlH-FRV0l?&R<<(R+Ci}GSWrza&n
zC${ZIxe9DUxe{<YRZnfCq`Uwl6B%bed9uI*de1u%UdQju`#bfyR0BTR6uRJedi&@`
z?1b|z>dJ7BO3#1+mBW~V5^%t1E>UY+n<-#$0K6?EWPu!MxxO}-PZP2ox9;m0)wSOj
zjWpVSxRq7{!3+kM)%^I;26$`>paTF=o5FA*?Go2~P_+iWd>QFQeQ>R?9iyWIi=&H!
zlm=RL`jo+u^Doz<Nj}iu@78@9qjJ374>s`ksRumw57}#Y?eAn8U^oA;9yPxGAJZ7B
zeXfoe`Lk;tf*92aTJO7AvaPLY1DWRt8SFPxx6BUG&8*B|RL9`5c@h#v<EOU`lalp@
zFoj2Z_TY~|5~{7_l%H>diT|n~JMt7-#~4~tL;u6y)pSow$+!eb$rRR*gzaEs#UvkN
zG}%Xz)iXO|ThyxWSc`T{c9Btp?fw0d`b9j$y_~u5J^a+@?+^c-!};e@H+=6!W)<y_
z1AF_O?JjO~AM5+B1>Or=@%mj~<8p|3Dc^CYGuSVZGNJ6GjPLn#_g(c>YMO;Lo)XgF
zlQ4H>eTy_^ssreh_8sglc*jjiN`cPVij;N60hY;|3@npel%$X&p|5S_YM%so(cO$Z
z)=7JdBUk$j!AE0#f*RbOhd+D&mzIuQephicHdc1;-ZUBSie8>w&MBZe#ORWkX{+zO
zhApo$GD<UiMv+%fik6N=CBe;@alMYqNoKD5y>u$|p`qcLMs~Dxyb>g0`8NDJ(K?T)
z(I1p=Y<zFj*qCO+&-9n4Zl(rw^8)$E;6rS*7}-&t^@R(~^NIRydg4s(8RTUCu?fLT
zpeOEriiQWmC_Lb(p|h+~ww@1qSzvIP3iQQ+dqG1n)}sR9%`=oi8J8s8+=_vHO@T%!
zPAF1f${(CCHK+w;iV-mgP_?`133&5{*WS_nDF&tb@U>__7IY-@L%SCYMEKrY2c)P3
z{)j|G<iFrl;CGM5S@Bo=Gp7P;K!dv03le9dU@C?_T)~{f`934SnHC-PlWH3oop94T
z-{8T&hfwDN&0(9}kMYuYG0A`GaZsW&4h;)pFdB+4`;h0uQftQ}%<uL6Sw&Zv`N=kA
z=B>K>_^$U~l>8T+511U}`~N<t`7byh*p9uVc4&6i-#ech;qyP$Xdm-#RrY(9&W~BU
zg?JO&M8LQDe1*{%y@2at*h322KaGOt#e=ddooL1u^pyZSo7^};Q_%q3)zQ#VJTy2s
z3h8gIzs}PjtGLpv<}qmb3tP8!-8hhLC24S<6|Q^6>8ixJhC6<&qNZpr2^#5WJ@{?!
zaSo2EpG8htGT~@I2XM+us?^q>dOWt>3@e_{NZZ@nN7t7^t9Iqny~kfXd9nkI`ltjc
zuWQ%k?8oOf!+Tw(MF*g-X3;vCl+sOa9^JDD8S-P&kveppc<+*0-ooo3hKP4{2Qm-|
zOkp~yU8B@Nqz7TGWj%H=Y;2l&YLpn_y%p`WB65sw@9z&rCvMXoSph5?_3a~lI~q;t
zAy^eZN-TP`6}?)%I#0EIL}L9XSQ?tpw!JYTjz)>0hIJ3Yop@PbUjQI@C$_y|i@E>`
zfRNqnj=^6spfu>8r}8)=ksBFjLB<L*Z>n=>R)Fbo?T~iVRNZ2cn7K3(W{cXNt8*+|
z_4LA_eQRJuW0lq|+yO35G0%gwpN)K!;md`-Q_z}au)U@U(+Xe~$u-o%K3z1w(S17m
zBbta%=`TN;4u^_{h6ZH33%1Fw-X%XSjf#hGbcy@V?6E&)E@@82x`t+-2!$)4E<dPF
z(0m<rO-&Bjf&srvU^S5*8v|@eodP5hYGWN?h!*;>;NT7{{^IxZ^`}6u1F%dpr0KA+
z8(pToQR!x^2pgY&D)B`DbXgF^{-p@la2@I+<VFA|g6Sg3NMUShZ2XcP)&{*-P4ztF
z_$(O%%CHx>NY8uF;2e&oPRb%Zjz!7pJv5OIwbVxF?7(=uOK2}1KHeZf4bsr83Hw5T
zCb)w%DCv?$i5OJs4VMAHP=}lcWXtP?oLHa00uDni5<3G~X!c8_>PCw}s8II-Y=&`7
z7!S`sMV`W;%ldcE6ifs35TwM49+uEr-KM9Fd<S!!Qh>^DZEZzeZXm`Xpx!|bL269<
zfE$a|IiJrDvf&pbUY|wD>woJp#n?ZOy8kRt{QrZ@|5GgHw<)1O_djUh^Z)K7s<-~{
zC9?l>56)3uXe>bO0i~fNWwmG&H5&Cj>u^2<rn%|@6P%QuzU-iiKbZlYyDNntwJ+O-
z9`>a4k~KLP*cPB^zyVqfki!+F70t<|m(cd{0PK*$Eb83CkzR3X7d~h>d5gO2-QLK4
zTbO!;rqO}&115CEL#CA!7bhKnhFTyX9MplJkn4c?0Eg@EgA51ecgM!ZCqqAb0dyv|
zHZ&ZO)y}huLWp+n#;_o2$blY99K>qWREI=nKSY;xyREBOgk4u=ObjZ<?6@Gg^A}xu
zi^iQm&%6*b?eaVtedoTIYyeFH5{Q&TJ^zTdMx5%efdMpl8Q_>gBAb}27D#r|Vg4CC
ztX7^k|4a816wV&)piyykErR*RLw3U5rMXu9mtmdU7cOXk+jhw9haT)E=-WY;2;&UU
zNMKgN2s~+%YYXTY>L{jXVnU;c#$iIVJppmufiUw9bss^aAQ~qEL)cf(Z_!|mG9Z#d
z1IZwdNr1o&^Ly$z*S$g5;q0o9#)cpm3E^`y{S{Eo4!c_bd4Qt*8>sVO*taISc8xUM
zL^Qb;W{YJ4yAPThmyS=U^$|83x|Won4GM<RjJR7sB@UwjvVmkxhJhk5z*G~p=(3}^
z7hHUN1V|_#_(zNXf9<_zR8?8pHF^MZLJAX^5*3v!K{BWaB1KS8P)ULkizFZ*QBg!m
z3L>Z=D3TEhB}g^_ib@VDnWK`ElEdN7McMD$-EVc@zTf@E{c*=P24fV3`|Pv#UVA-h
z&SyRlZ2*HOywt>DLFMJ0N;zn6lddTUQXfM#3-M*$)BH*w%Viy==$}?bG<oF22}{WJ
zHKFD4>L^_EmQ>vg2^86bz-ikeE-qeX;L>_lA4)*xvN79CkrI~yQ4V_YWDoVyYC&OK
zn>pS*e!V}QTR_fR1aC!LdU|D2db)(ye7_MJ(QKO-YPDvUw12(@RFzRQXhlzmgZw8K
zhw#44?ibywPRKhJX9RftV2evgD6^N2Rz}#6>D+&uh_r#4S0>dRWs0`wvZRcRz3A&Q
zj=~ZsJ2+&LP|y@%r84+C1vIF<S*~H(^BS79!X#i!Gbm_GN=_Epy?b{cxJFa}ykaP-
z27$2FeCW_2VIV>_C{MCLa%jkLpe3(v3azY|9S{hnFALtKq$5T~cI@&VAE%(^A(yTR
zns7lhqQWwCIR0MW(7-|pwjyZQB#Ly9Jl<<!Y8nJJj}J+ysiJ08_idDvmGd*UQ`*(k
z)y45b1|ij{n_vW=05l^9Z^H;d11oYTWza$j%+Hsmg5@S6SUhDy&Iid^WgHL)jwLHp
zCnU&i6SX6Jx{Ql9viy}*wnVZG!Ollhjpi#PJ*NH`1?JC!I*AxTr{y&Pu8dajWo*gL
z&Msrtls)|@KAr`?4^09}**%#DK_S~id;0qN?m7*jL22>nkf$aKo!v5gd)35I2K2>r
zI^snsLo<DkJmkA(gN$h3Bgzoa2jPa>4ZNWI#j|Hm?MG-xSXS}Abw!I>OF5c|vm8J|
zVDsx-RZS{7-K?yvfe@&pl8qLD7h_=$*ajN2tmX7hAZrv1K<P*f(2dy^KfXxlHn+{U
zfYoUn<vx_BjAXFjcT}+#?>eryxXl8pRhd|#Fhg@Fm6QqOwsajikHC^`^=3gaz+Qpa
zt5k?Jil?Ne%Jl`}Y>`k^hdm2c&Zruto7r44b!9dWYghN?{d+O2H}0rp%Vq<oo3nXH
z6?o9iM_D^{#L6l~#GB^vN?4;^VT+%+4=2PNN0F!&^dpsGL-Z<&3@?aCLMTI`BQB$<
zo?>8VXbVxtEz8%j+eZ++;<C#tnOnN0<m(&p5wy99$WL`JOwLLl^RcbS%XrqcYSIxh
z;t~MI+{+{w1g`C39|?ej{4?+1;NatUFz1KBl&{xual`wKJ&Dx7t?UOu61_Fl)yE@e
z9OfK}Xa;VyA;U-kC3rr0c|*wmRYNY$yQ~Sj)fepb_D4;hplt%i;RYtA&Lbtrh@nt$
z1bm`Og@w5=!??7x-!clnUoWs39UUbqWJJjxx&r->`)VW{k{^{anna~0$F_C1>I{`@
zVdpDhG2`bHGEV>sW|CiLEe1YwwSO~3Lnq{Ys$2~yu1%R$Uwdb+G6Rb3^<J3tZXJj0
z)0r)PyZQLC$x_U6iITg+zJ#Vg{=49D2i}}hL|N)qR8-3&OE-0?$!qg^>gsD9Uw3Mw
z6dnlf(S`Rl5TLmMg95Pektl5l7Q96Ptlw0q8V;F)txe0rxfi2+2%f?wFY@<Gyh~)1
z?pG3#a5T1x_c~)%MGC<}yG^)KV}MDJNf8h;;dBfMf%;PoIzd})0A<ENxAPGZAY9~f
z5*!A!siVlr3G+lXT_}t|Dv0(MDLK$o&F_}m+W;q#c>VgS$;{JRILj$>RK50d3A9dE
z>;iI_Y#;b(dc;Q^K3uUYB1g8EOC}n3*0(TUvXF?T{KY3x&fYGxS4<2{l$WjM_lKPC
zp0j=4zM^`?7NUVd22T7`x1Iq;o<TH!1Ox^HOJ<2W(p4_6YTV9Py|$)Xi;_zr8tI9+
zf64R*J2L)4_2}8N@kClyB00!=>J^vHsZ*(VE?Qth`axr)Ou3&Z%b|^wta1`;64I#0
zdYSv7utr+%cW6W<wbLeY5e6i}l0w~0n&YSR1e84=piw;r0f{c&tSF5{m|Y?nX^9LP
zn%y@bS_uzD<?QtO8m`-NIu}dvM30{{u3g;$!tv02Q$LySlne`$0s(-N?OwBwlN~7|
z_v^+Pc}Y@9058&A8kz-lsiN-{e4&q^zC{$P?#j96Ku+c%o?ahFJGZw*!))M@j8BCu
zl;J(GfKvEO>#$ei-Mg$r6qU3&JHHdieb$dIiG6Cf`D<$8RK&p%A{_+}0|d{NR3`8D
z(5_rNg#s8I6(+QvCq(w)!gS0+4Ov~Vi%te?7P74!Hgsuq?t`!x>^c#@?}KG2-PMno
z7F1G1Ke*Z8kz=KlimstsJQ-UcT1*UjJ(vSfZafBkt}6N2G^mEuShNzLuJrmwIO^Pb
zB1ns`nSqh)OILt=>iAIWDb&D-t_&d}!$iRXSm3E%7|1LqIu`ifo@>1KwGJHcg`Fh?
zY4QbdR2QHsO498^p+vWZtSll%)Y9;B$i<J1jqNZfFeSkkuJ9nE?1&Fc3~72EA8t1S
zZ8f2i;kk@PlR{Jvr1%%V8b-G_%CYiDhGEedwEI*;;qZtvA#Vs#CSST)KbA=Kgtqw7
zLnE@UId|By`6*O+AU~8sghxC5*v8;X-P8I#AiBFKCrAdd5S=J!w^+62XK(9uDVdEX
z{Rl+l;@kT>2~}7Jwq))eq|N~NcJlDNvd-vqk|ry7x#5-{9tXI??>g-*0L#&A*ZGJ(
z10qR8OZ#z&<1X4BIlrzZm^*|0ED>jdQRszWEqQjZMzh{|^egm{eb0QmXIBpzg!kU1
zD^@S*QBM0OFr*kZQdi&M3Zi2%Xkh)p8a{TFY#cn$AX;50<OwsCji`66(=BV7R-OT;
zq7Hf)86?8Qx{?<gP3+ynW-ruw`uD#AVB3IBptUS}lp^%csRrDi)$tt0&i(i>gJB6<
zf_+ph*btiyT_0d>LOj3_uq&)VvcsH(b4_`eKqHRv1?rK)`!M??otT5oMjkaGkC#y#
zQK%r8v-}G*7lq%v`Cx3!2^vny*u8a6&c2iFBR$<{9zfWLhs7($NVynXypD1g3V#Fx
z4k>L1K}GZP8y*M`VPD+bMj8U7T}NVJar^7e(?9<d3kkt*0b?!<muk>%4~YO5OQ1vq
z119_uTyC^Zm2K`j+g!DCL0{Mx&LS@`z7TD=WH~M8feo}~i8@Tly-sF<nkMn3F?SL=
zrK4GzlD=NnHfmyD(!;!M`*yd1`#MLCIMfcQpwq!Qqyhn2FM)wPJmyrppM3fqKcSCM
zgzblrqq73A8NyHzDrA^q^KPzg1g7G>_rbxblnm=t@}lamoTWw6kGxwMUR-%<sw#^a
zDl>mYQ-L>h5YktGg@J}a4VU^-A2UcT5*k$uH>xrK%Nr2GH#{}hBr;7<D<G;*P<0Y~
zJ8txhw1JPjTs;m`2l>D^RsXdPkuPifFI}RC7^LehXiR>X|A82rXg3@EtnN_67Ns!K
z(=~vjrz9s!osCB5G7llQ@zbdtYvGZOK?@t?enK^rZ@N`1xV~oLu8L6;`i`GphLeHk
zgq>606cAq{P4pBosv_i+-)+3t#6{wO9A#xatHVZ{{d>GWpu#5qO_5<oeSJM-8$^<d
zwT$nh(H==hbx<*#z?EV_+3F^7E{Jnn6S_6>yq%>pxJuErW4Hz1#{EW*e^V=Ru^}qr
zQ7hyk2~|%4`eT~LE=<LNsp&)O9WjHj?K#HQ-d$K1yrinNWF!;blIU@%M9_LLCZ_vV
zm5@Qc4n`+fp7W$!^erCZ9do{TRjOL?1`%Qm-LAGc6$I}2_X%h_4n2miI~;0TFerg?
z-kzT0`jTFy{$>0)1?k<wXpgWyh865%&I%7QR?A|S8-gY96Kn32oOM*j943hLyycCN
z?ajknAOo-IuN80K)@;<(fAIbLIe76PcncNckwFeHM4OTJeQ(4~L(yMQL5)LNs+xZD
zR?Pf>IT7;~eAYoPH}afesND?)#u?2psYGA~Tk_bkW7}1Uj3^BH0|dkQE?F|s7%M`A
zn^5+SKsqoe=nX|#=vyWuTU8>ioajh-8I`xaJHROS2plI;5GR6rCA4uN0ZgDGqQ3+l
zfx@eM{lb?pO)-mVZcI9|Lel`YWd}?1@>qw9wM`_e9&;XZmIs-w<9{G?<}O1Z$nl|K
z9tVLxbWEK>hf_5oQ_xq0dWZ%Y#4fh0BDSdoV+uly#{o-9ST8Yy*&vETL|zo~7o;9d
zA}eB>ix3eK^$L>xqL)LBI1eK9j!a({U<@mihb^m$PQO}4WI!=oBol*9kWz?4IFUo{
zf=))!#9dP2hTTOMXfbi2<-;(Fx^S)eB*=th9_%#6aTy{F4hohaKr1qJLOLhfAE2!z
z!8#;lET_y!&*YE1Cbxa?pn~TEc5Q>f-3q5r3J<m|8b6K>tMWJ^21#UFx07DSCfs66
zl>x?25vh4#N_gG{W<^8feZc0+wrM^NjT;21bt+t*j{qf+2}%|3-qj)R*^F4$5~iKJ
z>*P&<M<Sw@iE@)T>0&nnE*jk=HC0{{qMoRT6DeXMD@4wUB$KF=sg}5>sUu7f%0dqq
z^txjvvWfgXQdc)Hd?LQ&-@kb?keBjdpOK2u=3r=UM$AolZ@8y#+X%IH>klysMsS!`
zN9U2TO;-7xj(TE<G!I^XTkoUzpkM_RG?x(4AZq()MLUf4zj}-YAx<41ZGV4C${KC5
z1cxV(`rbCz14tG|dixnV1B)wNdiv=hQ!o-rn7;Q27!9b-)u{t}qZWnw?W*gpE^@Vd
ze<Z}mTLNYYb^1K*TdogWgxqNHOdp?WEvg?`Hq8=bBpGzAW<bB$t_p1r=|MNaqOqd_
z*tS@75)P7yCI%=gkSQx9c0_Q7J=BXX1Ib>O)RYtw<`ErT?A&@#4+K?F!mx2FHUdYK
zq$iq0=hynj6>yR6j~3BOvUqZM8c2djVR%%ZKYGf$>0z9~SAnl6G&DefsimBRrrE|X
z=*bp?J~4@a&dx*;m3ShL0j<7rYD;1*f+?c1NA@Nhy%YAgHP)NRBm)jeFPb<>>MoEZ
znV2d9P>=f6r+G~z07O{b5_^)c=xs~p&lBmtpQt6+3#4jCVmJUToZZBmVn9H0R1~|x
zxwqR1MRA)YvRgtZBGPS>5%;zcmrd$HL&dYU*zKqorJL~3#op=Z@9BRTN*B{Z_RA=&
zad(VgM;d(k*Fa~K2Rdle@eyly#2bY9L6lY#a?7xJFo4!SQZc}?1Y%SJ3wl3B-u6pO
zOv;ZwU8DISPk!vGI`v(D{jK65R~QnG_Jw6UGx;4aemnO(Ag%Ff$!d;JD6yeFp`)!G
zho&>4r$w{<#fMtml)VNAf+n|q1PMXObiqnW2TaflU#Wjvx?OS;x_2m8X+mp-=mCL)
z(Nd18guu*6yf4I3ZhPD7WcB6EO>@sm^BB+L-l!I6P8Zho*25}h>}XoP{%%kI`W|q3
zR%}I#N%Xz9S;8TaeFqEG-F;lV6>@A2pSz<q65-rnGGli%{QMLcJ%w*OZ`9U4@?X&u
z_o}8#$N@{|_WJ=g0-QDMW8{$5!5eiPIUXd!Aho`Yyyv9ohkz;(G*tw#21DPWsY})c
z;^gvFYa#UT;hs=)DH=m7dj7las+f$`LWNok%hpswm(&UFPxYK96L=olp`>g|*nSwB
zM^Y+F%jEl>6DP<ZMq*<S0`?qPAQeiecPHap@VYgb<_b`}HgS6zK+?abzk?Y*uC2@d
zki<P5gF$IQK1U8U`DTX>r0I1i#$i~Q?pjV+JzQr44DRBzH)INUK4FGl(`ON9keO-K
z5FK_T%6^1VJXEqktGn+`@>obcWy5Hrm1ZANu0$_w3c9FC^JNX>`|J>I^g)G)NXx30
zC}6}H890eVfi%dYHtSr@;o=%7h5;O(4ce`;`ydKjO9rH2XN!x8Wet?lSIg4XH{mQ+
zP*fjt9+9RO>k#;Fvprc8Q2~e+`9HrZo(6e1IO{leE7AN7-G;JDKWY7dcg?bEmp3^7
z>1^mD(ozOhB<Zht3M9qB`T>*={n9t4fJ%!hsq~`S%B@OVG^}t=n3n<`J#sI?OoqPU
zOkHsZ^2QXDME17WgBXcp%mf)IcB^8Ghp=~Wl`F|r628PeAau2I`)jJJI~;t~q46}^
zjexfoK^=N#iGDluO3U3)H)(n}C>@6Sk~WzMM{FCU;XxvQo0!2Ed!@Q5dXuMtg~>Xc
zm5@g45)fk$HQ&%IjP!(YjLz|bjU!U|lM#mj<dV@XV0@Ew2(C7CTRF=&;dNVt6#*+2
zlMrQQ<^n{#522(;xZ4B*pGS%%ePJxg(0wpF%q#c7gMIMF)o6qno1%sg;XoA0jv`jX
z;3n6j^Qa~xfg%d>p^p$!CC$8M5~9!+{f{5~Pws~QVuXC(L<%28r~+boR|BMN>#-Dd
zjdqZ0N8of!ZmPqVjP(K->n;@!T%HV$#6mi@Uv_+togE$(H9$**a3F}U<Z4KNDQr@2
zem|k=hk7l{wv%p1;t)_AM1j8BAsshITx=io`N;f45~`s_MWmgH|CkICFL#HLhKw{j
z4EtiDkN%!h9es3?)+8Y6eB+}zhSo&tRMBMH56uJi$Dtl^u1p#Qx~bldf3%j5Z7zQK
z(^@LysgntDrpQUXY>)g9`wMefdBA7PnR7mavAXGs5W?!VGubl(JMJnj&R0WpgmAtF
zHAv4gHgn1nDH5d!AMjaL-4*=Ud==$EV6mWw2;m$3skV0}%Kp{L{mkCxmu2J`z^iaX
zUj-dAs)DI7EVZ4NC0_|=xeOahh$tdMT+KM<{Gps3833Wl>FD&J5UX%D8oNw7Zhdu*
z&8bPjV8v&aKcoIQ;Og{Z#{2(d`A^n0I!H(N)g_{5cnL)yW0$7^_rrxN8HCvO|1WTP
zh4w!-l!V<E0!o6njyPEo1P~Qdu9GV%7q2c-50P*aWdf*{#*?5A+4g62fvZF#3e*7=
zxCX8$0W&~J`|6msH68aKSRw8;kSqHO@=+WF%xX}U+olRdO<j15`5GO{3dx%^TflRZ
zh%LiIL%ZZPFp2^|!oU%8SZ5+6HjeVd38G#BwP<R;8NH$+;f1miRpZw4%{Gg&dnmY(
z{%IqR(JN?JS?G>$wT-~fHoOPu0=au3IwfUyT?upWGY#`H@k<$t7YWHeOLS`wQIQbh
zcei>nt{1q|F4cSh9>?I=ArwmV^f|}T*@mBO$=(Yz0=OmTNJeY?-RTIy_h0cJ>mHxl
zOJ*;Hj)Bsn3*d<O9EH29DRhLQdzvsl{&gO$I#U-zdxUc8E>s>Ci=2Y`J5kQ<zRlpX
zK$s6N$uyP!Zt#%e<5Po+8+U$6;qDOmYeB02ds^E61GWKoq0~Rn1u1lbr2mZa-0mMp
zcdAcL&Nm67;RJ*}#i`x9XLiYQk&FZ^YeE)&sk=ykgq|5?eI*7Z>8_)YC4PD0?+AlG
zkBs;y!r&77lW8EnuCOD7L7!a+0<dBKqxsbTlYEDQxi#uP#0((H9m4>lSB#G#Qs$!o
zj=t8Pq+a4wYZ3Ql;q8M$WNBPwT!?%M5(3itN_acYD0t5#XX7eH#OcRYlCjL7bP-4x
zwGLJY;^V{Lj+8?l1aU~{BM7gOBK_Z*4x(3RYMnSigwrP-P~#wDfJ3A)W~Tu*^RdFh
zbQMVvP%Pju+RKcu@Y?TV>xWiTG6A^`raTqOvIO~@LFEt>PN`VC+~%{n&D&KG;GC0=
z9_ogNlMQB)o@%D2tp5yh{~>5F60o2g2Y|f$JxLVM4k0{;k{~}7KsP}|5Ch5eL6`$S
zsz>1`Noi@)5seJgEf0K41i*O=lB(u?1J&{cyJ4DXf7-D7MenFcE&Wy7cIBC9u<so=
zR-L+8b7q6Xm8r~K^=x@*t8dt6)%<FedC!csPGCW)!r~D>x8SXu(^tYY&NdWA+BzN9
zQ>))yf3KqFVNLwI?Mj(Budf+MSM+4?|KUTOEl=hckbFBWE#EEGEtL^*d7yjLbe3vb
zxz|pidv@Mqc4npv57z;iaZhjWHT13q<|RCSEC5`Eg^39i0INdqnjLV1=0^l6lw&&-
zL+V;u*nv0&14A!IX$E!mA%B>)a)h_0Xq{kU8tsV-A$VjLLaytWj#7>?8;82OIt$)a
zMe)d&A{Q$wD|-e8*d1*V8OIGH`rV0fZlzX`<WO8|Pkh=hguHQs3{v-59I=k^Cv+5U
z!tkan>S`S#TS0Mg?E1eyRMK~7$bVwO-?HwoT|vRq^c9phM;#lNVwlHJa8KWL$qQLd
zE$v3q!TvipJV{+mc@r#CZO_`@7`>v3Czw+pdO3wPjIuV-p-1bp8-;seiG^fPPEOVj
ziXv?#j$_-aoRt+Q*Hn8{nvo*5gN{`rPRoDiCZ2CqAr@A<iv%_aa4+#uPFTG>YTpBH
zVFn8CGeWZc#~ax3j<X%Z<@vW>ek<9^+5Y(PC&(G<M9EYOvrJ4a^YPoR`0C6@DjS9M
zHf3$t=@7ef#<0T<pUy@g@mC=@E3fAq^jSjC)chKSuP)393qm{KPT$=u<KONz8Yl*>
zk3GF7fNpS9J(`V`^*YpPUPCj23nC;zAQ=)QzFhyurz$TAK-nA+ya>i%pR#h000L9C
z=NG713(U;&_+8aZD_H;VS7v&#Ae%EH#(fHhC#DiA0bue;OG}$z5-oOhpkB7sF$^vn
zywzmanlFxI#-MZ+3s{r>)bjHCPlxFD@F>=OtTYobez4wQ#JHC$+jsdD^S+9X4&(QD
zpT=B4$`z56l<dr&pPzTYw2WuH3HMTtaZ*pX(1{;plVQE*eXAo71E)+d4_O55)hiVY
z`8UPs@hdC6#yF#~iHRMa^TKK-yno{YAQIi?t?_EQrMv4*lvK^E_rmbUAEIVTt^>-`
z-q);{Vl98Sv5B{2V~)PcZf#=|gigT_PR2EBl=^pFh@2c70~RLhP?3RndZR%}k1zDq
zYo7sq2({`R98{-8`H4zyDN>l)*D~TwtGebWAP}Ey%2{xPX7upnZ=JE3C)OPDMc34c
zvGMUzIPxy&l-JbNF+-+DvNs<^o>JIM>N9XfM3X{~T-k5=_=@vpE#@T_toKTN#FD;!
z`vz{RBfn9xEM|OcK{BesZ&)Y^6UhZsk*g>@F_9OU!$z{Lj^E=6>zxrbY2@!0j8#T&
z!<9?k{uYO#qoZp+eUkdt0*YrDJ8~J{V`)3PmGfzV{Q8`|Gt8&gT#}})6Qk(7`s6fE
zXLTK$t0A_Z@rxh%#Xe<7KA)!8iigB*{?v7O$M3x}40*pQ)@|8QaCYx=otp9G=MMNJ
zi0BnE)L+`ak5Bpy9o^kB<I!~#;e)sb-%=UMbBvX+T(fDILcyGb#UphN2%MiyF1;ID
z`KjVw+39z-v7uHDsY55zR-Sv#VL^#pmHlxrag+5WcI<ZJ-SsB9G31|M1$>gV0$RPS
zGI!uxSSbnn|2z%%@+B7E-=;XmezwYN))Z)5A{D!O`Lt<+)Q#|y0~6W-;`^c<m*1Vs
z>|>`Sant?PFTDf&0=L~V4yQ@yYB5n*Ll0j5qUPfA+Gy_J>yeRbj_T@6F70yJ0Tvp$
zTAA6-eZ`CvEytFNJB)UmRFsru9}Tk8q0Q<gt)Q&k=d;5oCEKTZ|3Qj{9=+J~(Z@?K
zZ`w`YFw&O2$$dfzL*pwy2^9Tto+05fMZH+Zy7^=?l|EeJN#ny31rt}vHrexDPM)K>
zD(h4t3~xnWEcYxL6QTsJk9`|qWpm2fT|c(A>~Z#%z(z^dVJ=4uVP9>%*SMnR<v!Zq
zP9<imUYZ$sr1I=p@5PT2D{58qOnjqgm!4Ea^k~0xd7niHgNT6NL%DTIlWCgK>fxVQ
z6%uDZ?0uT<zYP6FiUO>;jSt0_#ciiqzu&H^CM01SIOHhLWRSUiCQi=J^yHX|yVLw!
z=1kNwxtgdUnwMjK@+zzbz9oCpQ#1B!U0ieTzC;*qSd}?+NhnG-WJb@?Q^WVZg~sG4
z^_!D6MBRS#JFcZDzF`&9a>m=2IydSnbym^3dpckE_P!HdSL8-}-QDlNnkChG=8%t{
z&dGgr=DSi$UN|<ql-E^QGp_veZ@<1<{F~Q9F0;>B((|X2_R@<z`u^ls`%O>g*kpAD
z26$&qf2+!<<MO+}qj+Gq5F5RDr;fMvq)|1;U`TFxR@H9-6Xnfu?wV(9RZCn>jwgRR
z;xe49Az*GZ$0%xjym4a<_Y#Xk;i*lE7p?A%*bVaK$Mt$UA6(F!HI?*>TS%O*8Q0g{
z@gmz@)5S?nmn~cI@h`E-YIh`FylJqGRm@f&+v{4FkrB}OE%S$TaIvFU<6s)?^L1yL
zrqUW0tGG=`$8x?W3<dH(XSs7g@(G3a;dPcqiMHq4g4S^KH-;xTW@R57Gm;!F3_hUn
z7@yPjtm3!2c=GEyF=rL#HL-0$G5InJFRwLeWkq*+Y>)19H#@SremC9S{J53NZtP{V
zYuh5w9T?qIb529;_rkmxYd))Fd?o$jr@8G`>27%FYAy23NK)bM!(HS**K;?;*&iHF
zyxwqBKr8kA#e|{XTO`k~b5fE}7tSvzuRPuItR~CC_&Yvnwc}AE4hmRe(O0Ro+CGKD
zP&4;NW_I=kX9}gkeARO2<ftIZ?zj7uDHi%z>q}PxDb>%GSVS&a%|aLJu(Dcs<@q1W
ze9FR`Df2CkRi&u9JQegkoK!2{mN+Ys*z!RppJ!s=two3fYxwB8*!**Bu`zDn?o53(
zqJ4GkxVYQ|wD}VeuXI=}y=&UJL}Wc4&X;aj-f*t?gX(G)>gE`y0lkCPzN3LIlm_w@
z5#G}gqe2Vcg+@CWy(Q&GuW7yp1jtruBI1QIuDJoLdEDVap&jEkS>L#M%4Doggd2)o
zPN}f*3HE>1acMR2F_{=A?g~u4%Wh%9j*1a3!9pILt4~CfojH~~U;3yxsFD558&?&9
zl-dA4E$6(o)v(MV?X<!6U`(s|mCSkih*FyD!rYHhkSr7cdeo%7gEmn;&E#mivh}(>
zCWUdw-A)G=;#<8i*^Wy5IP$31iC~~dikfREDP?74y^aBEz+9gr?0EddW@gWOno)He
z3&V$IlZ~!M^@MHR8qDrcaOL`*wj-&9!YYPe&+`p~_9QHr_a(H33dMdt8dZVD^SQa1
z3Us>ik3E%#!zV^L7i{FkQ)M`3-E_qkx|g`Sgy_qDp9tDp=9RfHR~43VW!D*8T6T80
z-@9v-2jt!ja`d2uUM@jhW|XQPBQJ*n^~4%JlL{S6uHi|qdsQ(df_Fp0N2Sj^(8@m`
zcfW~0nVZUyeYncw`|e!28K9cLJk0_*t0F~s5JBHQeXu_JR@1MV;|q6i=z1L5cT4U^
z(H52HcQ!tlOcg-l$hdNJ+XgUtjrg)0S29pL!98`KTf4Ez=P54j_OafB2a-}+mi1sD
z!yek0=7QMg&!0^oy*P@xRXGat;m#-OQ}G?AP8FG{9%C8aqB(pvDXm3t>)gmy1@`v&
zAr~@omWj!MHr;Km&*Us{cC;()M10duzQSrI<t@Xm88_SBsc1g^#v&<$m1SZ=w?h*Z
zCDm(S{<)8}AlMa6^UpuN^VUL!W<|E}(m<C$NrU0DuYUP0s70gm8DRA;%xk<x-WU-K
zQSzQ|^5&Rl9<QZKax9Z+Q^VC1;_|)c%Dn?KJihW_#_jv5Foo$bSIkQ^LEdm8VyW=g
zK$pS@^&=0yV=uH%7Ec~jvgA$Yn(J8ZP$|{@+B&1?w!<-sc9m<ndvDR2@Q7xqYAm?*
z+!tPjYnt9Obq*l3sc36!dwdUzd7sbkv&+1YT9lG?nolfCzwM&D*PLq$WWXHIl2ST6
zS2?T%+<TxoR|wNSv>LlHXo^R$VBkqlPme$NOFfT5z!IC!Jo7eG@5vceLgpkNR38t4
zQnKT9Rq5u@V2@gNmfd_2bQhrPmudFlZVeR-)xCm+BQKxGyI5H4gq#91Xv7`^tp$65
z(6J%}<#7C9t}y%QBUlpyZACJepU8%yN_k?I7wh{&Rmn7(e`hDNT;gSm3VR7^>mA!o
zN=hDJ_68x3D@6iH1u}Lw^e$bWfgppE(rb`2*+Gc;eGMZcuV))5QP3{u0l8s4s)hcb
zik3ocNW*O&^XLiT)BKcR(14R%V(gkQmf?!L6V**ko6-Jt9ZU!j%u)+?g$z<+zLQV>
zEtnI4J*61-$O$oeTYdW%OGhs<dT;~4yCmw>_HP5&gfv^$8~3%N;@<|kd~M<X_~fh_
zSg1Py+akZhq9=c>qG)NJEASsd>Bg<JL<O^5K^)nW00Fhm@_<s44FRS8{#&+e@o#Hu
zs|f*B>dbqQ>6>!}^DW(5v6-gl0$G@7To1@i?8rYcqPZ}pX`0Src?G<prc#L9$|CS{
z0MRANATzM9t&XGre0snE1-T72n{!Ib?(J3XToo4nfb)xl;=(J1xmTCFfuGLQc^O(;
zTZ6XhI_7F@%zg6IDc-^yM4T9Fc}?5vae8`s&rB&TSuH{^2&V4zm}yZ_k!kSTkT30n
z9UUJ%U3d=5C0;LGu@Iu=_cqwY{5!<!lri!QA$l1ib$uYGnHYH(wn-yq$ZVe!>!{z$
zdv--E84F|Mdb0|R%5x#Lc}+%ca>WuO*^$m|X)6^HnYa>lIC)aG!<+z0VDVhQVgv)f
zEme{KeiO1In>A+LAgmuyTwE+^-=Ub5lcS6WIoDt`-=NwA61b2Tnu*Hc0c5^r*BOsR
zb@8j474mhPy+B_rq1;>p%ZgWi>;YqGe1svZr>AEoFfoy)=%3fOf0jB}vrrf25)kFx
zIVU-#vYdA=O4%dos9RvtZR}(6TD%_xxI+!0^@LPTxqbNXArrMIy<!iyHEY+FqFap_
ze6Qf~{-$yu<)$F?FdfOCErf@$07>-FyjbM^ea2?BJ4D;U9iMifL=ixeGYF6kmt*cQ
z6N-Q#sD}UMHXqbet;uUhwNLq>X!90Qb#};_M&X-oJKLC>n>(F9A1t-EII#~bFUe!^
zbiXB^D=Ro>yNTox#o+IiJv<m-f-3mv8=jEYQ0z<{)BbX*`HBs5+eO;~Vb7AmAzLBi
zec4|n6f^K}$A(7WrA;S?T8Sw!*WdWO{@#GAOTxXSNIOP9nue5`hR8`|FJ&fihe4Gy
zA{>o0PYbEU?GkNKIqegN*rso8-@s$@h3B$G(jz=3j6Vo2Zb$t91o7bh^a8<G*D1SS
zx$1f==i5ZtY!0r=zPUN}!8w>a-E@7HVfXe~<x}s`L{<uRU0i(pc4|e^ij32>&jVe?
zlkE?7rZqQO@-TjSGpM{9P>Ew9dhFuL?SR?O&15WG@Y{Gq1X2etScX}3n|CGr7Sksy
zEKV$`*X*DqoU%#?jM1E=G|gG?`-|fr#U!Z))X&c{>m_JKCr=*kzF2|BO8^C~_Sv>c
z<q)XMeU(I$+nzN4>!YkY)>gVUJzI69_P*MWZWS@{@U~3>W`FF9|9XDi7zl)%#7oT8
z3e6vdEh=XOO_B_id?FZd*ut7my*aW+R*XV5Fs)POOcBa;iH#6AwFxdSK*(z}U=$pU
zGT?WtT+?q<o(QIXb9Pb5%2oO8-th~k#w}lm?TW3)IBiDT@<Osq#;1>j4uE~TN@)Eh
zT>#cX-U*e+g30)c_k_9kZRZo>0za}{!vtcl|M{6EKf|M<v+r&{>v<TU9D?ChEFjXn
z{o~>1i0PAsPj|&?`AfPAcTJ7fHm@qWe`nfs3A1n;osO<8!x*dyZYM-a9(;AbjSa=`
zo4U>#KEKnj@0u7l)nHy!d`+Jk4g;I|7jO~GL5C;&To$s}A)W)KCnAp3?HT61>qcjO
zeEIfTAFSesdf_Am_*c;2?8&^o#hb@qb{6pDN+W}oNuKyNuOuqe@xCusiS@>4dhs3B
zW#hcLi!R0i3U*5F?(PGqLKwmX=smMP-FUVk>;3FCv;V#-PjXe7Lo<rQ?IjAt>!8s#
zh(r|*C_bWckPG&8AYUruok`{og+8LGzQ0(mFDvk<uHeL{h7=bnGUg7$gRLHQ@t$-U
zjUM*kB~A)$^i_X))OX9ZU&YeC2r7CHS{ns`m9vd-V5i`NSs-l7{#2TSTPV<F+SqDK
zeN4;Jo@Uzu0aC(1FzY}?z(w8_A~iSnf@^M{2j+ji&zJStWm@<yWOw!xJIT)sM|b6|
zuI6b&|4KQ!C^kZ<K!44eHLo*heHmrYZHjU}Vfv->n!$L?uZ)bWzhz$7G93FkYx$#4
z@}d*&V-2boxRCexw~p8qpIFcwNMY1aJ|G!)l6{zaHBPU(mMPY{ro(t%Av0B8l4UL3
z&aep|olr{jP3^%ar!;>a@AE`_CUCDz`B7DRCuPh9p7M~B`+0nYN!dMo;x<0f*f`^D
zqU&kDhn4<TY2e;Lx@3MjX`ZKM4GYW#eZFBe>tp%N3d1cJ55<46<r}^bdR5>-DoylD
z=RSrc;bz|hXY)`VlQJ!1!^!^k3lxcN9N;<a)|y}2(tUI3<^5t&a!5MQSv(4FJ+wFY
z=JE4P_>%w;kh+@n#J2)MKY#a&M#B@^#<a4wF1CbA<dkAZ_zGA=w|Y%H{l~w-mv=?G
zlo^^EialbEJ^duVplIEy`mE7YC(r==y6Mj5wU$U^0&mi^43{tMdm%*Y@OE)M$?iX)
z$AK0nkstA`>^`Eupw2dSmA<dkoz@Yv=~WHqeCGCi<pX1p0>${}Co53Vn>j!?ZgNRs
zs+~)Ps%hH#_LN6~6y<Ct^RK@tpE}iB`Ch4;&g$(h+8=gy(i@N?yF9+fAbgvY0-49J
zvi^|XsTd^Lz%{`ft(mupHIlOSEbG&BGlO4?9VegBv`lenNld%EN60CeVjrwwew=!O
zw)sj=o9l>@_s&k1q=U@jm*pF{=5{shXAoV^T4A5>EKW0^vod_3MKcaRCm^=d1s`i{
zpip?0SnTx0EQ99|{irZDSW6M!@8h>ePnUrb_=o@ciZl4j`c9ONDBFmz(qdg!r#>=T
zzgewZtBR<f>^3{Y+jR3o(#)({Ia6b}R!Ix_<2NUny!}PL#M1d(&kW;p+lCtMMpIth
z?}t$5-Cd@U{^ce`{aae)vNu99d7&!#_BVE=KBfC!g?|W=sXkND_dCWAt@sze_(-2F
z-5dBztm^T~EhmR0PhD&HROaflOX5ja0V%Oey}oY8z`yqJ=5?XB>cn}npO)#QS2i{<
z#L{czb~Se#W;nO}?zXQ^AKpB9>X*(**=UTD{i+!DC`v@^okQam#;?ubRO^E-0A4#2
zBl<E_xx69ENd<U7^duKcuWCj!N0%Kt%KXc5k9*20JnLgsk5+EEd)9odfn2QYfddCf
zOzet$!xSV|bS5=^M7345i4e6j5ekzSxCMk97M;WDI~_}NlWWCR(czi`ADlSfSlz%-
z0d<gc5{)AccnxtC6R6c?f@+ZV?&0z9ysl<3Y1S(g_0V%=)rXxLBP(Wn{9n9y5xLi{
zt!ET5j;X!<$fJG`MLs<P@7)yHtpgOyt!^KBylcbOb)12cUWH@lFI{XPXAxxI#+PVa
zrgOgWQ25tLKiNF^RT3~ep!uK-xz$Ec^jwHd<TcR!=K^fZpG8-ml&j^vjxhm}=Lf!!
zlB-VK*ag?XH=u^a(q-r51n0%c3Jb4){P?kH`W<AvJ+q*E+sT3z$^0|AJXrDu!9p3h
zA2)Q=FeLMPx7foAu2IXDm*@Nvv{OBfZxc_-kqdIN%3pk8BGwWrdkx}ufb#z6-Z)5t
zXuV96_u`-le(sqii8;tt#|wVz3$%0+Tjdf<$ExM%==ghEp%gf{oLC$OSk~7_DNHd{
zEEAPpeEOgVDg_}39#wGYd&o`S_nPTVdzY{-a&dL(Satpuapj(V<ySMg*XHvEBoHa*
zPf*0GHb9!=U!oMw4+>-+7Ri}>&Z=!58cO4%r?7@goUbn&8p@P8H9i>;%rnNil#)`#
zVPVyl?`)cTC1@u{#;FK(If|kwb8Ir^6T}0WNZLks_ij_;?em;T4%E%!h0|t@8|knk
zlimIvn5e&V^WfU~d8Svd;&bQck1u0zKIoJ1ZR0AYQj^zn(@g5>@wufBmtNM2*m1Jn
ze(u@9wTcEcSA6recCL7TWA!F+ic|+SdV}c^h8qkR%g`WN^U%+-j^B#Xhk~)<_-|QB
zr)eIJSw8tdA+-#y?0@~jH@*~f80GhO!<`RORM9rg{b}lan>17HA*W6XB{VSW2WRi0
zj|6lPX1iKO&VJ*}!}Dz_Ay(b$P_3l%+4kA3u7`Z?+W9syX;}M1dydEq-`+G>XvvOC
zSUj3svKWy|IwVmXojaocYK%SV+~<#UlrQ1;eK!)@q-fYHrN@NhfBrJI1oN1gz*aip
z<mB`kY+12RaijS@T~pxnnXaQ<&k&EM=>JMlWEo>mTj3a6&his2$t@m#94X^s%YQ0|
zA50u3P!U;+5;8J`&~dRr2(5PSqohPZB^h7Cw~QirC5WxO@?vncqy4I9404r!orIR@
zKScnh_tZ?Ce0J+r_~|e*p)4`6+xXy$YU-MG>&h_DTsZ`2BAc(TuW9-xu+j$*!64zy
zpXyOjAoSRz^LHqeuQiVL6?a~WtVG4Z7FfeUY!(tAfbYu%Jn|Z_AdC0TD0^KmHBnN+
zzbM(+4fm8IMBMW&a&DwxsgDP9Y`Nv%!ynex{gKuXE?~p-X?65!NR2*V^xO&VWo=J?
ze*i8#C@(!Vl`r3`#@Ydt9|ve~2YRLD=DHv;?Q~g9nKoNhF8O!K7`x1QeiHK_=`l^u
z!*pWhVc}wssWL$k(EG!9-?9dJik1!inB?CjP5k=LgomC{gY^ns9U#vlk4{5QCMHND
zg<!&KIkK28wEG3BaJp>+D>~W%l8ZVp1k6Ac0|73IbfbtNJmZDphb-pFl2qkb0o>J|
zZ;MXre-%vpH#-`x@aI`u74r*FZ{tZ&=PMc<s?6U?@tLWMa;DQ!adA=7)qPzP7f+?p
zY8{l69^5c6pisukIC_?4+O}RBk?rcKsdb}JDtHz5`X9JjTggBX7NN`Emz7O+yRiwq
zlkdl8A`j!w4&>dw{a**}|Mj6>V5EY=!{p?#9K9@`^%U-KLH4zewAM$`Q7REV{O<u$
zDOMV|es}lRS(fp+t*s|?m)P$IkV<*(j{L;Q`5?R8N$H_sAr2L#4XXfQ3cm!PV^xvw
z%=_|(f1HXM0vn*Wit18MI|pSIvsYGA>TfL~sSP_|H?)B@SXi>Sr^m-TamH=}_*;ML
zf819RN4`DqmsLxl7jH21AL76N(UScSQ1&kY=+oDo7BjKJxpTcXE-tXOl<;aS?2YBE
ztxOnErxJ2Ic$UOFOjq`wccw#Oj>967{_nBm?+1PScG)jt$+2E2h%f#8nJn_S8vgu;
zgQ=3*yN?7$1iV~~XmRPPEB+=TCbYTQ(Yfh()72~JxV@b^eRdt)d5hLR7=$FP--_aM
z*x~;jULLjYpO*DZvQ$4kCHD4TF6)_cEURRYbHB#rZgRtw3_+@Z=jtTMvfl89wT(g~
z``L{^<v1>gjpyXavisJSyP}s9Af1wUWT(=G0H*&+nK!N9&r=6(PGu}!*Ne;~O8k!j
z|NGBT{_1-2PA~wSuC6@~u4>Fo5vBp^cMkj~o@~3#f8rOTnc9dwm^|vk|M8lA6h@R1
zo9oeks{J}d2+GjUM2u1<s9GX_#$hPg1g0#ZaPti*hWyKT3r|P+FN`#=!6Z5z%1|R_
z0dFi2XtM58CD@!ZtID^vpe-*GlX4^{TO6UWd;dCGC}c>|8kWwtDc}hICZ4S%B_x4I
zO~^d#`*n0nKeH0omh;M;hV3g(Q%kZ$dk(v3WB4n_1X<2R(H~pXqv2=G4EC%4X}~!g
zRown8)|Kmv!t~cnLPx!gKqouqP8t{(AQKetEC(T%6*c|^*`*t`t><ii2#B*w+y#rD
zXWCKLyn^Z=EgHhqVDfJaxEl}=<#t@GJ8bkHpU=`$ZZNsPgV@>zV>V3r_pk3vQR_vh
zc>iy<n|vi;Z+|d`e`Esrw!ooiYbnKIMLU0j<-SId`zK0>-RG=baIa3s+o?L*GD+Te
z3g0GW#$^+}j!H(~DfhRsWW}77%KQ}+O44RfCkYoM|K;zi?`z@#Yoe`;SFmJ`!(AUk
zjFw$cRobWXeWijtDZ@`)_1?n4!Qp@&qcT*~i+1b2`6O94j3GFSh3cB_LH9&g2dlDI
zqM|ubce;7SPe)#N1I7R9W_~^4IRXT|T9|KNkn_L^)|HS@I(-_g-?D{>?E`n8c|ya`
z0Iw-DJ#!isx{hAv0%}kq?}75`8ZHGdk10}Iq-7h;X8-)&$!4^|j@Z$@*&WYInK6^)
z56S%cUREJ_>5!uC4VxD~`qBIlf@c-|ol4M<_SvHLfP{pE#Rx!Q`mJVJmcpZ)=-x>+
zY3gaWKmqe_)2agNWBK0XbOfcgw9Ae6;|s5&i(|L^L}o}SUq)f5vy^z1tE70Kz}L~a
zN3~0O0*SPP0~%DOnt)ME1+`BFJwrRoQ^A#zzNkV;&&<wJXWJLNAcP=+4yv3qwTh+G
zU$tNS2(!W$CAHYRK;Ycsztd(ljqtj&m3PNQkuQd_gFkpC3R@3*QrnT@sM?%vJrsCV
z`_;#DX7^)sd?^O|a+ccZllNi~|G?rbqB7_w3M5sO{wI-J73Mk_O{X>sjJXVc-iT4*
z#P^%)AF~nV)L3i!?v9(+(Y5Edrg1l4>TUuzI1Xz_#)FYq&EuPB%r(@Yc(-mv?YWqQ
z1$G`(Hj=yv2ndWq|3r#<UDY9>eYkz@L;GvebW7e!kZy$#slb))qNZ5i8~sDJ?6&BS
zxksBmURw{JKx8BTiN5pSp2AP{uDs_3Vt623&8}<Ud`xzxC(Z#3Gk^b{!NH*RnaAxF
z&-VK~HBa*m>t$FQ<^+VW6dg-ETe}dfg&?rwLt!GO)Lj0@oj^W@Na-?Z$va5?mm)f3
zMgRvAH~Yy{2Dh_mzQ541CcL?{bhQa~8i_p9)aa77m~t$Vrj+1JzLUZ+p~w&vu`!;a
z6PURU1&m)5W@<vZ(5)1p>Di}UQ(u3}S&FjpG|p!rX&j2_-j4~k1Ae?r4QjL{vjrt=
z1fW8ViK1Phx186SH75{*cPq?@_Uhl_o&U^v{;#pO{x^E>|8oKV=L!6O`~>E_FPU7~
VBa;}y0s1LL^`M4Q`hMd}{}-3hDRKY+

literal 0
HcmV?d00001

diff --git a/examples_skyline/accuracy_vs_fnr-skyline-order.png b/examples_skyline/accuracy_vs_fnr-skyline-order.png
new file mode 100644
index 0000000000000000000000000000000000000000..27de7a95802e9783e412cc0940848099eff9e369
GIT binary patch
literal 43584
zcmeFZ2UL{Vwl(?ziUAQ75k*3)AP5SgB*_L)KypSxE0QJW3@W0Mv`9*poRK6sm_TwW
zKqQN#BIgWmZMyrMbMHOpf9JjN|8KlI#_ciMvdXG2>~HV2=A3J;{X$OWHu*uCgBXU9
zOWe7ofMLYV7`8WNKPh}--)hqaFMF(SNGR=x|6KMzd=BpqSlm&!#xSyD=zpTtUbjv7
zkk3X;%|_8&-^TvFl^$k#-^SvRxy>V^2fx|rSy>yIn_XqU!p?c|H$xj63w{oczyCYC
zxs?G&+&upe4Eqg}xOH90A!4%EQ8U=KZ)c7-oaOOv#_Z`u?{FSui65T7?cg&WXJLP+
zuGrI<cURiE$6l8#Ay7|oIXss0^wNHrqO7;)6ZEeP85T%tgm}d0rWE}~-&I~AQYhqj
z98c!?dEQBwUifh5isops@Tlhc;FN9V*M^h2d*Jd1UKW-V!z&iUenyl8UQZl}A%YjO
z{V*Ju$KE}L@KQ-k%mpvM6A{tC%X<%u9A4B!J<wQP{O$i2KeO=s>O;-O$at%~y!=PG
zt8;z^vyhP1$B!RFba?e@kN4VkYT@D|B52pv*Hh!;<5TO>jtveDY8047rqyB19&po@
zd!rnF9%zWTrnt7YMreuiPOB@5Z*G<w9v!Xw{P}Zx0w$UPw=6e$QddbsgFv{FDQRc-
z;q19{Y)e0W{`|?BnP`?wSm~7$xs9m?VB|5ai87I?hw1pgH8sih7TdP{{OUv3RJOCd
zLHLoMHdN%kv(b->C)sz{5T7h>+@2b&k!SeMo8IXTZs80GE+{ydiI<o6_O};4K8gwo
zIJ3e*(zVsqF>eO9=E<&Hx(UPn^0GeX8KI>05O%H9H*ZdIn{?cz<F{!J<2G4rxy_KG
znx!qXP&QX}x_HQ6fYl}XUV)i%IFD(>;rnW(4%xA>v5XR@PMsq3mK4A*5<4>PO$4}a
z>U!1jJ1?gXSv(&bGqfFRzOK-9e}7w&tg4by(B{?_x80;#*OZ+^1?@9V(;r;<Ms4CN
za1mm4TKUGo#kLdYMMOmQ?%iv&zBoRdxsy@Qr%+(ln`b#Bo^RIsV_fKxM&6y4mKN%=
z#8T5ak(~vR;=yO9e~peNow@mJ?Az;e{%aNM1h%D_ejOqrqL=>uDo##LUeVr+m+pPI
z`}+0kgZuZ(X=+B4uZ^6pwuXln#$y@=_kqtdQlz4$Hks}zI!H#Q=yT#yP(;LJUV3Y*
zLUn)LiAx$R!ou3s{&dz(47vIZ3WkQsW<|pxQXP9Ssw<eB#LjB9$X#vis0$Y^a2gpI
zRaIA4(+WB?Zw@i+T%P)oe-#EPF)t7AveIjRo{g>UI5l;6ax%-pSX@BZ=BA7FNd2i?
zxItZCPWSb;=({&=Jl5^nSefC)-IbSr%~|z^h=labt5-*IO*+{Xm6RUB6*j!2VyiEA
z-MDLHlp-N3%PSqm#o7JA^lH_~T-yniwsiHYxpq?;@X#MvSz9;o_Pgj#+Lajfl@`_1
z*W0ACoj+gYvN=~B>XJ3pUBGM7nQ41BfL`cF={g~;>ZI=a@|4L){ad)45qyfGeqG3=
z(5*?vw=?A%jhu8RPUt_-)s=>uz2>x-Jl>k%vo@i#W9-YR|M?kt70hwLCA?ylSwZ;Y
z$B%tNIuoT&FMOreIjQS=&2I9|koEWX_n#a%c<@Tq$%Rj39B=4_oZc|FZ?>G&z1x|o
z)sU*3Mq5|7&?wNbINs*R+1MRdYT5z|G1q=ZTTxE#chW;Ae~)roH-bwmDk?gE`Eql*
zW>ITUP|%xvlg_ZpO3`Y5jy(A8(9n=>N2Znu8mT;^w!7Yk>FT?34d|k!Y;5vsYHB!P
zg6XeJ_a+*(C5|jKiSXK$^a?J2*G1RwvOb;!gV5))AZ@+1)EN$|>tso<*)dAWhC<7s
zh`#DCFDRwm+}w(y&taMemPJXD^Eh>@lC?_gtnBSOzh~>VrzkS(J$yK3?!FNo<wUsf
zwW_MNp+TBXz)nd`?ah@dS8Qenu9q$UFfsm-mqNp9mVVm2bnay*uTd+bn7DXtTbrV`
zw)Xl`mU{+T`0z^-(nHO!&fG+6K%zBWJ=dn+b;W3Q;3J$*RZi|zvCa5db8~Y)Isui1
zk%mCVslKv%>gwt=JdeIJbqcv{IwT|}O5$W9`4g1WRE#!P=WvW1931fAWpDe_zRt?Z
zg6DnjZGd8c;QaF+`6fv)jN}Z$ngVu{7J{SG)0x!T4%hqo`ixuSYs9ja=I0gk^%G^H
z1XD&D!qW_zqWnjFIoDyLC&ME!hB<)35aW21^R%VQVOTM3%+W5f2Pn^L%4fhb&*HbA
zPCqXl2Fs~WexlTIfy-u0j+~BPX?l8^M$qBC)5>%PTxD=c#9p!!W1SV-`J@!|7vXY+
z=Dtw!TaWmbZ_Z04H8DYqhAH1t&ETGX|MRo)k#O^nf`S5Vr*WpDlAZaq$zJ<@Wh<+k
z$4{ONPfcZ9P)r;f3J|g5*VL^Ix&Xs&^vA;^=JJpXziQzB;K76AM~=jJcwn$trC?NA
zJW2M^dY_M&sq`d)ITV<bWRTE7eeBpzt}>@3<E5!?RZmY(2qn#E!g?Q5QhK6vf_Ahi
znnTxjq%ksWb=ASq*;&V`m-77G=2$Two8?J$6t{$&m(di6&w0N+RnVU*70Uk7&#xVp
zGV9VdT-4BNS9SGmNF4s-Dj6CnmP1vzcvWpjF~~;qOH1v)Y66)g$f()zaMI%^P9#E9
zFod<~-$^gxrU{Y9X}QZ_cxGniWW_{gNuJA^Riv=128v8JZPL6crrjS@l}#Yo2(OQw
ztL}dRH*5MMPXbm(Yn9(=I@@T0?TvwtPyCzYSvLZ1+`037etw=-chr34$vz5xX<pMG
zN^l1jcMS{-On!ZP?Qd&8^+TDSo}S-j)uO15#G^L3$D9)**M<|&I}8wI=u~KLtuF<7
zdwav+KJ=A`1C~SD^3kR|b}SZBC<I%b<FSxSQXfAiu3mKCUWoS5o*5Y#SzUwFwBBIu
z7I)2hWXxY+`c-IX=yd6V%qd;k3Dzu92|{;477e#?yx(bF<L%A0FcJ4{y?%4Z5ok=>
z(^O61Lm6jd0b4>m9<LM}5@OsED@G&al;=+`l$t0Vu5DMOO4wZ4?y+tfFB*-qb`s{{
zQGrLH46#<fF(L&X9vz>BC=OrhXx$LbD?2wgrvb?hB~J+jSZ?b}6G>t|CyXFftoP2i
z*|*z1d-~My=a&~wJ6kJtuG^bC$L}bV*iGeGkIGc%af<IdOe+UJ)1BO0>HA=MFG)9^
z?BGG0#WDG$CHL8n`}|iN%G1-+eM0tO!5`@jv0$t9MI#7q*ADM{Bjo(SZt4e<L@|sU
zx9hqMJVWWKlW663b#)Qik^+8wFikUvauiRJjSlHG>x31_B*Dqax&G6KBPcL1j*?ks
ztX);7?AJOO8JR6i!hGX)$*8C(jWVZv2#*sE+XWR{8Tw688A}tLX?b}{j*gB$at#88
z)=Fl|?@<UXzA7y(&0M{VgY`Sc;lAcmzWS5Q@6^>QPGR#t!L7wMSrmdqMLig=(W%Jw
z!b;K0?Xvm#<p3NfMLk!aIzxn=U4|!II!@eAL0tTq?&fsKOh@$2romKCks^FpOweeY
zYTy19D66EPaJHm3b8*pTcw{6*z}>~AR3%e01zlHi@|~=T?c&n1vi1u19jj3Wc$O;Z
zYVV}Oc~aiLmunK)DQ%-tKQ9}V9v{!BXJ8<Ob6X#0zNA&WnA%T`MIacZ0JtI%f`ul`
zNyUkdjC}NLv@0NPvs}0!0ju;?baXVDWDun9!mQRPcPUz)>P`aWws+q?laG)0;^IF(
z*=I8OJ^SU$mkL?hW$ey(bKp09P7{~DR#&GvE{woLEEo4ZIp|O>8!a3LkB;7Xl5=fi
z!=%!K$X||00v-lFQ>!E*bJOzLP<24(XtaCgJ+mS)046rRS?amsfKJddXV{+ImynRq
zzB&h2*{JN`P>8UVeZQ+JPxw@<BgDgW3gK1J&)yXFnpApdEq<nVadF8k_6LXncde9F
zI{#85-#7u@qAWwDZFl+X*&C1=+CEc=q=#I(XS_B)jP4Ltk<#=goYrYH(w^)7r#<<m
z-6}BU?Th8>(N*}J<h*h7*`e2PpO7NbGVT><FKrNeEEK)9?zblvC9wkBqGjE{y}nYh
zlL7gqsF*gC(ERQRf|R{Z6B(K8dojWu$dHDww%0o=Sb2CpeDogmBBO?M=x1`4#A9#7
zt^F|{4^U8qa*9iZT(qPK@b$IfuOEtlOd)&Vz=4pC`?+0<Iy*}`)PCMxUPh3yI}T=^
zx3sjZt*^)T6j_@rzA?p;C-|Ns((&Fy>oC0h_n{mAEjZEtT;zXGF8jxGBcw|No*%E+
z*{H||^eA@gR$(zMW|5+^MNnPe=--(pTw!HmX1)ywgZ07%OWO;TRA)rv8cW6#!x4;D
z4V9k)6d40(sJ6df2k260q&tLA0-$~fIb0S4q7ZLY-Ctvx`b&iN;?XGQOyg6EiFaGH
z?0c2PsFW`X3Ti^QwZ7~9>=~<whz^9-_Hwfr)*~_US%3gail0he$eM<vOeo)&ZZpX$
zv+L?)xMp<=QbVg0M_IgXwSS8JOdpMa-6PwegE7zW?QW#_+j|W0r@9NwRhy!OfLAAF
z>sHsA6-eP)+uEpTX(!)sz)9N@rGY600lQ26`SXG6QoAalzoLSenD{syT?QnX>P-kD
z>oYss=7=f6?b8Um6e57|^ywZ%g0r);Yinz*>UV}ibs%o3Y;SKPs1~Z@c0oep{qLTD
zQ>T;`d{15FdhlhxL@RKMr7x7SEL>bl5G6YRso6qoOj^18j|6w70wo4M%h(?U=I&ih
z2$sC`^~Jy8)WsKC>Po-8BBtZMmA^dMWv$1+<DoJ5$?XHs6RG8#imfSg|8W%bfn+E{
zAh+!{Ex6`NmhZe`MourN23XX%w>V$hahNIt@(7CmCoZb}rlX@HcKfzY&Idqn#$TQv
zJ$UGl5&-h0?Ts0MsPHR#FU14sGtl6HwgAW!->Kt%NdjG;m6a72z&cPl2u<$Wxj8vG
zzYs8!{WKgx$H0(<umSKNbnO+}OIc~_5kz%|<n~1fId?pL_AHFI&*3vbv9JK)^=}SN
zPGOLgbMJp9r96K8*Bjo87wd1`z8ww#5Yai{rqaLv9;04lm4KKt@KA2Jmk~(RTvaEp
z{$7c-y}P+D*SufYdUMsXE#E}74XD+^cv3V__On+lf8M$ZoC&z#?ErdiVEcgHt|cZX
zf2)rc>EE+w&zp0%{ojD@a2f_;0q{1z@i+(-64KJ#Kt2)m9{lVb9(Ow+ERb0?3|Pcl
zKvF>b>YtIH_66MPNJvTY?KR^Csz~K_BHC>|<p3F3$*<9+Z|4Hu)C4jM_yJoB3lP~}
ztrfAs*zMg!N^siCFu136ebsXwz8Hs02yB~6OiXMfAyoSctQ(-`*7Lt4U~-<m3UChb
ze)N6;Fxn|y-?g^r?T5#X9m`Dq5N=jnvpH>Fan+zPf)l1XFMtHDiOwwhi`8vyZ7!hA
zB))&&TngF!yr5tYj5dE?XHv8qL9JtZYr~HbroN3>-SYA>51VQx*VnIKN8sK#;rVX0
zyWu^UCOyR?XW^VFfUMvtB?HLga#}P5o~pvPb1pzI7LWtCQR^)P8JgeS<OYgXR8nFB
zj1wt6Kuw0<>_CoJ2JY}{W+n>-6O{JN%9d0dot%v8-*AF>R>EjBTpLWlrv#ptiC~eH
zmGxr~(bg)qX_+6c8y*^p6PPZ%nUa!%iw9Xj#(8<N<@wQbz!M}TU+UMtc?TR6co4$p
ziW(YWaisS4_GPo5$mW-qRRYgTXA%}*(uP)4RJ6T;KLC2zD%b6mTWDO$WgFKUE!~(d
zL1d^MW(-^q0glq<PbUx*8JR)FrjiV}(Fw$lc35=i9P1NV1xjrY;_@69^x@y6fmo>A
zz3V$&vE$b8_G&e!EWkZ9J^*URLEef3C8ZpM5?1wPKi<CNcQ?IjKYzXn=+0)cm(=(m
z5smF9fQnSdk28YyR;>pAc`bN9eah$7*x1-{o53|6m{)Z)oCqe!O@BvF52sd(AWBn9
zkB*KC5DG2I?%%&(TUQszDE;R0$D=VoE3*V<N)?5Lg+ZB!R+5v8(JphkW?Z<n+`ZH$
zD{?KQEmfI~+pI?ow^+-T^{um072;W#pyPb7yWCDSyk(Y+vY`=hoR8n!+`Py^=3x>!
zVW8St*H{A=gm7N93#_^;RV*AFvWTp}W6exWM6}P2G{V}AL$Jau<<Y+D78dUhL$G10
zGiyuq8!}*EU;q`*|LASMv!Zb_hTR`BfG|Gvwe%hS^-GCwXYINGaCC%u=Z323+-0JL
zGtly~q63CqGG=_@+O=y34<5uZqWC4UwHRnO`TZTxHMr&LIDC$t6q<h}_A_>P3!oFX
zFIT)Df#f0s&kG)WCqy8VqQ-CE?j#9a0qWDS((j%H@9OG^2Q)^Vo&;;30Fnt9vdO3q
zR#3A0&i26r2U;jZHZ5vwwPb*s*-U=tOft`JobgiViuz?ODJ2yg9-anVOUq73HRn-f
z`~K!+N+5d7FmKD2I_|NuvbF+`4|@IDQY@?NPkQY-U&9Pp^L4M8s*_VO#Ij4BoUl|}
zH>L^@7kKI8lmGiiVm*C*V-US|H5Z^x;<CbP@h}+^VU9?#c5lpAl2D9gSJSsaAYqY!
z$By(8tFII6mo5dvfq|3ya^`eTv0l7*(Z28Xx!cAdlW7!Myn{)o?J@_kMJxrlYe@f9
zj~^c;Il=`UbF^KTl#v94LqyCKLIo}!NhC1YoE92*5etOnA_+=vu+p-rriw}k+&zc7
z6cF_Fm6<GqXjc`x5-k7>J88h+u4xgQgRTg&@@u%CCD2)FySwkXZY<lTC&eCdXR0IB
z^uzr4-^s^`qwffB&feThN@@&*3CIPWs*DFF0bE}@TUKc-1da>vHVa6Jp8NG6M4~ma
zA9!iIZ`tM>w%kAxK>mZT2lF3wwQ@E+JwTaf*2|AZ3<#_L9Av^2Km&;UBRX`?ysW)-
zDXjwdD4HoZy$>tP$oPkG8O4GO%ohbbBB((>P|1$(+b1uw;Gm#*pvESEL}aS(kt92(
zFXq5d5@0oTfTV{9wG{A*En>xtHINKZtPkPP$>cHpk*t^`^8+MHMk9!7xLPMIz#JUv
zvc|@#kQIT>3HZ@oQ{X6_eT;%%^@4qmr8lJB7<4w!2hfl}aAlNe65dGHc3aciyJwFf
z=-MHn<b`43*LDOfF-(rR*JYsrF??H4hf9i&BOtJ`Jk<u`n5Nc2xl7+k$T3km5c*;m
z+*Yr5YCC77NfFWPA-1{Jk**&1{P}*{$?x)TnPp>Zlar6gX}Ipl%ge*VO2cR8VFSjC
z#p{ERP?bPx;ew3X>oDY}Ts&#5N>V)A1j0|6jgY9;=QFg_K!oK-8zYxsP+0l+)BpfU
zN=qw)tU`-mjyMQ=Z9*$Oad0Pha}An0fP87Gabs#Xvq1aNrC8hxx+Xqb_jbN{zqTI@
zcPor*p3S&IlH--DS8eMq72Ls*9;S6@S-|AFTPCoW`Tfn*-|FhDi&vc?uC#saX1c)7
zuK}_<j`5OqsalduWNS;@ZIe71tdM4rA4?W&s%v^oCLJxTg;WVN&CsZ!2;%KsEw+#l
zAJ4+XB!<faMCA4CS%*zZ;Zu<Ai%J&Uzx{}3e-sF5o&d#xO!heO<_SVT@vM2;yOgQ!
z_%lnzwZDn#{KmraZG2h+WEbk*EhOdomtBEagBBWKMs^LA2M=OF0A`Xndh}>W=9&*7
zG%6~yu#itR)N=vvW!hR|Z>}0D$24Y3E~T|v(B6z+HdWAZ(y~n{+gY-F{r$U|D?HmD
z748)vk$@tVc=Vh&UxW3pZ%3=WCUjuEf?VmE$wGy(`e}m<blaocBPl5fiBa8VCDI7=
zfgA!wqNUA+CX=olJzVW}=gFf--!(=GOw0~`mXwsVvTVnGCoemUu?EzAZ$DDrk?v}|
z(P=;~ul4<Wne$2q05uLk^Cm45#z}wO{8MdvBU%!2H8~}v_?{k3h~}!H3LKzVOoW@4
zTiTw&#l!qygSh~5G*J0u%f(`X>b-TkvP`S}%^rJUu;v#Q6hWWpfnRIMlvxw-7huqe
ziVvOAMfw&9IaVeQx$dB!okl;K=Od+gUujm51}lH_M#?Ds4j(H<wp{-1V{9k)p~d3b
zbD|_C=ExH6|1J>sZm3`MbERD-Eiso|bzxD__*^xE{arxNTEH)|OCur$>{-vB7YDTq
za4kN+Q#%dnS4eeGR-57gD!~2d`(=q%D0L-6J!~vBbPbIil+dVzD<0MygQ%=Mw}}~3
zIQ4Rj+2&JJi;~#-f`SZ{o_G`j0dJ)*S{2*RWP&1e2M3BZ3Zh1fV=c<c`hE^`U#U6P
zHgQZ+TyPfKiH@n4Bm3l>n<w0^XQjEqf>k;fE`5+9w5PX@-t<G7P9i)_m_FRPRlYt^
zb+xsvmkN4fp#X#yD<nc(t(%eH{I_$6q5wC@f>coWX}J39*K6!rHZV!NyuBe+1YR-5
zv-G-cOd(8#>I&5+py8#tZLOz(3Qu!Y-xnSSlY~h}x_l_R7CzTNwwi8=T@MO5#nZ(T
zHc-4AheD8zZ^dG()XV43n-O?{+-w5z$bZFlx=e7P{wkp(BOIz>i@;^Z>NqMO5!Wq(
zIEMzh-+ia7ZF6UHR0LEwcOVvN@GSUk33$B;5;<X~$b9XUs?+9UfTof<+wV_$lBT|B
zIL~xma_?TfwyrL{HeApv3)?z3P&}!qsD|<1nZ^gb89Go%ah@)^&EPWYN$>dU=q2@>
zTR5As<}*l}{u33l=I_%7-hwbCXn<5O@8PMcA^9IJF&#!f1(E|Wy|6xqA#x$8J~<6|
z?py3Kh(-wmsN*D4Xqjm@57Kg>4tJ#rU1O&Q5s8wfW;%#bgo&&Q709zr%abVxSd}f0
zwZtLC7ns;NfN*r6t^>KC;e2qb$T=JqFe+ePQhP57A(9_J6(pSpYYUiYR@&XHaFG3k
zHzuzFb!wd;uz>W{&Uj%wREK5s`0*e=oecekuyG`TLi!r!D%Yz!v5R-OfXTr9*39)p
z+V@=ms0;Tb?M=auiFevZPQ&Q#?vBtM%w0uL*8zI6!R#eGJ8dQhD0O>#rGiaB;2s()
zSOjeXv|4|80nz{$*FgCQQd=hsfl=Fe^3yzb0eA*O%}3miCIKnvfbmgn7co#+yGE<C
zgVbfqn$n1JLQ*~p<zh`ME$5YKWk3N`r%$t`OfD=gK6EFRJGIC^QDz46<6<$tKSr(;
zxU(TX_p&;}2@nS`=!;N%SbsUPvb=1!>5_8#6;xqUL14zQ7w5IC)y>Y%%2`@w`*Nb1
z4}yNM%2=*mRfO7V&b&%y7_1k_CkS+BsAb<M=y!AW%VaS8HfGE1t*z>OB*xSD7>%)M
zPb$y^P+tYOXMy&xUAiQVdkFe7R6EqQ%^)s)Fzr4s5i0058v&J4bf>Vmq8Z1bw79<9
zZQh<BMXjf&H@-1b5y}W9GHt7CyuCJoNOgvi1gf3*P;+!h^Y+G{HY>bd7s?R|c+AbM
z9HeK<7L$dA1zW&ZsC0EzzrJ~wJHz^8TXCp>lngcWJ7}pe1@|_`6c(=L%urQ+9kcCl
z?uD2!W@}4JM3a2Jrl!@aKXYw0$WJvGGPEYtoh<KG0IEmgF|0x^`)SRsVGj4~(t747
zha2u&6FTj2w<(hJRO6vUwm$orf*Xo^2m$I;xFOjd)e?h(j&Fh1sscF0wt>5ouWTAb
zUnR(~LrvVBnsFd+ERDC_hRZEO{AOva<t9vHnA=a$vQN|5lI$(9-z8Q#s9*>&?*c2U
zBvAS|w74WtT3`HOl9d9C+h4~F2d;y2Xej`+ejghb$L%otyBQ;<bgZn=DJz`I2<7}c
zS!s2LjBf$7(9$Aie-BDPT^7*dJg%Km$Ev4Zr-nvUmif%1hLw3(fTvuv--MDguy7y@
zrcO%)J$QyH=azyB%AA)7%I79B(;!+0w!7WVN^=`t5^&P${rLKqD7WOD!u8zdGfd~t
zv%)L_{JPR(9RLe@oPfh3V9IdAnC~N^5zXLr-*!<A#jtjNZB=`VL;p)ditRUmAXJo;
zagbP$OX2$U>;B`PK7C@<QU%nUSu)Kpp#Z=i<N~|)I#k-{2DW}My1&pe4yu%OY*7rN
zTT1=TTJW6@X1z{d7vh!&S!qS-fJu-|eb2s$;{s0!5?Nj_M_E`ZtD4RKstI**8>;dH
zX~+w}xVzW_Npdv7IZH}b7K&h@fH6DR!Wi-?oM8bx>^4$d4Ii{}|EM)+{;cjpo67LS
z8yNC7Q$tr*SEgX@F~BE&^g<fIx#hu8;LC}Nhx(BbR3ixOX`yB%Q#>$Bzj4xoAO=2-
zGAO!-YS{y@MM)1Bj*P6xR-Ud4U`#?!Q7$BjxP9~vq6mG(=jc2IStXPaF{3{-2A5e6
zA6iULD1@5<E*^(bgXqHFPyyl%O3KKH0aG$^`orGJua9^x2x?l{b2#U<Ot804^trBR
zq0>Vk@fB(o<<6eUyI|=Ud+_B!Mu{%woZv!1d0bF)nf^52d)Ey+L~VdWwvqGp#6U=&
zK*1c<H4>ykp)RU|#5o)exA<`%!x(53?e%Z3+5+vFs9}yS{gp#OFPIEEi3S8SMTiI7
z55FBj8OL>fu?5ME$W8-`xDBP}v9U29Z4q07(o}zi4wS$7-8KsW3!>V43g8!iA!($?
z)JMO16@w~ppfv?Sd;s%DyPF!uTCE>=j56VIj&{=IWR6iv)pVvpr&BNJg2ThZ09F|#
z7K@9Yqo^6bB;h~nB6D5d6W77quI4cfRXwO$@0RgEfmDFv8qm@N#1?UY^YBnXL<y-n
zfT0tz`NdtmP9;lQYcOYLt;wAg%551M`8cJHMMRAxS|QCHKY5Zx0>OiF2rWL^3{+GJ
z5CwW2M<a`2&^gwG(LygR6+}!1hw=a$1BnMAC}>g2K{zQU=27W0Vy?2o+<xe4BPg$h
zRI$~829@m{3x<~QJ*UBWP~8N&E}S}q6V<SO<mmadS^0z7lVGdWSR2f?1pW_0C^<>W
z$aDZ=HL=;kQt!y@70QU6lZ@?r$1`}+FedEjy0->RR{HF$1uAml@*Z_v0i2*^wGGh}
z6&n3@SRvB=D041AA}$o`A&*Jm9)(_emJK*$9l)I}ASGaWEZp46V3?45^eA0;XVZEM
z{sk_r185+JlPS1m4%6}>JsX(-_$_}vg~yFFWZ+L)Toq90g0asBe7Gy=Dq9TFb6}ZA
zHLlC18J6P4Y?Gz;!~B$>54y;}!H~oWe=t!<X0|Q~&Q_A3OQ3?a1jsQN@JMJx1#sf^
z)uDhTxOz4qH-J?ArK%gEgzkZo8v|8wTIRusT(>PJs4xm_Kh)DR0$@kOXAuK|6w25f
zy_VH<KyX^En}m`emyV%?4@nIiRFNUd3JS3$)5TH{tX09dvj~zT%WuDlLZU1=!J$3L
zmp$SFwZnFR83_P0k;IK=N9sTuiA6gukcnYbo0%axXO2}V+Nl=^1ftzeT3Y&Y=fIvZ
zNjPTV2}nWBO&W|I32UQUGlG;n=jsckEwA3c`l-HjC(^(%(zYc<G0B&+>NpkETcn-B
zlF(|I09%N9q@KD;-|{rX7&6urfj`2vL4N03#|7U5WLK*iGA1qQ%ciZ26)WO8)vGyS
zKqbeb<aC)2b(}|)1E|zKEa--n`Km68k0i!l>P~VtC8-Ielh$W3FwPA(YBe3l2v_0l
zSI)oOg_UQmKOSICj{e?gIJU~$eEEGOn>R+T)(n%GPRJ?fnPSr)jKu7L2dR&C=jG-J
z)&SiYA~JZ`NcRE0U(=p|m$EA<>)@IP(jzeah?I-#Khhl_LB*iAYH5j=1q1m5<5FSq
zp_q?xoysEXTD?_gX#c15`2UHElN=BF&v76DAfg$=ilG*Y2B!n8=v2TYP@HdtETAdJ
zgyDt3b!zAVF>`U?3579~r=Vu61o;gFiO3Phwc$`IsAYhc0^!Drovj%yrAru|efLNI
zO3MJkY6E(`MC%JmW+eWo1|7um{gcwjFboemkDwRGyopaBf`A7v$uaTnTck#fLFu7m
zye$dri^XB=l$bFS_+Vo+LHh>gfTMTcv^4F?5&P2}@N0T{b65{Va0LYgMhTIf&0L_e
zT5^}znM6dQrxfbmOA~nD=Uk!yM))lTJaqYac~#->G7I%#+)}{(UNeGaLKYAZaxnqo
z8{*RW2kipMLVZ8)t2;UIcPE70567E=0DzIBU<m~{0hH>zyu3gVYy&kuV&nihc{`|}
zOcMPr3x1$Df~#Qor4><(3b=elNq&uwr@XytK={yYrU<Ge8cq;lg?2U<G(J3f%Q(K&
zsnZGIIt$PR(*8lcdWVz(&1+a8y|)D>N`eXt0HE<klaL{7HhVJe@tT4~F|m%=LnA=w
zuf+}P>ff^t*JbNB&_IbZ8E9V^FjKLn(O<s|2P!?04;SEVo^d-XwbB7No+?N_G4b)Y
zP#}innNeSV7YG1A$2OxVDDkO-<p;6N*4EZVD3bPxwG)H2Iq_%=$E(+`(Y#9q0n`|u
z{2D?<65tO0hM|#>M5rQx%PxZN9pvkxSS2`3AyzP7E(%V`8#ivywt<cKN3m^Mn;?pE
zVCP!|4l)5;1PTCJ(xK?GsLzP1-hdg3)CNQd?tOUF0%k+-)v*_q2+TCgN=Tsk2;(?d
zN5IxvOqMDp2KlS&eEy;CJ(wKzkr<96$B!dBFkhmzwKbF=esMK{Uoa2GMM9y%Z3_Y`
z%96;lA+vzMGAE}z4y~oa!a~9&`jGgfq+Ds^Qy&h1k$d3e7}=MqDgtQjZNRRS)zp&n
z^0>ef^2sO)Y%$2-g<Av&@(uJHz+IUv)NlvfzL+Z!Xy(WOUC?V6UTim&3Vtdmlo8fO
zqdP%PV3a^!o<8>-SC9})z>J7^3@Ak-KWo7;E=S4tG>#V3+;)(wBdd_YUEKu4w+#uZ
z#TVfWcw&$i;PwAZtZA0}{z$>0@Yl5VvSDsf@mUNbFv_Y@V=<)g`GZ)fz3@avYVll>
zIB*~nR=o%Ra{#P2U;Pa$K@w2A;Xz9CID*z>2<FU<J%%x7_LJxHfH8&jmIo=}&1Y}Z
zA302#<Kxrl#hXIsV~s~m4K5&H;Z1;S7azD-o?_(qGu{;4dG_Y9jZqCD4{eqoeOR8B
zE_Hb>bLW+ULj3#7HIPI(nY<|X6Jg1FKp*o4JrpM?6)8)4@zSMc3K2~>=z#+YDk0!8
zU%ul&ip3}ZaG|qWezeBCrOX8vlS&T^^k!*Uoyd4rEK;GD&=*?EU2}SR;V?G#EAku5
z5r_6;qSCO|ZSPEU9C-Qs&g%mEH;lZF&fXW!o)ycur`kJL{~?mctKt;<3Uv?FwtnFw
zfudnv3>zI%=@X@EE4IB6OgggAW%nZpF#KLIJ!V$@=x-`N=X374G24vB-^4IRkVyDs
zTh|$HBe0n)DlzfL0l;&?_bY;B$C#eVj|6KzaunV_T@Q8V4tQBbgc&}E#sXU0KBl_r
zx;0h?VQpba`Y+1D&_@Ajzw*pG=f?`ql340R?wVOCzHxs8Z!Cz<z-0mb!Xg-LWPiI%
zf>|8+7ZXYJ((kzvgM+vd=dQR6{+TGPI5rY2W_ytF#x8T=lxJP7IF0EWgi1>(+)$3L
zQ{DY{w+9ql0}mu5Dt>iYA3RuFHL+ZaRqkC4A?NlGYWDCMR<a14d|5ToSc}Cxo4a({
zc)xQD5j9!H0DBjeUuR@6ktq4lJs$iKVmJtz4hy}x!$XQZ+6Ahlu&B(t<ZgoG>chl%
z=fVXAwKu<jGrz$pNcZz}jQ#C^AYtZ<-nc=o|D8tb6o!xa^>9=-hN)cdX8!LKz;lxs
z(8WAOVKruZ0M3QM(8EH4Rt-<5Zy(wcXL&ylZ$B>i)PE?{l$rM!HsFLNp`KJ>Yg=zV
zyc}jo<b7xM^T^?tsJ}6^TAA1y;nx{ohV^)l-5G1_48t^sX77i?Y*jKcQa9&!=DFv*
z60!uH0prri_dpoMUzL~$q#ay~cpOX2lrbt!jA@Q~GvLrCt*yjcvm|8)Y$;%B8d%8$
zxqtS6haGc1Lg=0wMXDifas$j!AxeVIHaQ6^%|4H?fhnzj_=iA27Ik6m#)66Exq12d
z-#~|`wEa;a8&XnYZEm$YpOx|9f2W?7kI$y=p6OUIGmD<X%TJr1f1T&`TJyf3=Wvku
z2(d*<M+W8bqp#1J&s>%;n_oVxc<YvAdxWTAZg8*XWr2HF5+^x1Umx)^{;bUB;l%RJ
znb`A`Xa<_|OPZGFF4U|o_bPPRC>-NezW(O}=(b|zJDMG9Ilv0D858Y2q4Dv>bC}+$
zFHkEC;lr<oHPvf-#E>9F(!SdsL-_D?&jlv2gJgj{J+<_T3VhqA+Mfl&su%4IeI@$4
zxZi#BWdC(bP4M4lvaV^_y-6=yvvpUx3$&PrhHsh$cA^Vn+D{I9zlKKk@DSt90|V3@
zT_W9y^zPC@N_AxWu$T~C3a;Pf6<RsoEk-<koHO_1*C&1*d0bfP;p2xf(Mz{INH5mb
zWb;h3Qg=|M@+&~}$_HiZJqs89M`4<#nyf$J&LD(rt38I5vFPrq9qX%;9`#TiX82%K
z7h;i+RAK@zbi$f;4+gw!g_9n34G(WV;K<53HBqbn8U=eW2&&mkyOhMo=fpAGxLJ24
znXU1&fXny9GG|I<4Gp_T3thQgKT3G#b_jPZQgiiUe%XGKW)(IkGBax(X=v@=(cWl!
zxd-k9)_@#13jd6WW2q5qpDA{kw~>)20F%itl?xLtb-WVvU%eX1SS|)BC=jaP5-hwP
z378&qzp6RSQOLI0V9{kapFxv)0%qvv#;8j>dn&D}+4$Ap$H#A4%CM)e{FjPcHNEp)
zAbu}jyg<b@sv}1R9<vgez%eB?qmm~nyRKZR-CaK~o*F7$K#|<hIn<r__o5-H$ctlx
z^4omO)Ey>FOiWIoo9Bb|9F$*HAh>Vez6HehhVv$<+syz-79ok6fEdS~dZN$@R+y69
zugwgNT~T2HuJ<sy-appD!YRM$X<_63@`$T6Z2TLXoY{7fR4-_NxBnAhqr&G7aHJ}L
zLFWk-ad4n@?524WtUT5ykfQEq++${eG|<;~AJ_A1iZ1c_^W*tuT8-Dc4jj;{7V!28
zjubq3LWO5*XIp1&C_+>%3#1RI|9;`5qoqv;v}`rn0R97j7N|UkS8^0h@LCVunNk-U
z64kSyIvnF~(#!vDvr;qa>C^cqmSRI5@d$r_phZtbI~|C_#e&o+(F*S0^u@(Rn)Jh}
zXGKIRK0XTNJ=@n@b5q^ezBO6?h0@p=x<@04`lg>v+6|gteUNa4EbKwb!oqR~tPU1;
zM@L3jSy+;L9NPfvH44lq6{dgra*JuZKY`-h*|X56qneQalaP><#2Yd7AU9?8qrtGz
z&T#6@uYRX%7R(v$9I(txn;<4P4#(f5{Y3PykTiRkIf!5D0D;=GbVLAyD&0`PDLRBG
z@iZh{llOvT){G(N464;kedxx6Zv|v+wkuZ@khvKgyCZKnfTe&k7#JQNPL{U!V)EFr
z2+c4`<>xo~uozDoew*ZHhv^7h%V7NGmzI5dC4rQUI91fpu!~N)?)1J?$Tptqwfcb?
zAC$`PDFzo8vDIC=8<3Y*Q&qMH!y8a(<pfGu#~wTi-}<H>I9;at^>#p=>Rzi*`9iBj
zLd&g;lF%dMG4yF;xJXT-{mP(R;crVwD9FDS9CitxPGoJ4eu?7?B^BKYdG(u~o<u@|
zUcgfWB~F;o3|c=QzLgucV22Q=>WO|A%BOz*_PYov60A}=xq!$Clp)qjmuziMd6+OV
zOKEGfhAc!5$<*EEaA2}#mlnENLE;gnY)-QHcrO`9!_26_gT(H%IwE708;`l1KM3C2
z!=fY18XTT;Jk{V{{8&fV0oucAP0ce{jDCAJCInV>)BbWT)Yld-5r~H;01Dg~94nKZ
zJo(gt1Ff&m&2uLWn@Lic<t%j^{bcQW>(6uMF(NM@)Q*u&09B<>Z5mnSy3H%52(4I*
zWDA^R1mBI7fnJ3KG;kGyR|I+85w8HKtAXHd8Z=0-{&^$j!(|qOH-w<ssCaE8JhI9a
ze3Vcwp;q3DDF?Xi!?@G>NXMXIB^h)u|8cOdjDbLNkb<H!=K2#&wNseVrFA+CpYG(m
zzZ#r_zoxhAx))|=o8j-8YDY08ia-AKbc~OWPdljd)i&Uzgr<|RPcN|;r@zLe7aT(%
zRMm5qL5&_6F1)CSM3-Xv-<`<?@fG4+QN=(e00O;kBWqg;LyR!R*$h6dOmL?yHs7SE
zIT;5QF=UQv*CP@Qj_ZH-F6W<$wvj%4aqBQVz2M_RCQRhoVg$*%xbQ5NOus0=rSy+*
zKo*=t8t}v}Nq}t)KiylRSI|wTi<STx8?guBdd1C~LDpxhTlwUa(FFnF>VQdscvRiV
zI54u^fv$*5b}}rLcIz;?a<Y4Jf@Mr9oP7&SoW|e{GzN0Hy_h5dx#tq90BLERz?817
z)6t~<x?VFsH*ZO^fS^OnLu%!x&tl;*0Kb!BEuriL^%d#1z)Ok>S+Qg%bm0ejDTA&E
zirwG7Zxt90-~rxv4ModOp(_J#ekYVxm!Nc-3bk3V2DOn;2o4Z(u$bDGndVna2!h5{
zwbx-<8XBqBpO9Vv@vdU6!5o#7p$-o<_>y9-m_{_HMy@z3nO`h;5v7_ZxU~djGV}~b
z+Usue_4M{GfwaM&SXNecY3mBOb^OkrJ&Q~j(vkd2v7-#ofAICo7ZxbqLa|&$R8%w=
zPmWg^&CCyTJ$p~Q$}&UQOl`nPwR;A`I^bF^NjxDXZv~-L&o_Iq7>7G1v^wwvnPNaO
z2@VTO1(uxTh|GCVtTTjytkdeCH<)~qQ!mmb+p^ncC>Lt$ZP3Y?1FZOyLJf&*Bx@}3
zh*p3X3n|d3g#*ksP-sNmA)&5Xuuk~xCX=DnA`R41s9!}uBdW&WCWs?6B5p3qA&2q&
znkM;nX^$#M2u$J9l6Ql0a+!2%=_x4>v?MFQp`*de1l<zgkHUi|6UxySBEk8VlT}w&
zrwEoNU4HN{fN^iQF&x}~TN^7AA@|Hua*92FJU7YcFbkft7cC14Vx;18*G&Y3V;S=j
zUl+KaY21gSoW*_<>`s|Ew($uGsPI2OJIk`N21y5^#~qy8_L>2t1=X%%%O&986L_9$
zc6V#*0uPcUi2KFI{C45;d;Jsh;$Kdq{H~x#IB9Kft@_4YE!gdut!X&^1b54f2miUG
z$L3Wip}vF32F{*_xm>6rz@R|oD*>D~lejGC6>0@>i?$8y7N7>R+V?>gMAZ^1b~a$|
zFJ8RB#e+l+9#3_LeaU9y<Kwpa;Sn3+^P7>Z!omuYFM})+s6Q3fdIi~2`OEjFEwAnr
z=Sdp})-DSS?Bv?7Sjx9M?o{_bVR`xLxd!WJzgw|yJb0fQT}ae>jhh-ajWbbY>eipi
zH+}m-&@D4m><3{u%%sHY?Q}<aaYiO@zPi3VK_etQ!{~fSk8wz*{qy~w*0wk~uRZ3w
z?Rn}==fd19(LyALfv^?5%Sgju#DXdA>{;3*Q#{vU{Oj|lV|X8ZADAzUi4m>X9OmHI
zu$7RIsI94Kf%QfO&0N83ss!jdlFD~PekfQ_wouw_S8u3Hz%Nrv6p$<`n>}^TaLQtX
zrx+OYR(;{)<GU)%w|b3oMn!e%jq?fM4=9}!g!(y(+ps8Ap@9Td?@`$kA}b!Wt;hvb
z@P%X7J;=rzba}+{ulq2%cPyToUKWUv$mybWSNHJ^x_F}cwxW9`342xzy{8*gxFZKY
z<m8|(FNK-DvflL{U|j?Ssn{3f{qEs8&|C=dGL>|;^VAaC>0J-Cq$^0m7{mQ4eVy7L
zlGyZ@yGBNDKxfdWrw5SWzmJVLQ!{c?3@2j3qrNDB8LCMyANNQ8C+I2zBWKPdo^a&N
z!}&Dr>=FD{@47wlivwqyee?wLtmLAALrJkQotb%3;)_KN{5u#`FmR@VN6SP+9GLze
z`vUMYL<>2qfQf72cd^QV%wNJgJmEp%Y|1M9!rCo5%vt*9F3Jq#kJ-=a+qU!f^vOt5
zf&(yNV<yS{T_ac5ZkURR`E7Iv{_y`+X~I>zvB_H->q@e9xGz0Z!@IjX#pr=!(a`pa
z>b*U)Pv9EBUzmX^){WHy_R~sfDj9-mM`z>29dG)ItLK@=^nVO*SDvtN7SQLpyYq?4
zIEVTuyJn#r7-SGm);swOidFZZl2zNLM}(spC+DQmYbp11+h?w#n(=vRO4Imf0?$EB
zPdE`E!8|B3qU?gBWh3G-@9X)jOmx_9{@F>t9K(LmiYL5jX&r;hcE(}fON)knj+4m_
z#kdc$a&;;4X*&$=jl5<pj_TWHrE}uYk}yHxzWOAoj+oJLOn67tdG9*ai3YEkCCJQh
z>aE79DLj{D*Ol>LRW;4Mwp9;KJ{ufAOSlOVE2w_MS300>;|k7W`V?j!o(@Nw{e5S^
zn*rYTDE7GfN?O5gbML~P+(<Y#a{TT`FG3qBvKl7USqw^V`D>3x(tK_T$(mbs{pnZM
z7|zvot{F;T#k2NLUcNjStmU$oeSl5SW*@m(Ps`;od#w3=*yOcl<Tyk!Wc1dg0WLmN
z+eyN{&oRk6jacx@P5b77DmwBf04m7S(7c^48nv0oC?J5^iiZG~93|vje31)V4U*C_
z(r!|XaURoiJ-l^T*vIGMz!_;C(T%mW3)ilxgPrh4wyrno>av~b>xCy0_4@5N-`eU8
za*Msi-r60b%`*dmYIIce_+K)E-@)aEy7Q3J8e~!w(;3INR{B>g^gQ+{Yh{PN9^`XB
zd-l`o2NP=tZ79mpn_w~|wcEYkD^0qTUbA-$Wb$zc(k;*x{SGSc60I|CD*_a%dmV_v
z8_t=t(mSa~!$XxFI!opt7bxJR+YrW$98u2>or(q6WC&$l<5+tF$-ofIW+TmqL=?tW
zL{7ksG@#1>_4I)xZ#ekLYaM(ziwn^^nlq&f{z(AP_;3F0sASE`pvq=A&y=a2=6NJ{
z|Nc7^kimf52sbYp8Q1@j+3}%-YUR5c(lIjBGzVe#N2z0Ok`$4LlNcN(yDPp&GbN{W
z?)9uM=}*et9WRpEptb;7HVAfzvlH)rco`ruynFM<HW+vR*aqYLA00=Of9#%d;BP<k
zAiZB8B`=SQjvk8ZogRM)3iB2C$E8NI(AKtuQo)+bv<z)+2U;+Kzd%KNw|zHN{$)wG
zor$DmFZrCA!~=u1aA16i@1d0VFShCNUpNnAs&$b%ty3jalKrnTO9LIRJY<=97}d%f
zPZ?_|2B#-zxs?}LVwXR|x32OSB**vIDo8lKxWcm#E?5s{y=42NM@OoNDi7JaNi>P5
zFT_jvk>rD5DE&56vncfKg}tQ3{(DLJtB3}Six#AuZiR<2YZ=oflM~Sxrk0M?3WZD8
z*5b~dWAgbr$H`WQu7{0#X@2PK&8t-3dzN_DDn~qPX=}~rFGCY2m$J6qa8$$j;StH;
zV1M8jnC&pbPT^)@Mv}Dpx!mJj=M_$<qd(!z!wHG2j~%z8*fCL|!CE0q_*1W)!GzJd
z%W*Omd*IxKRumoqF}?5J0pm%0$+m*6?*BnYLxV1>r_8EJY8p?3slFr&ZfWM*M`Kv~
z=);fx>#5UtdE4y!?UDSc<tH$j7j^jpG3=ImXlC`a?UEqS(vX07S68UG55u5TK%`TZ
zFoC(e2=8>UFxt-!y5#DWz|hQvMB<nhD{EU74vjBV+P>&h%0E7pw_XUNU?=w!0HH6-
z+S<>wPYL5PHq&1Cm~h4dyel5}92AL~lZzZ1_JuyHq=V;fHYmMUG;U;sn<3qJzdbzK
zLN8t_%r6X1`9Aeb&qv9p7^B2(|L;o;yr=l>*<nA7qc8m%@wS~?tN!v!Bd=j0nw~rR
z>1dQA9N4h1Nb^h@%}=m)Vm7~?Yw5TKXVywqD?c7>93D^0qLp&ep^b$vVypZzgIJ7t
zd5y)hQC8SsWPRkcSvC<?`4#<eTC<^`wGhu2_6*z~sBa3ik1zpyGe7Iy22gM(2ste_
zm#y@5S5gu8`FJI`oKCX9<U~cEHb&N^S<*@?iivrlEl;4jUt^pP?s^GuR6y0rx)FA5
zXo<LHod-@nwBo>q5xI|zwDflSCk5q|sw@pP$)8-jIv<Z3VsMi|(a{r^cs4hQspuLm
z@r0!!dl|rB*f#`OYe3M*0*hTyJ!;cIX9JUwBrdaTSwlAHpab2B5LlZYnI(;h-@jkn
zyXTf<4%F5ava{9MsgJ}Ofgef%%J2)>?Z&8Mye<MjrOD3r7E0tONyClv*u3>neQ^JR
z#ld};g#Qug->bYDqeZfy=?S{!(x7D-;cTEcRIs7I6%IBwDcG$67Y{wbVC3s2X^-Ur
zG&dQ_zqq-epsv0&cj6MGeMQ4$U=>&}FkzO97w<w#Qe!m~>cEU>{Z=<$BRC==9V#Sg
z;Iv2k!a)BPY>Dyf3I}XnGcFQ=(VRY=1|K+OF^oJkFv!&*$jQH+xYmr*y`PePv8LDj
z>PEcf5GZ=H*Z+>clvPv^{sLy+px2lz9~W`^8)}b(n*}e!50ZA`KCkwJXtZaqLvI|i
zOFjd`A|$F__nj?Na6|58u-(A{q!As04wg7DBq2Tx5n2(Dlb{y=0F|0rZ2?*=1_imo
z1p&x$4uVtNch~|TGWy=d6UbXecH*MpSSwbUw!5er?h;{_{jAaY`s1%?37P31#5{~Z
zWegS5e2_8?K?rtY;u>hXa+QvkqfGAi7+omL;6b(c>#IFpB(t_X^~W(xk&pPF2Vne#
z60Mgnahsd83{!Tps(V0l{dx8VX8SDoBN0~aaSnpxA+{$Zm<w1#MI-O)SzsO<5WbVI
zrXPZlL|Y#H^;PIUZ+w4+538iMpgxS*zC;I*rn?OqFox~M_(}tO=L6Y`XNT|cdBK&h
zK3&(Cd?6O^LIK~g__fDFHaY8Sl-jM@6^LE4mbBN12G|HD*~}+i&kNCdlDg2|mcdqE
z{`DK8zrJ$iYQji^R&^L}#rhw=h6uq>{hAUh|MDPZYcBZrqZj&bdi?csNDArowy+U_
znT&ew-?LXw9*p`Qzk00LA*WghvT#U0=>R3bczc>uBihSPU?i{@@u<SvF0SSJHzRVY
zT$fx)(A|l0?|xXJ-_CR83V{AwxA;C;Yi+p~W|hwj>_;VuD$14T=FdCRn|!v=hba4b
z7z<i{)Y7f^mX+N}u>vRm^Dw#<9Im=q4AYQwcE1y_TIHs*UZpM411dE5`5b=I&-rbY
zFZz|a6$X@8KpQxX<3D`ljjj^&_**WuA?GpcNd{v+5M)E>)ZlJJyV#_{3`N`k=&Cu1
z>VEiz>>%x!kCKv-AwUdp3<ajEq=CCE2rT^Y)zJu8(x8IeMs_mze+g{-j_)79&$t-<
zMRgtmM@GUgY&nCrWkOzSC}px+hg?*T1@lZQ2t2rW=y~m+On~}wB?+`d{CmL#PB42!
zX0VVLTEkSK?IkBS7nNs0M?f9&(CTFQHZ(XmAwB(~L@S&%I5Kj|)7bI?iYk;GpM{Aq
zF48??K2N>8Qet8%*9svZLB|=pHM+OJz;@X7!gjn>40RpAvEgntM}X+EK`g-?|63rL
zu4`=6k9Ig(qP5=Sp0gyf6Bzq^e|4=^j}#R);K%1RboPH0N^1UND3N~B%bt)VrKg9x
zf5*|$zuK-O36<RVi5JDd?Ue5@_W<oj174@y-F3hMi^|L3z6)hUpMx1t5q5M!_z!kr
zNP+97&!jahy>cZUcdRoTcdQ!95`E7U`HWwIh5<Y+G#k2g1w;zeE+VfWbn7%XClE<1
z>;fH3@C8g1LH4tsdiXHSZ{^*Y0G;MHbVS51`Qlu8N8R7K!IX5+)K5Hqa*Y4P$+46B
zJbrw)?+z|MqvgoF@bH^lUyWdb`5oq4aoXsM5e?@-4!xvt=Y-GkQ{xWxT5wdEs*yv%
zK_%L^<Q24IeH%1yd9HuH){K=thVL3k1qVlE?h)0+x#x`dFtKwV#_#n<GhUo6R=R;&
zN)x~n8N!3jhI)q^lzzAF^kE?m-^XgrLM<P&jbNi_oZf_yV#l|OVjo@viJgEymEcX6
zlhx6^1d7q!Lp5}?iT^JC<R@B!f8tOFy1xgzo{m%6PZeR%JXEv06Bcb!etCBbj1FI;
zm+E1l@AE_psWR>1!_AXXkW~1ph|G7ft2WclF_Rd$K0`xveFMs8*|SM&6RJF)iKF$Y
z&2&}qppDjPF;7o_%gWQ}O5yNb@#z;YTv5nTqh-j=DfBqf_XMq6EK4`{=$pZ^e-fGd
zzgT$}-iQFt3g$vlNB5HzA|eD#`VfIH%DdRP|1*5y;O-STt4$w4iZgsQAb`XE+ftw^
zpfL1~g2P_TcI%;b)Wee^+9DFq?HCXiUp)_Kv>}+4jgG<fH$f>)5q0Wa051G=f;kJy
z<#5~H7F)mCQ3S$Pk1$lY<;bA>z+9eF=!mErsMVf^uT+13v}Ys@B}RC&3-v{*R8D%H
z-i_>{>y}=xQ^OfmGcKRCEGN<2y?W6HUPlX2YR7)d{<Obmceq}u-QQ!jd!jq(mx)n^
zth{a7Qz#E!ZHOo>(0r99Q~^8u*uw5=AOT6<y^9B}W73n=ll05Ma~72LP<CTbl9zu6
zamE<hjdL_IzzC%Pt+cSooaP8%6d5qLFUtjc?0XOAtlZ7xP)kA`1;~B~4rOH7frhyS
z8+ELi1@5yF^=({$ZsWD@#`&qq$;m%T!KKIs+oFL!t^yXMn<4u#Z67poxM{9g#h<_X
z25qQR2Hhq|!T|PIX1oARb7)gK31p^)sCBwfK~u)kl5^v0-FsOYULZ!RG4J1B4n0@<
z7r1Eym(}Pf^65mm%-7HqAnV;CgNlidpPw?^Q7^S`8ZI)j!ENpe!bfoO%6(8Ry8v4n
z?4zJpQdfVA)E!XFLa#wTg`HbD*e!v>mc6UngJpgB_s|P8rL}@k1UoBbf$!y#PI*t8
z9*GB7h)6vC6JTkAQ~*8QMq@cx<%A`GzrU6j_Wy|H{rf+&Ao}~VMHpq^<?Lr)r(ZMU
z#8~CM?^6m+#Yzf49U?+U9WF}ngpU%4e?KNtpKJi2!Gi%-@T%Lr1jGVkjt+XCS^_eH
ziF|9~<vm;}H3x@*iRi&_>(-mo`vwwL?~4n;bgZNRP*Ew`nGGAQ@p*_pWzff8g@?9T
zsUMh=6aVc_%>zr9`r4mi=VDrRj|i)|a_=LaK3&iAp77msD3oE?zoNJQ`M#SyChd$?
z%xaosU%s>w6dwX(4Tc&1W>*pv#U30i@$Aen1SQJY)6|Gii;~oQdnGBUSuH6^J7g%h
zyK>1HMf!XAN_XjZj3>GkZ~L?861ZN~-@S0T+4kb)W={XA&;4zC;oPeS!yD#cN5R)1
z_PrjhuV<)k;oDu`G_U?1nN?q+3H~Q;`v*Mn_fYyDpLOA@UUo4t2{$+Y0F>1phYkuO
zAfh~JcR@&t^@62fW1Y2Nh}zU18F(mwh;XmxbR9Ak6n>uE=e8SHlX21JX0*b)76CaX
zl$TtNp;#0n`42hh&#yknl~7k#co_dMn@?C@&ywhjCIXz82U*vnJ^J{U{|N&9(@W0$
z2x1cS`JiUHhDd>e9}Cd<q6&gy(3>~OP#G<Q#s>Oi*t-;qsx+NsSNJzV419O?_`ifh
zQBbde^zaySeGPj$K&P7k!@Ud@2sO&#1yJvXGN=h??@*&P{B{gt?ZEYjPnLVd{b^HR
zODbe}<|tdlp{<_)vSs}!1;_(k2ogAGMaLJI_djE&%3p;Y8QP)zBpd2U#`Vk<7al$_
zeymhOUjArVBN_Lz=u0|7QyWjURRw(RpL~|&GkYGkTj&G<7nPNya8QvwC*cG;K^RBq
zz`a5Mfbv34#}o)jNV_xYJ>iq~Nq9~eqECS9Smbw(K5tiOk~vr{hF~HXQ=yGMH9h?<
z6tuvL6;>AkTu49!`gO{n&7!Ya4*|cM2rm8KlBj<ghF~6e9*HsekH{t(cC=f-AGQb7
z)+u3{EHFtq^I6zv^SgK7i>;ro-<kbRBOCzh(Bkl)bD%g1dX;usObdDwQLhtn?ty>b
zt)>Y2aB=O%?XV|As|VG(f8C`9x9Og1D18922S5KvhY<OI)kb_Rw3O&&s}(Br`m{`x
z?kj8DXr2gTC$<rKD`1}qmFZLf%HZ_O>tKl3S=rf+CMBoAd$h0lrzUju!wy<gy(Ow_
zD(TIAWzJe77hwx4fX3}m^;rC_8-Vr!rBChPEzFhOZQXL(UK<6z3%PEP-6_&OrqhwZ
zNvloVBa<1jRom_!d8jZ58yT8J6u>rLbg-QaY9PBAGJ3T~J#6cfD>t;2fzMQ3lSa<H
zr;8bP?h)!MhH~hlXs`u!S))MnL@7mCrqMak&Om7EKrP@n)UuBnh}%b$R8$DiirTrk
zws5*wFnX}Db&|SqZe>Lk8mwVkfo#%@wotg%ZP>pN;I$|;->KdV-4{4!+vll0>N;jC
zeYzmK<REINhO4?~DhQ_2;q0#uUVnX%!ZU-!X9Yz?y?l*{4Dp!kl=O6bmfI5Bbk?-k
zn5}4UTqn2Lp0Tu9kS?|7i%^peWHl9dNH(jp_qCj-^4muqL8Tk{M$x_wAZMq++=JYl
zA9ZuTHMA~)fpNe^6S%_r_wQ*&)iP8fMclQa?MeywI8Cuqv}KVi+J4Rz`dcL>BwjP3
zy$p~W81i5H?s}qzMqScH{+XLuCO;41eV{Mx(;Jv>uyi;L8fkTbBIP&gfcuvdDsj(P
z8j^xSLr+5k5+~YkbqHXKI5aZD!F3Aj)j=|t%Z=WuzkKMk8f*i=sb7CmHi&Q)o(MJS
z3Y>$E9j@QKE<e#8rGvrq*6z%+JPZ?xQ}-G4<?r_7p^7xrBhN^zLyH9}g*p}T_gfCO
zPX0+1BBgS(c5SL?j|JG66G0~^81u2JG{IR61^?lhupNb!2rB}?$^RI^vHv2<dYnc(
z5I`^-tpGbSNIFnDGK2@(=y*=tR}NY<881LDIzTnaJ9p$@+g(%{ct;?rbX^n}m<M(5
zG}@dEXv(!6=xC9^rK_0a5BqX}H+KmPA`_@97Pc&<N!Gx?d$IeAO65CnB$!|Dd7{Q>
zFLXG9$-p=!6^ogMG9m-OLteD|&e~`_^wL%j7y0@g5`u2te77x#=BPxwe(@;m#I|*~
zP|#hL8t9+0duw_P7z!9AP@QHuqCsB~uoCEO*O-rp#)(#tum_w2KWHId{uFG=XI{Q0
zj@A^^*a@ECH#7s+16Vtl%__F+P&3jF^q{5*8Dy(1!gH8_Z-I|=Lr@NE3Uupqki`gh
zp`;1LOJkQza0;WnSGdjlN=Y*hVg&lF!{v>CR8e5yJ*po6J4p%><$rxD=;wc<_y07v
z{m;+--xM8ST>s~1|1-J%OBDRyIV6TZVY6t|OyE#_2wR#2A09q>dy!c-)DrbKKp#Uq
z2+Oje#D!n~3KbrsQ1+7o3tuw`XewD*l+e|PIyWE;r*(D`5y)=G3@U=?-v3+H|JB}?
z$78v+Yd_IEE@jA+La7vGN@S=Ek)jaEkPsP?%(H4$W-BU`DMAP<k(mZVQsyCZk<29X
z^!Sd;T5G?<e)s<N{`U9J_xpYCKkt(JdEEDXP3L)>$8ntR&d^hAlb_5od6aWW$GpLw
zdj$@$-T?vR#3s5`7;-2E5>O5?FU&&-VvT?|XK->KBZ4lir_iquR}*3>XJi!1eFcOy
z0`?PoPe?OPL$h9!ZQVmW6TpW&)~I724#SBZ;Sg~*G`_E>c!&V%Wu;^iLPlO_so^v|
zGY>nEc-Wx!5WpK@3$bJyKmxA@bqBN}b%3~_uV_EK2ppVjmki9z2{`n6!AG%c_yJ>f
zqT&V4M|8bNqr>$wfdxQMLCiXE+FCVS`|^4jaaMw>0C@aE!dFd-E&y5igA33q8az^o
z=qk(>%1?C!D8WPrhY|v)YR!kI;6ZQ(dLSI-F~ny90c-<q&85ke66Gc%E4$>$g;Sgq
zNQon8niQPyv*z;(etzKXZ_m#qmb5Td<K^eCoM}ac*#=&9d*F*l-b;k*<aq*P5IY>W
z0mXt%04L`<trnPQ5lj`~HNj)c0(VyQz$zh@P#{_nO>f+I4oN#X9#1Y=alkX3knIrI
z%nn()C*c%6zmKtpe@$@9B#!wB_^7EtZ9}RMlF*JO%}Ub7F-z0>`1%SVD)&X6(S^(8
z;BDh8MI8+&LFvQ~hcq*o3POl>{N%Se7G-rmyug<$p&*ALv4UF^7oH~zPV>GyPd?mw
zM}I3QWP%6-F7MMye-GoVLjWUu^{79COAsXbL4d@Js7wHMu=-%>#xyiKz#G^PkHj1x
zF&T13OUo367*@D$mEfdY|L%Y<xJ#^!+SYzApd5>dj(#g@+kf8*?&GXr`j%jJ`9kW<
zis!AEAyWH;(Amep08Z4Mn)d<)u^32kQ5O<AcKo;>yxwbp&KziW>|y5;5~`a<u;>d9
zL@rU$&)I&UurceojNT>z8_WZ`h%X#s3184fSP(W^%}~F6`zAWDe!~V~N$RvgOkQ4V
z7E&uecu_Ossfh^)9UGhZbKU%A^3;rc+9uO;bK-@;6c$}W!>9Lt`|TpU_#2x!@KA87
z9h`SWpp(4xBruT>hQE~`^w-T<FqkBdjnJAIwK<{hm64Vtf+3GRiuq0c7U9FKfBfnD
zIg8BJ>pr6>)whbGt-vUZJ(wZeI|Qc3V5>ju-$o$;XGPjq0>>m@$X#QeKmRzhNYHxg
z3qXN{xS2uTyJ{i6efw(%NWuNadMa7r*>C`0ck1MY4I#%o<Zh9<L5#C{x)5S9OU*S^
z`a-(a5dv2c>v@Q@?%?lO!2nMylc&r99Ez|2J#a_5VB{gZ{vBNuY~#Z#Q8Hs8LrRvG
z0daru!#HT^#q5=4Vqy|6E%9YG_y~H?f%WnS!lC>d%f?VvWx<=p7rR@}=jqWdzmECu
zhvZNKt%d`d0W@{>y2dC@lt2WgQ$nIG1MEF=FwUrkgJ2i&mp=ojK)|@>sY1#9J*vLg
zMOg}^m{#^es{eYX6*fLsib_gY@G?nX0bg*f#Uqgs2t7Z?i!xCZ)OrRm@V}#9*N*WL
zmhOP!xP;({VelQ(hg|Lh+~<WcsSCFPWE3ul$XzBTCDrj`Y)lY4M>pGJ#Nr1Xe;W4K
z@vYB5$}We3haY*UnZ+D64?E))%r;xeOykOJF-GMNgOiz`oeX=sEC8W+%-9&LI|$td
zIxiSW+W4<pv4TMNn2XqW-{N&f%FD^+gw0xgwI-7e%d0M4nF9jL-LMcNq!;2QUS1%F
zvP*rAty!jS555UB*<fh6%jdDgA^cSXF))WZJCKww2N53e88f!nv112O;ceuQe2#V%
zX7!2cPRK`yWk2ce@rOYOF~wg0P6dKFvJ>ZHV+)>luN%*Y3Ef^sAK7zB6KOD(o=EEN
zNqa@T_5#Ash04YG0Bc>eg}~=f6(+&%pwwkH*nm8F@&x;roxl1-me0ULZZx+zpL{O!
zbh)*03)9pKm{e8J$=~9XerIMg+}hO5mQNVJ*qu<?%^hrHq^B2^s;sQk04XXwU_YbJ
z*cjnTB8nyyXd*U*ZnPIkZ^bX?AESUg0K0>$`KNKUIx@W@Ou43R@S7)|LLianggI;P
zgf}V-6t6ef!Mub}pRf>qTPmaAFs#A8T@Mi^aUcuq`ecZGSj?vPH{z-<G5_OT#aKie
zc~PL|iE*&r`TBFKhq2nk!}5u{>wwz~z~l0IC~RoXZ`gW3npnl=Wpv+meJ%p-X#>37
zIN+$#eE3{_SEBakU|wEcRIGXjLo@RI#amV$0sNhM5C+sp!j+JUE}f^S$7!MH4@qy%
zU}FW$d*FEhsq2jhcdBEsPcI+o`9X~RiFqFEvv?FRcWMYR`<x%yI*_Nw;p=)~b419|
zreVQ2;V|hICNm4w9UZfAZf3xYguvV|GbR3O6U6=mC;wS2z*m+XDNxW~0RfNLzqPx~
zS*WY2K>?`-jxiyd!at7q(~(yQb0CcO^=)6>uwjSmX9B!P)7-}=U)W}aLywf?w6F&f
z1_0@#(BU?hoB-2msf>xcJ`!M4KSKUYtcDbbnGl(z3N;Yll8k;o7;zZNT8M`y06Clt
zB%UMCO6JV})C+=B0(sHO5)=1vW=)K^FJ2%<d)O+%t?h<e6HsaI1xE3wg;`ypyL-=o
z_THl0mlm&US6BUDAqWKABa;_rQy0(Tuo{P4*9?xGS6zot#!QCez;5tMNpt`|{~ISO
zT)`eAwjS6-1|FHERKv+-93G-Nlfs0-3kF3Z-d9?a`O<PiV8GJz3Z{ObPgM~Y0?gW_
z%$9h+_u#gANaqMrlE9LQQ;tme@IeuCh}1`yj5~qXkssUh908uGA?hNgdhm|mMIb?}
zqNwO<pcp7Q^KA><i+LgD084}bLoi?%YM4q<M9edRQ-QdC5NLyV8Iz_B7^%2dv{7z#
z4IF%Sl@e3Liw`{Gyp-(kEW{WOMkSft3Ph-j!-E(f6VwO33spD^_!TN{(U4jrA`!D>
z<6>_HxR|SBcx>@Sz3z)s?mEK*#07@zDW<T)B*zJ{<HgAx%)Mkw1s7I&7*pcuAGeEq
zxw7x+O}WZTFLt9?0W@u5k_A(X*zYQb*QlI7zbC1FFLH0N)f`~i9DLBEYCGX$&-N+1
z_xGJ;F+a-qns66h_YrwH^yt_=4ol1pz7$yhZiYvL(w+o;Gm&khJ4k2WSxTg|$G^vR
zL5wESk<PBJ2Z4c|s`5J}^byx!GAEAe>5zdibMXUl<K=0+(#J$XL2QG_W&(Ez6TAx2
z=_9`Un8cGt;SGe3LnZ>Ky90zaU2CD2ICZrtE%vnDDid>paN>X<<P^lQ+)4FlpTte9
zouHSfI1JV&aeo!UQ37g^Gf|I|m$&KM2FmxU0m_#TR?BBO|7CT`I+S#VAqF#nPw}SI
zoVQW(*x`bg!nSOd6OlgOA@@wJ<a)-R08mI%BW?v^@nQ>z3sJ}OqXWx5l8(Z_khcdV
z6P6eZ%)mYQZN#Bg7_S>OC_+=O|Bl#V4_G>Z={U)tFBh^&cqWTE+2_17hUz%yMjG3o
zpxxgmz_L6)rSHdNB(L7Lp}C@9r?v+WnsdYR*l6M@&^Oa04qI&Uj%pweXeMc-p2J;b
zSo~daeV`%#PW3hHOk_Cv1g_CpcJPBkIbpNL_NH#ng}jBH4@{18l2`WTrxZAhq)KiJ
zgtH}}RrVteD!56)aD--E`#YP2Q=MbiENP&D%wkF=fpQJ*Q<Q7npcnwEY(y6+VVCJ^
zsD$YtR8@zAh>-Pnr6oysS7r%V5w>XV0Rw?Ez-{g(t)y5leK^ZchxTjbqxRC4MVIeM
zd9H<jBe4F^0w(xvNJ^vNeaqCE{1iKsXhT$B8~FC^F{D>cIrn!Zfzmbmvz%W-=MDxK
zvpW%Cw{s4C)SK#oevbSUE7X^E!O57IPl9oZ9W;V3UO~)tHXYfCgw`$nt(aaRI2O;d
zFiU#rYAp8JE&<uQkT@hFa>#okWQD^S)3U3V)rB#?;!xE9rUA<VuS4`5uifsXb{W4u
z^m}CLdXW4Kk=YUBg`O@9IZPAI$7Q~RbL5|jc%0b8;G7utnScIW?D?cT7ACRCT1x$#
zzTJ~gaYEOaiaJQ+oy&R<rX|5*F-L;KV=Ag~!4oJiAq^VBr-lgc$;yY^Zh^rv9$Ow@
z>0s_YVs=g@ywiAZ&fb?t5f$b(EPW4rA$RlfRV2?)0HP1e<E6f%Gl|;PG;~(&KF+&o
zlT5t@k%ff?QU>i<7`CsEDF5`S9zC*%%AR<UL0S>WX30`GUCdA1K9Lg+AWUr8nHh;0
zm1kQ`9IY4_c9vOjck@WicnxQDZEU;tZi5r7Y{T`SDj4aFdwi8}+JynfB0f7ANO&@!
z<Psb93Wg!mm$0<Oo;v`qzdA@o!n=q~)FlH0VzNZ|?8NzYY*Br}odC5cHq*lug1DEF
zco=x36K26F1e3Gaoprvu<Il1jxaa{{5s_sfLYPgPSoVxLA7p4VY;JX1a&;7ygHh58
z82iZ<u(RxPZjOaAl2#Zr@cHQkZgU+!a|)I-1e_!evwHKLOysQP?t^BEgn2l2;?vX9
zdBtc*?<~U1CjGyV-I7mY|5B_xGUiMCr(n*Zg?&7XoHH=sJxe?qP_d2&B!Fs1N`u7+
z8kS&htR??Q(pDZG;)gAU1dIs1pi$CBl0odTainzp%s?bu+p$QsN1~#PS;=!cOa*Eb
zEKbuS%_K5GK6M4>1Wv#jl<3jXhqtz46J=b?Dclmg3>9So0XLW938W5Ks73NZH^lG6
zFrKfAl&7#u@+uJb5crDqyIcji888VU!T7TszAl?84qpvoScdQi9*1ElhC#8waRsAF
za0|@qiFm061#JE#+{O72O^QCaOT?1TZCii{*ArVtgvaD$m6|D2PKQ$=qMB0(vof)G
z6TW=8i2OUPx$#{6IIwjSU@<V^d_Ivu%n5<~gZrZp$kw6k?(WY+(YUEApE@QBCxnRk
zA55~DrJrVDYL@`tjD8ZWe%iEp^))x$7MiChf#I~j|I{>7B<C6{zOxpq6H)+TRfc7K
zk_3jR5M@BJCb;+@={pMYM2Up*pi`IMA)8Wmd!gO|+5%tQ#%n<KiAFWZbG3EXWnem{
zD9rF}Z~<yU_DJrstC>UxbTGdS=^|-G6KE>o{<|VroS>G9EOq~}fj2v`Lm{q<SFffN
zP-o2)YS5|4w5@`;g8}%#5heq}w`wxXZP&~9lXpvE288q^u_5-FlNsP#k+2Rmor&Cz
zh0~Gl)NJlOqM}(PV+&-_Ajl&w(t&L2Si;c9C~Z-LQeQpLYg<4RmO&qt`Z^0!2!DFu
z1KL-mfM_AcLBMumq(K&zhMucBk;nB}NTNN~QWGvXSZ15gokiRp)m<J;RwR(FM&e?K
zgCqfI7)-@#;C38~H#q<&KiiocTvu+IA+|^gNI9sf2*DMNEOA(<sloaj%u%AeMJ0BC
z2xK8-tw;75Y=PCPMG79kg02?KTqab_Kp|iX#Ag#-vaT0oE1U&>MqF1UW_x2$45>na
zvE&%dYli5?>H(L)8!M|JbX5-a@X15;9anf66EyC7;p8KI89A0a#hugB&1P|WH9cWs
zwu#N35d?1lVt{>jE9}}1BE)Ulo>7-xIEU<fdup2|>W)E8#23(gJg3A@u<VAD;Y*zb
z_1@oLpp$_hfi!}G!5&$n@h54nErWGnRHN`VP=$0yNNc0`lN1jp`Bbwb*t~&l{zr_B
z8*<-}oq=$p2F(Wvc)(Zb=>o4s%rzM}6p11OpTdV3Mk$?%nKRJik{$|UJ0mM0c800T
z;G_%X)Rz2>?fxJ~_?;lZDR!o%wr=#mIY~k7oqF~Nnm8s-5I<*BO0*HGTboIG#Chqi
zvMj}UAo+LaLF3HHRxC@t@~V|9Rw!H->V-g!h@>%ZgW!OzzcuhF5-?NZ-4ACu!qFua
z=u&jyx)0|%BqU$cQuAT|H8I$vN@Oqy#)sMTa%>Wabql;@ub^QXIr6v<x34A*&J>8J
z^`HAXxP`p1AL1{VjwVVP9G|AxBKgKP!D5X#NAB2h6bGZ-d5d)(cgd`lyn~{W>mQ`D
zr$wWp{FS4croS_+B95s;O)uq1rig}XWDSz}hdm3a{XiX(0VshK;oa)g{Xk>9V{xH_
z@QqPJehRRrK4%Cp1E~@>XIsY;^bt9r4q8d5BY$5iNa-!k5`@zh!TsJHJDxN*R)&jf
z5PA^2^+*a0DJ#jth&mL;PuvQR4_Vq(WHy;U!U=|?XB%b?ar?enc%G87o*YfL;6veB
z_I`SFL#(2LmqFf*b~O}8c@VHGkcolq0wC@<s+_&BMhfC4i6ek6;x<SA9B7&fODJY0
znnwk;kmgtSp=L*~zZ2Jm766~91PTG)!>*+m2+Sx3lxvN-S)A`p<|1&`R)n56abnR5
zF)U7U(r51;H&O6@%v>hz8r`T6-8jR7;+)>ku}^+_JD`2*q4P#|6}-+I0ZCZoArP=J
zj=}m{2|S84cQ`Gacu!r?2_8e--VZCqIyhS14-HK~74;A{)Ms4dIa<{Nk9JkuSlNG4
z=GW{|UYNecVz3kg8&dUBalB~tl$MnABT2F^NI=Ec40{AYr*Y!pfZD5&ruUt#1aoQ0
z9TE`;c=JRaWP(Xa4!~6$E>Q%#AlYiZlWj(7GIBm7`h?L*O#exJ8kiRuLX5KM7!0;h
zAkRR5O9cWHs9hhW<EKO%O^s=s*-oDPZyY5D_D5)QB02qq7BI63l0vjeu}vPk05&c$
zxP2t)nuE}TP+N%keBe`TkHmxzgJ=T^q!EbfAkI~|LwX4jLkhW{KmAk5L5P5UoM05_
z>rsnYi@}IRZj)u66dipGZ**y+gdA+4lc|uL{e-iE1@_KS|I=-f-JqM^34}2u{2|7S
zV$;LRvP?+;z%Nj>9gDBOZ}YW#Xl&3i-iUg3V=GhMKW&UJ50c;*K<*h`-K)OvWzvKI
zxJAzy4=}s8JpIw6eWax85fz25qp4%3FVC^(S4fPDbrc`Q+)Zm@8RTTK6oBqLAH6bA
z@u)`@>DIK)m?CtY+(Q5~JkPAvrvz7c#QEy><o<0>3^CNUd9&uxu1P@`b}8C|8w)N*
zC~2=jr~%<|sfp;Ru%$qKtIyWg#@OU(^M^lJZ4|nA6KV;_r~ukID+kvYlAJ(R&=TpP
zwUOZKMp&}{l>rE0ze5RQ)Q0(#7E0;RU3K>fP$K32+}Hnm?5qD5s>B?mF%w#^5RYvj
zD9Mv9@E9gj4JY{?uq#Oe$0P8wbKqE_M*=E_x?9~cf8ycfa9a{tMmRS4tRo!1Gm{RW
zk*<?VoJW*yJ|C~A@vz58dTb*4JvdI1=0~V#nZR}n<(u$wHlN~X>`V%Qq@5?=6=&F%
z{SYIvDN5B30B`i$whg<OXJ`@hDfeTju-QWbLNAaA1jcXUs4PJ0LizFlY&#DYlFkGE
zh+J~C16Tl<v4%;{Wrvhhqm=IDKT~KMH1Yvs!Z}K-4oPsxLZ=boBZ1K7q5_~2$U|5F
zS``9e{4QV&Xu26*y4<K0Ts;ftPC^<dba^9Pc%R`^->E=2777Xp9i94^D<XfgNU=>(
zI7ra)FA`%J85trF41CI8IDZ!CFlk%E^I-)gm2e9=NHZHWfHs2(HmPKl__HU$=FtQk
zFf9<axce5VBa7)U2PK;c0xQx&4~gVm-c^*!CHc%5$ktKOvt=eW;A3M@2q0)qeo{_D
zNyV{^pyGd^{llmKw`F+gTEl<P2Uf{UJLix(t(bA<$-cg`3SM1Sx)EJ3xSPx)lGdG~
zSrTc~2K|k4n6gAI6q1;kfv9guniAFL<$2)H#y|9lr<DJWbNJV_&=Eh_DJ!SQ=+k9z
zju8r}(@UhoDxoFP;ky3@<x}StQmwxejb-$}WSdy@!n^S#w#wO3RO=R7au<7Z24>y9
zq&|+5<$QU6tx`~2hFsdKo8e}qqHdd@G%lAb>|LAQ{_>W<ClT=qX6t9-hX<wwUUuqp
zns&dJ<!Ncsi4o7rXp`f4*ljM5J|@qu#oP2`gT<Yw&pR@tcYK-(FgsSuZPDJh;K~tT
zEqZ7wT=}h!+ZWf!JgE^525=ue6n}|Q7C3T*jrGt~U*v;_un<|1HTfaf@z4GKj)Qtl
z8*=2)WJ_1x+1c4n*AZvE+lebEQLW2+g9h94#l^(LE{%2;gE1ARS?K0!jv7X*aH!HQ
zx1T|gGBQP~$`TBzF5SGmXZBJk3}3k42B92R4B}ErduB#Pabv+`hE-!LNI;_vnFYd<
z$j?fYQQw(Hokn|D0G4pSEOT0kgO_E*sWG>6VZ`122!O-U3xyLGM#D#3cpg0}+0@%}
zJ^Ow4bvimLO!i*Pd+7|NZj>&E6&SATQJ&e+j4NnexX^`$um&TW*v06RlasqrW(JL=
zSWzVP_4F*~%iV`+ZqsyifEzuN_O&-z139_5F}b<ooSb6f;s(fVWA{gJ1FHDz8^23O
z8UkKEc;LV<Xl)$;+({OFxt;^LX%!wC8hU|<&C_$beX4*7)nW56ck0XoSPFUwY>85?
zX>DagancKrb`kv3JNuJT{c%(mqi@_L9O`s*qqqwp6w*>0+6Hu)fH`{tnh_TWRl1F}
z8O5gbHAF^3L&L__*3T>#rYofdx;Zvih-q&(*ry(QFA~<*W<nYMjc{%~mT3Fjg!{XF
zNk6beD>^~+1=35Z-T(a3XZkZPMbWZ!r%v5&bSbETvC4v3ip{duAHR*!wYAa72Qap8
z+Q`^0FJt)f0Xziz!|6<(osp2>zteLDziVFRQ9~E?L{4JXL`$5ag6<b}&n73+?I=;A
z%W++PKHdKPT`gK#YlkS(eXFZfV$zyCwfQc6ee0<d_v2CddFj=Z*vm{CpXoO}f4!d4
zefi%%T3vExDXjAT-~x#ExYA0WIWMz%;-g~+Ma4!g>Pf`m5mt)#X^%Tj-}5XfyTz8V
z^oNcW9KMZ!PY&IGZES3ckO2#PojcTe?f5>uf+J{^vyDl|=1dsKXILMX-_LGRJb{mk
zG&y#JD(Um3Z@Pa+<Ud?41{`ympgHiM1hNJAvTI%!`^YnujNeRb7D~TdEPDUmXFzJ)
zksbu-iNT_=0<N`R9SxuU$iR2LLPZ%a-JaO8OQ8(Cgg^|%@K<|MJ5Ue=a`(iS;Vtv;
z)LWakL_KLYdP%oxBmJ>)6Nhcxw6Xdd{(ePORaHH0+_-UL32GZ%r!1XRvp?JtG+0=*
z^D?Wr3B9CB$P+Zw(u7SXJXXzWM*eB(@Eh1Fm7*NPGEm?$-$hsIf66>kkSAJ^%EmGr
zQOCTDg+<ocS--i&a7ZYup`k(4;m4QHb4Xca&CCX#Hcn1V6vs<WB<yk5K;s9SMRWI2
z^a<0<fVkAG`{Tsa)S(7fW)_yhRc<o+5-0eW&g^vGxWQhz-**pN)$UyvR6f_7HHrU<
zF6aU)SFXGS1OsG+7rg@mK7^W*brn_<WQAu9rj_qi+4-@`Yd}gM^sO0R4`q+HoW~ux
zue7(f2%$|G%7d4HxV%8kFAKfR4A5le1(K`JpeKB`g`fNjgzxVS1sI+)#_=qwNHVq+
z?0!VmQ{6YKul&t^T#ijjS94W#YK*7SR}Z<sPqV-1EB(WQ%P+(s{y~%Oqc-)4i_J){
ze@rD)oW*)c;jL}rYgK*Hm<FbH^MU#&{b``7UXU?Dp52vdGCr7A`%5tomYwiOb^NZ5
zGbV|4qFWFJsf;!EoGyr|QX9IqYq^Pu(2UF1ts`-}N*!r!W9)4$(?4!2tW_x=eAugF
z5x>J*8IKuQ$uud)&bBX1u~*bG>a-z$h1{^W_P(&(0|NKFZd)_qr?3^j-v3ocH0s{?
zxUZDv_Riw2yrGI^by<VO=Wm3j9iZ9ZS7g;rxmJQd$gRrWG9+qA2@%2IdUfS3`{KK%
zs_Fz=&xX3B1iP(iW54XYNVh|X1vfNt+2-z(YT23?@%0ivq>J<x>Wd~<FKjn1g7`|M
zC9YY2ckpAUo*kmljxYVJVB#P;<P@?C5z#XX)gf*nvvbW82TuL@Tm5IbEfvS_!y(k~
z^<JB(h*KHws@Uy6J9-BGzHU`rZo;{E`RQMl^_QxNuWxen+>D!^xSCksckT)8YhUaY
zBXL?5R~f`8AAv5BUwr?!U+lUN9#KaB{g|@-(I>K=lmjK4$DY~8_!v7mt$lJh$ZIR}
zp7c}3fBedy4hpX(^P(HHxA%w1y&HR0yMK53YC1{@cCC9K4gb6dcWAh4-P--fmMcba
zpX8Nuzp`sfb*r{R#?@P!f5C-Hwyi@6{u$wAm6Wizy*1u%-n@ODRvpV=eBWAmK9-K|
z-Mib3(eu~0lED+Gx5|vyS}z(DxlHtS;J)OL!C=Or;CBvom}`PT1{UnM*U_I*cz7;s
z_-=Abl&3EyjFEa^EJOU<*SNUo(^;=1+zhrvE%Qj*TUO&;=GNzR=dO>)*Box^tOQu@
zQB=<DJg;Td$?xYGB5XE4;8k($bo$UJc^b!+Yt-KN&YUg{VAAzT`*>+Q*xy$s@5iyP
z6Vsz*sS>viSHCKi#|?Hfb62faJn*%7a(Z=D)w3hFma`mPvrF6FHa0`m9rHyk;wX(@
z)bcGUi!+h6%?VO8LPC2hx8D$Z;J%x~=c*q4sm0ViWJUc9mce2wiEe3*>1zupiRPG;
zp*{&!))1t6+)|4=Gcn2d{Ya#!-#fZCX}@x*`vJr3ZC2wm$2vr&>n>_%4&`iRX7=g{
zkdkU>$dAD}oW>L%Ad{I{=x}MeY)EJ&SCz`U3X?4x<ZWnepBZaZYC3GQHJ|=W{%?FH
zruwDXWiv^|v;bk@;?+2Fs!s{Yt@=&=P_^fY_pkkXSN4C5Q9k9qP4+_u56h5<v(wsI
zk>-|lU(ee&i~33*_e=9y`C4YkI$kP(v%5Lyfgy*P=^e@j;SZdQ9$rowQ*z~lqhq|^
zKR(*>^6}*WGqcdP&G|{6vM*jHXPQVgC1kg=mB@9CZ<oHd-oEFhL>4CX=OWd}W%xs-
zpIN!iy~ouw$2<?LVliz{5~+UQ^QnopUvIK=@ab30i*cId1{B+5-j%ieabW&_Uv?iO
zJB_pzMq~RvvE?3}m~uxoZB|o?cSs+<b?rJMg<igGWiQa<B7$M*tlZG=>;IYWp1DY;
zUdpG^H|KpWi3*Im2$kN834K4gFVUYS^xy)!v}3)62NiV;j#S4_W}UPg)6#=zM_1Eb
zC7o!O>Ej|vt<<xXZ~1XiNLI#rz5K3d@3n?;<Mxzp^~`0nmelW-ofkT0&*jb5r~2FH
z59nAm83oJo1^a*xl~WUYriXk-*;-b0h7|&;v*Nq+O<mGXbk|vU_<K4umFqHsCg=le
z?^1&R2}V+F=17ITPFGOmykYFg7=QJlt4#5{QJpKU?;5vP5mTQtyyu+&TRMQg4s$IX
z+YlCe!!=scZls-^sGD8zkNyE}<1r8T0>sW`<m!Z_cz83tSLz#yN>7|D<7809Jp)_v
zLTO+J=!{RYMoU<`Zm#3=^z?lA@ZmHf!3$sepBby?wiFxQ4)1+`&+FZui1Q807ufHg
ztCu|g^iscwbO(u1%M0hr4MrjzzlVyVU9gM{sAD<D)T>vY4fvdLLZ9xoic2(4B-^4=
zjbQnsCxy3a5}SE^)RZtfNhvvdADv#k8vIIc{CDC_hUyb%F-l3*Z{_4T8TW-uFKmfa
zaJqb65KpZ2*Ap+iUB_R9XrQDWYCIm)@6WyQj`|Qb9QxGwq406{Rk7+B%8UsiIUB0*
zI^3X^a2;F#d7%>}3C9(@H}_iS7zg4x*5f&J$#ZbxIS6^mUutnoDgZ52HGjW&)a4r&
zhZiY~*BBk05N#7wsN83Ehr01YCCflbgLWT{Vt0M2o2?57C4LZ)Af!Ls6Q=qV_u;!S
za(tXV&L};T>Fub=Twj55eO1R%N57O8+jMpb<dldVV`|!szbG#=Un{ks>N1tv;fHo)
zcF!%qK4o~mI#Ef)xxbxi(-CjlI}+t`I-^7@xM^#@-P+O->hMSjL_9Oo-*}{$xwyDS
z?cJyB52BdbmxgH0eO0%SpRg@!)MA$um5U0_hZxGQOzyMs?krGOUOWPR&jY@7QN4JV
zx(%9S^mp<*D#*~E?wpOpwiP1bY<C2ng$yZcqo)75#yToBtJ8a$0J2Nd1Cke4LxE6-
z5$BEzfB#nx)3^e0&;+(eLd4O>g)@(FV_Olr`+%b6s3lbh;G=hN(AVGJpYSesxh#To
z;fcCLDN6BO(LQ|MXo8XXd2zLh*raYEuW@t7GC>2q-jZ{{%W@JD_Mk*5I!S&0vcem1
z;OPP5yLayzz$?>gB!8*|Wex)@ofjxyS~a>18gTf)zp)6s)J{|&1MCrP$|iFtUbY#D
z#SvuH9^=fvrdoJFX*RcZ1EY+L%hz>!o~Trnpy1Mp+{yTi>B?uYs*K{|MA?rWU51o>
zeSNtXTH)pq6BEOQ<{NBKfO?UIidhwx=SKcEuyDcsOdq<`ocYhZyj;_9%J}z-7cXuH
zfst*~kss1aqYThG|8p#D-BZj=I5-(=+t#aFHJmptF4vt*o6e4$%O<;kU2`NW3yb%P
zb&S6GmS$!?C{-(@!cWx~AaUvA_VKvr`VBjcMq+b>f~Z%;Zml5dBlFg*gJi7f=#Zx!
zEu%71OcmPrI9<?8_9S(Rgo^M&DTd;0*Mb|iBW*rh30aXp%5i~>MdQ#<Jslk`E0x-Y
z$g*hiur+7mx60@pBjwDG6x<IrU3~LLKI(h*_SBt2c{}dhz1s!6#yc~A$in#CzWWLY
z_{InOz2qaEgqnL^$&d86zt=U77AQW0Db_(9>eyvb2u-2#j5ShB!mxZX1qzlo0%e8N
zA>5m?Yrm>0HD!2sxR=`06cHXC{sOd<TD4P$No?FMf!NrcYA&Nc606OJn0p_~&w{Wg
zp_BIUV`L)$-)nqosuc9%PEZd5&<W7w?b|E2R@A3fQ%lE7E9XTc9QwVosBZ_j`1Yks
zjJEM~`rIwp+7s@oYveN*DwoSM5qFh#90hi<$qA6SS<#Z#UvD;6&l_O(;(e*i>)ju=
zZr$pM*Dr5oW;Tvso;=e(2HxJf;`e>?gZpbRtNPY6rXz<3SB?dgTYruGes!yAS*Bv#
zC}Q~Qt+p%ICF;d21oTC>Fcp7M-J;y3Eo!gNHhqh>o8M%!pPh35sXq=bUuSo`tm#`)
z)3<7=+Hm_fZ%W$<Cc_p`QLR8_EXItzgiYhcrDAW*TssRHdb!BPR6UMRr8w@1<7I{~
zX{^ITF@WK3ZlZE7uml&lP33mUOU<4ko52f*(az|`O~sSFszIp(A&Kl~arjA|zi{D#
z3zqleh727Yo$C5};gq3o66%&YeQ}o_tW;5}?Hh>`)chSp=joc@0(ubA{J@|o#`|Ld
z>z2P(YQ;zGxz<Sq`KXqj0*3}W$q{`vK@H3VEJhAic1e^ENiL|nINPX^wb!Hb@fCOb
z9Ju=J+5%1Dg;u*3i{kxT6`v!>|09Vg%xaG@vO77R!WWhyk+YN9d;PcS{i9sJn#9Fp
z*if!5DY$y|sz3NjMXc>V?yNnxT!m#;e5<8RreA{HSJC>1(=<Z0n2AoP3ycN^=NX#&
zWP`dW3WXqO18P|U&wi=o9e7CHxa)&%Sorl8u?wBnas89qA3*wEDp+|$aFPlJwbvmK
z4zWA$eNWeJd&+&)mFAY(=C6j>ZGpx%!6pQ!QKWUnVXN_PJ;GJH*}m#Dj@$nhmznwd
z3aS@L<qxpqTD{qt;1cppFQz)COAd7ozN}2SD64WbLZd-=f&FRO<#j#s`?a^;E2G!7
zKK!ydq#!4pmgw_J)irdM@FD|4LvFVpw>XN(F`yW{;z4!0Zpf>H=f#DU-B<OQdsT)$
zLqvcX`#IB6&jmcF0MEC5nwW+u*VGTV%3c@?u9RmxEbt<Kc_nYM>T?`{h<JwR{mBv>
z1*cv}cdogy#7u9lS@in16RjW%<HU%IH8G>gxmO1%`&&jz74|NDl<m)8sQ?ni&Q|M6
zc4sLFzOwMsUQIeBz3PVfQ<FZNoE3-Tdyl_s*mKUGx-lvb$KIw*n@Zomf1fo4`b5%2
z*`=G(C@H?Rbcw&#!%F#mTJ+U*Y7P^1-dZY{HWrr77qrpMf}lsAQA|eGzr$^)(Ji^g
z=uJAkML><fRXbC?`VZ9|XOff_qR){4?!CO^W=p5q$DMd28y9l4!wb)$^(d{%C+}DC
zSC|ipXQY~T_l`VMk8?Ydgu6O~keDo36tlfi6Dx(89mmOs`x0!<AJ{M35|i%xIJ{XS
z<NCOL-CotgXX>_@er>3tu@OWNJjjxA$O5vyKQiOw?9^MF(Bsu+e!8c5TdYy*U|m6C
zcWhxyLsN8wmbyJxRpF0yg}<QW<35i=+5iZko)>ykN@6vs|8bnkCksya;L>HYK6>+)
z@;Vt(Jr5O(`i%lLI|KoP@Kswe>HMmWGXqP{s^dX488NA;{Gh=wS-Nzwc0tn|fD$vO
z+sKp~f5CVxzk1FQk7&G|O@NVTS8@7t@aKsrIzSX*(V<b^&nF$=)OS*kWyGaN9Dp6W
zRd3ak>0-V4;%U6Zvwdk_(iPKqOqma9WyG7!Aas27ZR~b-$E%Sf9n5EClydTVhll+E
z@u}y~i^c{HdTEvTog3=Ya-0=^<|ZL=n{H}LQ$gx0AsLT5T4w?!&Xi|WCraonJ6$>H
znt{x3@7ci9^+>wo)rJ#{PhU-Tr4drQ7ig2{KejUD?9R9dBt&m#z05n{Icf5Yb7#{Q
za<OV+QQb@PSHBxY4$YZrw1vy8&M;3O8|>So<ZNQe|7)dSzuL*4CwZxvG`qbOPe~w(
zd;7|@4heUnl7jy`yRbvApR(~(D22eyTWFtZWyl(&;wGB@TZ{oUP%g^#cYmvvlQ<m~
zDhQgu36Y$ImfnqwF)f9%tE1|q2!bKN6HxOWmu?RYlD}ez{PMStWijy@{MJRGCjc%6
zHc^M__FBn^c7yLmDKlTRGyU$_cnBa{xUr|S*JrB5Uq6T9=a{YR``#OJK6<<^m4Qv@
zfP7xTnktpbBgdDsyz#oV;*T_W?M9XmF5M2i0^^7sG=3M;-rPLHz_N7lVXt5M&pXfg
za0&2a($@LZZdj`ZUgKkikbwsP9hWZuVsndT<%Zn?xPkM`5j&c{(LT#6Te(h-Stx2b
zz#J!yx5l0^19o!p+x$}+nHO4U?iZWft*uq<^}PIP?>=trtlxi$usuN^N@sGh#Qf(n
zHNWap3G#BBx(-i%m|B}E!fpNX$6Xv0A<ktiAqQ8jpj3G-|Kmz&uFX+5n5|VP6u<X;
zY_FNE$bXd(Pny3WufM%s^E?D*F9%MYzPR%(e)l^4@QnwnF(Mb6!(PMR4!^R%uSoSK
zM1<$}%_w%M{{DgOA}9V^pK0U8&F4aCl|HR`zV*h+UHErO+%9R_Et*CfDXLqdZmfAO
zu*xH0P1Wn=xN1XL|CKlekCnSZ`u2{wNFpELpI6$YIxW^c#uFLf(7Pj2M|y>CO2jrc
z->uA!Ruk#|o(ik0UO)ZYx;f2rM9*HQC`h^B^W?W6I=EJjOS;QR<@&RaHLXzdsISFy
z-H7h(+w*Hj5Sk?6mo9e87>p)_053=FK0U!K7+%Q{snZS7`acE3sI&QxRZ2+_pcQru
z0hKP_1p%~vc>3m?^5bDwB+-9;K62&UU}6?du+M0H){5E$iF-c-4keQ4x3*xCo%#!F
z9dWF5-m9|bz5;D>gh<=m$2C>bXT~SxK*@UHipH6)RG}(k4fW6Y&h`ZH1!L3z4%!3;
z{EQZavE?^ET!|9QITb2yjgmV(@E|>^QPKPi|Eo`t9HZHlr$)w%$6uONgyk#_Cayhx
z{P>HDZ*IPrvaql?C?q8GH3xYB4`24TK!U2WLG2y3^Vm#mN<_+;xR2r4-xZlCbphs+
zA0;jEda2a6#!L_Fiy1VEKuqW)VLleEAnf{@v%nkT(GD)}<_s>El_`RJAqF~~j}NYk
zfxK#^m#DOJ1c9sMO!UZM>$R>2;(;X!MIp4b;`;DskZINUQR=JJTj%z%i0Rj>BoFkB
z^kc_DwCU+d5)ue47z!^|oN#w9oI7i@*HQV-owe2x%3&qdHD?|Ad-Q_(4&J7XT9*7T
z;aAjxpiN+%b3&p+fYs~4Ywp7ubBd1{`^1K4+1ZC*nKx|){nQHsMsz@z{zP3NhnGcR
zL_~7bOSuSpcML8C^(rq5wR&D15fyG@vF?=WHTaUmq$J*)oSc#j(hc1}4;$`bWEI-O
zX_UkW^!fOY>lqirN;xV^Z*Ai9rc&b^e!P(#o$f-+u$-b^EMoUMg#2=<u!9}Qmw&v|
zaQ2&N>eM=(k2(%dHXL8J9V_wLP5Eto%<C#+;1j_DcL>!d-jt|WEQk;=HiQz`DJ=IW
z8+;EBJfD1O=aKl}-wy6i7b=gAlFN#HoMRFajI=MPs-E1Z+j;??zq~y%pm*4RD_LQC
z^6c!_RntQk;+y^Bt+|cecMLqEoXoBYF*AMqv~T-zQ*=cwsvaFU+}T@wOq~9Pj;Mzf
zU)}-U1GE`K)q5H&l<GHn5iD>b*45jVRW;){wjjzgeo{f{YP6O)dGZOnH!1h@pA=!e
zS1!Bn&(JEHd>8z<q!-FAuehv8b|jSv^8qE{gESi`Y8llQ;qvxztZfa4-)<LtppWv(
zGt*bLam}b6Nvpq17OW@!O^J^z!V^fJHkEpOG6VX2kHpItj1!5=lUhDdrq?$-yGqxk
z%5JEfe9v{yVi3xAqktoqU*nHeWewgdT~EHQD@}fz;C5-{OAr}UH#IS{v$I>FRfdw&
zC#HfQw>441+IQjAzV?|4`-j=bC{eq~=R#;*1OHn7AgNS0G%&(?#23YYlA73)r76rv
zRzT1zrQLMo7Zk#YfkX6723m?>LyNfEMw?$Hev0nrs^V$-m&?8-%ZwD+$lmht@gaUr
zPQI)!Hx<zXFKmzVw}Z5(8Q!^b=hB;-D-1?jnwu|~n3%K*gU@u0<P)PLIdtpRaN7yy
z-g%TMJudC(+vZrur7C*)%WWoIPblY5hYUbrq!@Pn+t6rMg`fppDdpYQ?_q^=c7oGe
z-O-_U!!+~*$}O!yd%qP>REAJfO@H;O1iFLLb`CL{-6&g@>eS-7;e~t!%xZDniA0TJ
zStcENT3Xr{h<>daZSC#lrz$4{2me^%5AxjNQVR<^BodAducOpg(EA7d>v{?n^atKl
zHr5?;r0PkYY@cpk77e<?sa(awKYIMw0Q@&}p?jp6W~e}Fz%P)MeNMlk3Y<xTlar)p
zNP_|hpumb4KMVbH4WLD4NI^@U>j!QdZ2q^T3RJ9Zb7DbdMW9SFlH4a7{W7B7lKK2!
zsX<aG|K)P!KVK?+zI%+$hf!8Ym_c`JMiIuBS03@|P^v_K&Uj*5jFDeAvTD`35iZWO
zrVsTiD0^^AMzytLmk5k(@`RmYdYiK0{->CLxC~lpVI1*kS1gsyR(Y6mlj~Yot7`-b
zl4E15Hc~QhYJb@E*zbKeCnKNld&9M<*{H!#>KH+cG<xu9?MF(T2yuh>zfqH=MD_pU
zQ^|0Vnk<D9A6O}GV=vDpq2(YYs_DQ%ckk1=6h4+<&^4kGC&u%(q7)^eWpWoKf2`rj
z-k2*xDyB;do8)*;>JaGRx=Fe8bYDxeD|^z9?_Aml$}e+W&-T2%j%%sdX1r|20|PiQ
zMfDrKq!Fxt_m4pi<-b>6{^#-fuf3oz%m1k;tztn&RXXE+4_8hum;bi?c^qu}@&Y{H
zU*<)>yL!|BURRF)+w-sTI75SfII>if_CuI79lBJMW(lHf_#ycBMd<^I<d^0moWVYt
zSqC_S_T+uP&iD(>fe2V1p8k!DYS<&fh50|8pA>1GvK}s^VkV}0^S+NgNbz*41>{Ph
zc<)>~K17|;f4E)y{f8tc=i_8;ojN8Gl(+TvIUYO`Am)@_wXv93fZ29c%hxvRlbAm0
zYI<gT#ycf4aBaOqekNC#I~q_`y?cr5+d20>$$h&{w~k59P~di>b_LBb?{Ypiaueiq
zDF5#a<v2ICK4u_Tupy>{rn&sr5vZoxte{&XI49Bej)qpc#L=i@Ls@g0^GF#qegwTk
z{8wNCv=4r~B@mDh&2CppE>g-%*lrjS2_xkojcXu6JbC;Ndy*(9OIu)&8Gv%QI%%7j
z_$$RV5MUv|>o=e0UcGv?m4)TewKj<<eb*CS6v~=Q3NlC4TDJsiUsrg3A(L+-tOF`w
zxAkm0*`2+4Q#=WRa^5(cVF{W9PuvQ6cI7NSgk69PJv||rBSA~{$!>bL>319@Q0nXT
z35Y6WbOfUtrCQFfi6TNmPg9eQ)l)mvxe_Oxqf^B*J|fjW5$p487XhG`_$>U3Kl$nC
zUckl^R1oCAT?(lj+qU_WHq{X@;ZmixgYQ<<SV*1dWGrX4Qcet-JKNG)*#E6SjbrNR
ztI)Srf#e+lOdISdlmhtANJ`hpjIA08c#QxF`!Jjy&i+A@Q58M?_tX-+ZEXOif9=`N
zHD7Qxr0B3lP8abInn^WSC_Ul!x~&+q2YjrcrqD(400Lvcrrp&H3jFsMsfW;ONF>7`
z2V5ARv#iBk5o$bBv9Ge->&sjdB!?9&NYY5zOi~)G`rgemqXF)0mX75YI)zgmPvv#*
zxAX!#TiJ0)8B`!L;sjdfH@NNBWh4Jm3TU&{C*ejJjIp*;U+k;lJ-2Gxg3b^S+E9R*
zw>~MPAjPS7ATfxVBDtn&CnO1(MeyY^0Mrby+V#Vt3H6&Ym75QEtH8NNQMrK2q%VsB
z=!0+_X!s0hGs{d8%6Fop?<_?Bhc1G`Zvv#mkofALQcr}8%n{frjN+z`@WIz{ty-B)
z-$3H6f)p`?&A}8MP7s-Vz|a%n<RsDg^j-Jqy9%juC5H;=0;H%8c@xQ&`;8qp7>LQg
zjC4qazJv+%_q8W!6^+HFxTAl0B{}_OQ^`r)pMn4Wo0mRM@D5K;QjQ1!liua6FciT7
zQ-zBQqfG9jOx=8``bs?}lUq`+0P;g6pd?6H@_d6BYnjb(8Nt=Sytp`z%7;^SIoCbh
z^dGOvt*K;qeI5kE{v7jPR&8l^c6Kh!uQU|6S7wM3E<`1SIt0x2!E?E1OMg&m{+(2M
zp%X)O<IyX<Bc{O=*Odk6UZOmczm7le09-*{I27~h(!uCK6TGMuTT&d+yRD0iWS77A
zuqO&aMPSNyh5-=lW;<5%&%p~%eq3_$e(>pUJHU4C62gOSz6|}a;fSSoPxxDK&S0RU
zlfuio1X;`rl=zm`Uir>(+YD(vjqq`MTk~H7r(>jMW9}i=3qXa!*BnyXZ_sQV>7Iz3
zYweLk$wbq?hUEiED}N?x<eP%xzm3@ubay$Dmfa(WPge@^k@EK=^A#iHD}2oXl734}
zN=cMSz!*+_lPR1{Dir%k#>43OZ}hV{{ZXk&r)P6#jW*hNx%3|A@J7DEuRHdru_ze*
zU8bw<N}|^+*!YcI5+jbvx0B4+c(#2i_>*!{>VFx<IA<A|`AN$~yCXB-f|j;PSkaw%
zvH5X@#1lJ=*I#pu^g!Hx<3Bw|l-{AFaS3~>fja~}+h^zIxDZZZPbf*5^Xr)QBf+Mp
zClWkakec#_ZBv)FR}_Z!s!6;a?;H6tzT70_4lsXlFp~r)8>3}`SVku&TfDU@;rY-r
zHVWqvaR6^q#ZdZa#+fd=j3OH{%y-O7Bs3)Pet7(m`;9twO-4w1)#4H<-T-3xH%IBS
zfUO()qBPVJ(!5gF)cANF$@#CFB7gSS<ABi6bu`*@a{2DH3+!RK*0Jk6v<}Rkuoi&3
zjUx(D00l#TAcbHB9Or|%6tZP?9g41D5p*WIZzjkY{hV<*^7B2H)lf!b2^8IW@aytE
z>As3ah`t!r=LS)i>1*?i?hB2-ju;7qPQR3zdpX^&QOJV*ElJvzQXErAyNty<4tInz
zK(ueF0@@4*K*byoE0~2Q(d?F7VE5w_gihjYA18$p1#8>?F=6h15)Ay`i4Ok<(a8V5
kf&c#o{`YU-=HjN$7uH>0oNcNGADf~eds-&t*u|Uw3pPyB=l}o!

literal 0
HcmV?d00001

diff --git a/examples_skyline/accuracy_vs_fnr-skyline_formula.png b/examples_skyline/accuracy_vs_fnr-skyline_formula.png
new file mode 100644
index 0000000000000000000000000000000000000000..3738ce64a7a6acf22a568d500d7250fbe815d2d7
GIT binary patch
literal 43356
zcmeFZbyQVr_cpu`0SmBDX@d|@Q0YcdbkkiDD$+>jCQU3rHzi0Z-AXqo3Ift4NQ$&H
zTN=K3bIy65^FGfnzVDCk9q(_9_l$84N8D@eweFbnx~_T6`-!sRRmweddoT>6#9h0r
zieWpNFpMl_7diaKrNyxoe(ZF(gj3%I|9I@W{TTkd+xD7{1BUHAi2fyM>GfKNUkW*1
z(RNg`F>!RcX>W|(x#?(YW#edNe(SiivAu)2jWsX#1@7}_j+;3;+KTY-{PXX*ZR|~X
zV&_G=G3+>oyL?gIHR4yVyS9dL-_~5C%>FC;FDnccEIg3i!SpUiUZMVxiriJPm&OWn
z%ksX-<F}h%F`nV7OZX*t#YkZMNW{FAMLaG{=c`b6EFqrt_}D95&TC)BH#a^MSaK!p
zy<0!HQN+!oy;QcOYxuh=leyD?;`RBeUGVf3WOKGIB=A2{o|k;ktDiY8MG8NT?2kde
z-Mb401@j@>X$C(kcI@DTA2&!y=-|gYAB+-yXiNPcahWn{$l2=&O-;%xtE(2@^X~BH
zMdsx2J$&#$+pF(6Ew5v%f*_8Nm7P62Ihoqh(lWu_INtO|x@aWa^5^Ji;@V7kc~zdn
zi3_~E#@Su!s;WOaM6fo)sA+BtqkBui<SOMb+R#oZ*fTJ#+Q4<@jQq;V3V)H`&!4w*
z&0A&I6(e5Mo1`}1A(2urqUQ4vsXi8)m36@-)J`QUE9)>Vtxo)?z$fuo9UYyCzEbz#
zj}Fov;#ivv3~F#Vzh#Js^RLsAl9Em{efo7FXB{MdkG!@?xt?#`P`$Y^<1I({J=33g
z-l!_+&i6cv-lE*;qDgLChVQ;>rrysb%$km~va*h+q-GET0s_jmHr?x^CHo8gs2D}Y
z`~>@KlWjgfrFr_~i3;4g{k_rE^y~SGGi7TT_@(Ko3p4Sd1|p+T13cyHs#D+dU4$Rp
z?8-8J`t&J*B|b4RRViA+qAM%DJJ&K%a${U6IAgLcS*@Y+#M{_dbSoU8*ShmpS2hg|
z4Xd}ew;!{l!NZ3S30}hch6+cc)Yg7aIZdYLN9FeMTXrP5FHO%cEU4MqzB6fz(jWY7
z=r&A)Uih`Z9o1szDLt=*c)3tc5s}rjQT0WG^_52N4h=1RRu~q2mtH3k`=5^;Hjf&5
zyxH7X<2rs^s=2vYR!%N;k@zBxLoH!!#(S$c%6rodP5#CzQN(E?Xu5RCfQ*cc<Miok
zOMUK{ijg83<C8Fo`#x7!PbR#!CPYU^;}^%?*pzQ*HcG6eaToqP`0&xAiArioN9$JW
zd`q}b!!WP+me-U=xhZCQgcJ*2u9n=qY46~mU~Zm_Q&8ZSkdPRuxJx?1ytVARd(WN=
z<KyEAX=z_!Osa32n1mAu2M^KE7^5c%jE=VP+FUWsHg9D!dVW4Ya<c$;?V4?ddw-dy
z41t52TcGL~ukC=;Av!t>=jk5vr_Y`#sHnWuDaGe@6}x0|aB=aPo0})NEsU5<cBEGw
zGb&r3mut;3mRlI~We#s?Q4I_XG%DL%wN%sAHq*_uFuri%Ld18N_wzHQOMSxL9k3o=
zQA@5j6^c3k(q8zL-AHHjyl}pjt6p@b1RjWuho{l1D{HtUGm<6Sw2^7l-qp3>iJu>T
zVPWCOSaWQGWoO1nruRnb`Knib^n&4R3gNGOnebl@A2||!@7}%aj?OYq_l5DqXt?dl
zTs3oKW@VKhy#h=(T)-aX`f>4BSL^z0B{e*Hk<+9}N1D!%UNtOGGguCHp4)j!a^iVJ
z9LH{yt&GsCDJ$O~r(!T;@>;$L&n7G^%y#ZvQ_0rmT0%}vo5%8?FZy#~eQl3@-B~sc
zj+(;tQOT_?J-g2|Be1DQe3`w@!R+w{i2v4(&#WE_khCr-Ev<*|mAA8dA13CO_5S<R
z2t2I7l`B`+goV4lO*w~luEhpO*mFluj*55A&|>Nj#G5cq8fNCqn3&^kuC8Y5%X1Sw
zh1o5wt;GED4FziP#TU1Xj38RTo?w6I>&qoBu1`V5(6pDQB<9Fj-6WlClcLp$R32A%
zwFG&U+qdJy#Knh)hvWM<f9Fqpdmn2h@EYDRX)>c&5%vIqrT(>`Qhu-V>qnIIY}C}$
zWE2#sFy`~We|Id;4dvPnUc$9x=obftg(V+2B@^NB{SdpNx~%L|77iXBCCiSq=1)%!
zvEmF247y>t_%jHfG&D5ScN-4Ya1(ZF6kj?G+u-iqyL7_#ad|fV(;i<dE2Wjh<{rSf
zw3oOoWJgWHqdkAVFH*?vQo3%gYNkPHL;$VPB5Zsdv)A$)0qdTmygT1z3D$+fM@A=S
zVN56PQ%f|(;R3mnI<DK4YUmd^sw|Au^IP`_nWpbO%B`-a7xClAkDr5sZ*EpTN`_dX
zn{DD3-}!OS)@*fgyyz!Et}XA*b@9#RA>DU3J_s8IkdTs_Ezb_lFD|Nkc$82vOQe~%
zCe+l`$sJ;oS2i(8@Mjb?Nv6V$u8WiUJcab6J`w5Cry?)^e8-L*hxhM~i4=92`10~p
zce!`@Lem4o@9^_uW##a=PVPdWw6wH-NN0wADpzKSDh$M)UAvkf^hWp|RM)((&ctx^
ztS*bu^CM@pW9){iYQBG8i})VJXHE#=)=yDOR5TxNiJxds4TamW2Ky3G$j~o#*08cl
z)66u`kB`puT>ITH<LTj{r`TipJue;Z%!b2{gqi-FoaBOuP=Me=$hlo}FkL5GW^izj
zPRKSU&#LQ($Mk!n>Y#{-w5v>Rmp3;z8|GeH7vWMh(_3l+*@*$(>(Zp;`<fs~<5hA^
zjE%>8UHY9i6?g93IhNfh(fZ*&#o5v$u;y&b$_R&<m@-HysMMG75gi%&qN7W&p0r?L
zwZ-8Mjb_DhM(%UV=;y_fc9GCYJ%p^}Sk35SHr-Pwf?rI6fla96Dc7qyh5}&mOquAU
z@S8VpY8x6dX5%u^VvdQaC|r-;S|3dlE|>W53zqz9?~GR>N{qf#j4kjMhmRg*H+t@H
zou>UbPRqb8K3w2Vio$CF^X9i#!o}R?-MsszdyBcm#PncMx4*seOz(Fa#5o-f4vtlb
zV2cnLEPs7_FXA@;)2-9|bEfz2tIVG2kY#Hd8#^a6%eiq+{ruXW(r_(8d>*T!w`nV|
z8;V%Co@)``o~ktoE3kc}{&nQghYuenIy0luypioaqSrKOD*<l;Pml#KIkN6B{}TsK
zRo{Y#pcNDxJgnEy(4eWT{K8@7R%V;siE^8k74JqY<|L$M8hDsE*rf}$J}V2OmZJ@k
zgt+fE<@yjFJ37*@8$WMGxjZ;HzRYuNqD79UJ;%J&5$;cO<cJ8VJ}gU#jYTzMQ`1S<
zzxdx>rj3OgZ{MDb6mgP;JRY2wm<aPUX;1hDxeMiNLwW%#WgVR`NH4kWOV*=9g8d%Z
zDAl{Ux#`c|WC>(bXffQJKkB(Up7=B%z{Q_OuRz(CnmNSR*B3SbOEB{-<?E2Tf7%K;
zt}olcRmsM6A*Oe{n_@273o+2~nmmWt6}f#|lq#pby_YtsdR{oDt2ooXIdGp^T~{~d
z<Ht+!@$nq&?A0xJHpPgf#Aq*F_>PX3r>-vk3{=v@6AU#zefo5opI_ti=g+I1<(u&+
zxqfsTUHfZOqD)@|p}IZOP%_uDQ&H6UR};$Q$?M<7;^J5o!ufDsUS36WRrEv%A}P2)
zHb_^&zo&as-@Q8rc+G6?$7ezugbZ_txbQxzGkv9v?($iXt_=V)K=M(90M!CZh|i+!
z^ym<5z4rB$1*d+`rJDYJL)Zh3U9rM$^S7sZ3R_iAzZSB4E3xwPkiPp=u0?BtpMytl
zk<-)X&x0~EGdCfhbMo`&ejEz2Y`I6a_jdmGc1WqEo2!#BVy5-90#>j3@RMi#7=&Bg
zG=iFAuX1s5$;Dm`2nh&iywZ7qjxL3UOLHvDG)e^!Ck_1ZsDbN;d1DG@_h09U<E>X!
zR8%IKV}0?&iTG+J_g`-(MW>Pzqgq`0Jrgz4b;lu3W-fl4S=fjok-B2_s4*Gilakz8
z&qh#WW37Kn6V_a>zW;TnU*A-0O6DFtqGFtwR>fj2-T;tEK915cq#$}vSA(qEPhh2H
zl>XLr3EhEFlIPEpHw{2mqczI$Sg{YC`|{;lXr-8<`;ebt(2Ey0%;~9orMDI&w;VcL
zA<sKcbsMJ-&|Ls5X?**(*@t^%x%OZ4GBa%+4x!!ZLqcldv4$yM7kYr0!8;#mA^7n>
zhZ_8k;4lC8A^&Ic&Ohd@?K*(GW}^V{1jsHNCr^It9magFZ;S|*H}`LCy10a@uy%EI
z%@c_lz#NL@NVp*l6Z>FeE!GS67Zg9I%eNbnp=1)%1`v?-eQGqa{AMM&(^R(xFbp~|
z*LRSxEy}K8*4fTey1f8;kLLQ*@|uhRjY&y(%4wSaJ=bdkKWeY7{R&7{JAg8}egS31
zT}@_9va+&0`j>*$)m@$)IqUfSPX74AqXt4V3jqwmZ-EVsp?CyrBuT<^6+v8n6;2;r
zRl_yfru_$ERBWui=P8upiHIvz1u$nqdcdjb>V{vsbSYxtmveEH`?N4X=;SNUj&O2t
zH2)hvb7)}ZGkXD%?>up)->|H#4B=0ICh?R94|e8Sb*a98|GxI?SI35}^@T<ndU`Hk
zk)NxoCKehcLs`xmlxTadF4`r#HAD#0@mmD3j05bz58R~y&XE8i%Y33Od39wpTHou@
zXq3Am@Yl}t{62g~V}HJlp+c0H4x@;p(2QlS<GAV-85vFq34{7@{#MWe7TdK9&6lQo
zkN&<znslh)mua-;?=E5YE<omjqw|1e<E?v(k~LB_=_EV~>JpWKp0>QZS$X)_u_XA~
zaR~R#KqVHx@;0Jeg(lU}aZ%V()ONr}#*Z$(qC(1Pq77$Lwvw>$9%ji5a1EWLSIPSN
zdRwX%k0-HRix3C+j!RVZ`l+k_C=sE2Y;0muYmzOz6=+-L6aYyMrGK~t)PrR2jY-3!
zL=n;u+88DwptigB?&akvT_kXDa$Z1Oer|D*4~N5<3|0BD2?>1@78DcfxqS8N>tDaD
z5S6mGx0k<hBgSoOV+j%06~I~w@aGl4UC+-~?PCyXa{TqJ`B!goCxfW-SI5bYddKOW
zE|10Ls|f`K-{hmjy5z$yd_HFM6%wF=l$4L#Xrv1wQaU+iFL&?Sb?&M^!+Cb4C_b3q
z5dcU|9W9AU%z*^Lt>@=Q`rK#u;U)R51~Bo1Omgm&%=2@t2@1Sb$FgC?w*oj_Xp*K5
zgM|f8syuEv*L;;ZY<70mq0C^Z#~~n0#3`jt8_1Iz&{*#ED-;9H<<q7|Tk85e78HOQ
z)`jv2GKxC$(bCeQXB8v1sU;YedjZK*j1aonoo{1<=p`cF4nIEaUif^}@R-qaxAnPd
zg-e${lC=xmsK6G=x7Nx>Mx(t2Acynb>CQP1K+HCyy0x{nWNo@QAt~ug$?V6yRo3~h
z0ltkk65$DjJ9BONB~H}<=rwFADg|x{0+J1EwXc8yztU*w*}doCzp>jrMuo>??(<Pj
zZtf0X;aogCJbsL#ny~7F8_HqL9tOk!sMT`t<3r&4YCxF8#KiclzMVtZ0XUsqLpibI
zI+9ve*48Ek%UAqBDN1@H?OWU0sx~(_*PVCgHOu(MQoj}e1%7}$Zg{-Kk55k~x^sa;
z=&~|1Gxyq7Qp)0RFIlF#vac4q%ycBF#NK|cVsD={@--|LmMTE(ad;yQ3yYL&a^Rif
zbDs_Y>C^-&$1_0$^c`AWUS9VK{xd&6&uvkPS@e&4IiD!~mD6eHtsfvZ{}}cHo05`}
z`07|}cbR7~gsL1o(E<j@V5#4`98n=8;s7@fu|U*Jz@{$^?N`{0Y5<b0%Qu`$7aBwx
zgiGC*I=;O(LL3>XO`K=Wv_U{oIXhVfG@9n<(L~T3%;taAG?adqx^qG*)<6h0+`aqv
zfr3XE0jUaZpr)jB@`?sT4ggDwgoLC)5_WbN6_eOlMC)IiNYYYRQnFvCI_x_oHMN$t
zbq0%hOB}8ND>Z9A+3PywH$OMmJXq<gp{yJO@Ltri0KlNr^6bq~JdAd#Nqsnu5al*P
z3qn(EXQvi$!_wKK6XhPBo=&j9Lf-=`|I*mlIA4SnbpAO`X=!IR<^sGO8}{zo7j#|*
z1Qj$4Fq5t-{ER}j6@|P5IfhFg4Li>U&m7;idw26)Qi>GgUZlr)%?u62!rt&da5eoT
zym7%|i(K&Q*IYof&WMT@x2%*?b^k~7#-8!w#fzr;^(P_303+Il)8n&)m1urcRaN=S
zTUc<ah#Z!<IZSou*6CEv4OK^eSpOQ%?765iJUYrHDQTElzF9OZa_ZD6{oh|A#N1Q?
zU`L)jb?R*9ULW_!X-ZAJj+t3fpX(51Xdwcvt4lMPAisJ-lo{>mSlLNNHeSu_?YN>o
z*;%U0*&pbki^z?={o%uh9s8#Az7&zToPt*x1GWar2(gwcGpV3JTP~Da1CL)N6JLVd
z){>%;y7+;DIVmSc9h5~7{L~Z`{f9+BR#=?M@3#PMiPFxG&(DHizI+3+$TdPqNy%Z*
zY$%vLlqENpW09DW>@icv=+g5BHnWoC`s{rc4naXJW{=qqWH!ZVY4UJm*q<TiWzhd{
za;Df$gWQ$tFsgV~C+jl&4nJSR4hm;Hh%dUEVI*zv;z;r#f`n!Pf&w9Ks5)R0h<PxJ
z-?8(TD5#m*fJ&#q!a%}|q1Q@$+`Q5d=*k+Jn$8Wnm!zfJ%e}qD=f9jn3w&#%f6KhD
z#Em<&;(L)(Dx$3*8kVll*CF92Sj%9%N74Ai=A{nG0(w9UmmqYad{(;g`}>}K``X~m
z`AxpQ+Kab_snAT*eq$QtCIjqJ1O8~>I{0v5lT_s;a8*m<L^~b=QG1b-r8p#8U62*K
zbIgY#cwr@^8<rbDFag=c(O(jRE~M4KjFNd-l%S=3sd61RCX^&M)RCS55eg)95R4p<
zDBDmCi(}YU2&gAj7Kj_5L!+#@I$y`5@1IYuyrn&E3}mST7<=08Lu^SvM~mRiask8O
z5TcM+9Xkb7P8L==3ShaA5OAuX{KU;0bf#)$I&?q|dHeONJgB(li(}3BA5Yl%9LJP_
z8TZ)UKf+^>1~JYY$ck=}V=~M;3l2r2D6ds>vYk84e*fwU>7*=pUBAA{YlmUVJo*mx
z{6#QsP7C#d#l?r~U!9i*rrQ4T;oe^TJ<6xPky2t(;##onXUew>LiL@)0H)EA?QkT6
z@k@OCco)j)*(UWOlXI}4DQPcU269KoXQmwNA5Vt4#CG!ULyHSiPA2T%iHX$Aj1H&h
zGP<TPZ-G1A>gnG}Cg`aKg`^<?7dei<4Uky9cH+bd5H&PF#4B={nT|LF>lW7VSjctj
zX4pAtNOXX=vrePK(%y8ENVUJGSltH-GnyYzN@D<OT6}!4Tlf8~hjkGk-64ichyzh5
znka;|IW3A{Ng_KIX+clW=iX#Gc<_h)oHOpJzkf$dyc{bI$^mKy22lXUbW@#nRD6y>
z-{2PIy{CcC{kK)GQ>Q+DZ7ScXp1<?7kdQX8c412c$Qp#eKvo<8B2{f|?K`1Q9Zqv`
z1^QBK`KHNDRJz+Qj;SyT+kfgcZrg*`uT){)wB{FZ+7LIFFU{cG)nvMhipsgrT~$%B
zLB~aM&z?Q@)4??$?+{q!K)$o>C^~rPQ1Gi)r*TgoKYr&}Z`f7>6&y-BzAwGT16pu<
z1S?!ZmCuej`N?@M7q@?FBQC|gSu6@yb=jI6VuQ5WooyQMd+K{4B<lEt1dFyL6>bO6
zLVuLlZh&Bx`rW#?o-uxhRB-bo7Bc}4G`qOi)>G)f@3Cy_G})m5_(=`0MuJjwMlhF_
zEa#S?(dEkz&~(8*0Hvy?UW3$!e|uoLjkM&JwX>&_Cz+%9Wr%~~<E@DSo5W1-!s6#5
z&kVE#+kMNg`zh#qD~&mgR-JG1Ap^anwx&kq=FK<7a^^4AH#UgGPQ&8jeK=KERfi8A
zJPGIk$a|XzUcWw-Ys23_i+`tC#l46K21wC0*7<cN*@eXs<m?btKtGx0maaO+D2`VL
zalD6)*6(0x#z^}^udA7<BK~*D&zwGUMjh$|D-`i}v%9RIXvFW{-(=Q2Nk@U#Z$A)2
z@@$zR-fqBC#;CggZq?nTLW^BZQsPG0U5*gFMMF&+qtXGmY8y0yXm<F}q1acij**d*
zPh{0`qm&IYgsxxv=E<{X6#&k?U_sHS2y|?2d}M2)Qs*jhlFQR$Wq!E!%a?+I6>NeE
z-^(-R<V0O9MM`|0pP!GS91=VMYbDR)Q4IznfE*z~KI|>5e*~A?HB8c7A?`M=EYA<u
z@!$Hi3ra;B_eMyQM8}-UV>;W#f88awcu{$xYl`4|bw$f}=tn2)jO0(BF4u?gj;2*Z
zymgw&?SyzPEB$DninDWG+`Kp6bm1ryA|gNxn%rkfwZ>N|DCvHdI5va41=<g5@C*)K
z`*}wBCf*6ONhI+v2qX0yUMbO_6wd)10w#kH2a6>clo2EeFZer;UT~S#)Aw2#M$wH3
z(NDBG8b85t<?`h*2m>7;%=5#R8f|o1H4PKE6Aqg_BPr>$_-qB~D)63M`(at}<$?Q9
zt7!(@GJ0VPbk;<aO~=N}h*RISK*<|Tr<VFW=G1}jmE$Rm<H{(wXP_1Eo`Npv1eI^l
ztJ5@xgLq7hgL&jsFMa%+zSbZ*qpq$VBxe*F2U_m+{2u#fNVP8oELjA|3n@al7KjtX
zF4JhlgMeJA2mh|H>Mzp=2pkAAbn@g$$E9AER46y$L2g9kGS{s664VhQ4DjQu;)_kv
zMMM6giy&a7LZ$<Dvo73O>NKf&7BoqihznI?Pq{LhpfskTs`{2yE@Uh<vrH8z15&~O
zRuzjaL6X(YvkG&b={x(nKEbHkp989Tu`+%swl-=|_(E*+^RuH|0s@*q8<YW!o-Jp>
z#+)?9ofbwLJ0U)v0bL!hFaDXyEanjl>&85FtvMOl@X*mzlx9{&UL!<N@Z6>=>k25R
z1QtXH5gJ^0Pj6GA!o<V`mDpk^imAs>Q9&>V-fhv57KY{q!n-HrN?5%}w~U9izX4TP
zEltf3c-Zth?qYc;G=$WfKt;LNvBAk3q+Po6w;t%`n4N$}@}uRI0c6zzxZbIfQV^Aa
zfReXDU7`^`Js$40wbpM5)dYZ<!lQYp*rksjkB2l6tiJiu&HT_`UI08EpCw6>v<Dwa
zG!Voun!%f1^&z+T;w`CCGGLdAN{Xmp1hlIpW~HyMuVJRqiBuI}I0(|X7M)N5D1wIb
znZE@zL2HC4#T_)e`8HZIGLPWVOoy))nnguL$to#z0_M&Iz^?DTLCh`kj5#0M+0KO5
z@4X0LcTM|Mu3hR5Qc_ga2_^3X<N|S3mcT&eOTO>e4YyL~VtuGn2gf-uNxF6eTuFw9
zvIN1F5Esa<Gy!B#10Yf`)U1G}DiVSS1p5Z!zGLU4Ajg(L0zh*4+3R`osLFxH8qh=X
zyLZa4#{i)absFm-I{-38;9uBb1T7*03;d;qYl$HQW&rj{-kS^2gjyRjWmQ%CzG#|?
zF4!gRy%03Uornx1ut9iDWlg<`MuyZR4D_eVx4ck(DqM$sX|cJ!0`fA_eHTN^*FVpT
zFuM+rY(NOtXp9oq@6Zo}de;QNOtgU!0QA~eMY{zp6Uc(Z@BN#a@Q^L6p?W{&OY-}@
zu0uX=kPOwce{%5P!8Xt!5a&UNqdnVH31khFj?fp{crDNq$C{;K$#GBC*44E>K5*(V
z+#lXf7qNO4bI9uW!G}lN;JQ&sYeHL&xf)VKAUfnj>75^D)0zcSKTz6$&wPEs_$fIx
zlYC}s<35Wkq1^gC`a2rl0g|d~toicg9iR8$C8Lj(e($?#%enQ;+yeb-pB!T6WN56b
z0}76+A+5zOHrWu*bdwX`E&)O8%gvfveF_OoXUYXqtCmVRtOsTW993J)g9pzf_cE&9
zy7d-{%j(+NN!P6lufP!4yWlXdhXz{WV@^a0C@)3k1DA3M#V~$IV&1m<{!7mY@e8O5
zG=Jn6#2JCEGz`Ce`_`gY6H0#E<?Hs_ij&OqV|LAx9vj3?A_{M-%Yc3sLHseC^}s0A
zq-UI<I*x!RkE_H4Q^rS<wo_nq=m0FhisLo;`T{f-CYW78EvLZ*#YlWZwcwrZH~@C3
z-dmeT`Ix0D+nU_<LD&Kq2@tbsZ+JkFuXvLbrTs74FPmBVOiHfkhd-mgMSoQghnk+X
z^dwaFl1na9)A3z{;I+NXU%WVO^qj>U@-C|Hfx#jjh~(tJhx-k45X!Gyy*ioOX&^e2
zr>3gf7JHQ%WehyjvLlBex~aj`p*7QMSIvlWDNFE{v=C$sTniK|-=fkPq_RKjpr|AP
zY3}Oz^XCD?2iC0vbG8_3dIQ)F%8{jd(|KeS^$b)s)}dlzwdEoJKqj0JnKN*o;)gg=
zg(9Yqvrte0rwZ9;GD2M1kdCZg@M)avn2M+i!#5BWK<0{vvbVldyGD13TXsglk3GGf
zOT7qcA{LASGY1FGrj!_BT}srfj}+DF)N}A-0#4<Su?A5ey;d%ygItSt*-m|z$PgaG
z3{)jUVq^K%h7IiD@ovjZq?_q9MuP(eCW_t0#s;JudEo6uQ+eOpfU1l}B~a1}2#vC{
zu$aM{(uw19_UzpYL`t)i=ZH^=olxAT%gHe%og7@L9y%$x#i?!i27*o#>X0!3t>FL@
zNt~1-<!3RMCh&?yYbxu-#8KjMkHmm8CH2LN{Ww(sGk_Fm{-_;We0z735C>KY6p?|R
zzd~hE%Zx%90v|{#>H5WaElyht(OYdX7w@CWHEvj8t5r!dn1zx-NEu4Xgt+BpS5E-V
zrs9$?S-W@dMnx_sD7i!}jDD%BQv<O<CV=TI4n>gQQ1A-L%C54a=Teq1kRu5<@a>jC
zG57bJmE4fVFr?8QX%$V)5&L_7kdhbHK>kRIi@RKP4Ae{ufbZOs!eHgd?QoJn(G(RM
z5t-sOZD@lBw{Tj+QgF_!;^|p)oNqhda1S1XS%0wNt3#K`S=^feyF-?snW2{qT-u=_
z;PK-oSgVm+NuX!}u+fL|xH2ex5^KMBafBhU>T@tbTG1Z!PXJB7g&~TMi!&RjxQlAC
z?9m=?qrFyf!2WcB1?n`O<>G1umkz({><w|ClOkU01&{$jR7ix~l|8byvckc_av7An
zP!?1SB{Q9#+cCg&Gk7tlwkq|pfp5_RjI!-Tn6yc4bzQab*Xp{fyuM%4I;MJHYpg;M
z@-QP~IwXxG5S>RO`XM%gwC0Xh91Z*hWb4uRp^?VuOz^d+!47E%v{n56?&f_YS$o4|
zgBR%x657E)G|^j}Z(T4*4pxgvB!Phy2Nt{*!rJ}){l@xG2`w5=c#ZZd@mH4`(j%eJ
z>IgB&5?oRZhb+DB06)VI2Wz454X->YxwWFiUxLTDyT67J0xUBF1C&-=`7N%MTL1j=
z5>gsJtVrG@8n8;(x$G7~4w<u(?59puwyZzg%5QqmxLYufR^X5Xd(WBm@`oY$I?N!3
z(HcER_8=sjf;dtdGYS@sw`_G;X@7)k>!i3%0K^KKYmw~+rboE>*u;8{y-r`ByoB#v
zX^RI&knL|2bC)it>C~GJ#bar5pclL$PIf}IRT;KsubWZ?iE<+=OI!Eu`%b${Gv;8&
zBsnK$n(o0UY2lAQj{W8QyYW(Z=)oaA`I1JnU(2FRLbDQDlq7WTVgTWQiQqJ4;$7aV
zF@SYFpD@28r5ScHof?b)`Hwt!@ZdB%`&B3-JG*dDs@}YLGt&GgsDBQ9UHi<e#+>*w
zGZsaGmVl#UFRT#9F5^%jQRG}eFT~w{JG#(H#R9EyTP$An`@e<9|4n{Nx@3R^P{+Ci
zJG3QDC%XYJ22$dE3TjQjMhFbSHY#`{izT^(KeTCAa7f4mq~$hn3L+&gDCjWMkqf@M
z0e3*i5fmwf<mBYlO6JXh1PhWF>l0F>Qm2BVqB6jhmkiw60+=A1{HqwKrWbsD-(@9Y
z_u~P}IQS-zNN_b7Qyz6KDh7+{R9$^N(zYCgA%Y-ntu6wU!60F_0(wdT%!%B8x%zBa
z8x9;CQV)^XX<U1_*L!QDJInYPn^II7fcM(UN@8Ulm^WG3+2sgctK;mj)eGO~M#y=2
zNzBMGR=TT}14)iKu!oAO1NL=f9k6Wpf9Eh>n?5n*pd_?`y8(<I$Zv$4OnM`%Y;2t8
z&#Mp)A2{&l;(hXVtGRb)&YtyKiuWHBPwp2Yr^D#peEX)2WOztFh~lWKsquq@AygQ4
zL^nhzmT~a8Xo5ZfDR$D|o+L(bdzgL#uPt;KwE$<-L`V8-h#84c%~yz)(C5<12x2K+
z{_zYCk`@Yn0PPT91#aTtFpLd&F?=M!{QmS|=~YE}%;QFvdW&5;QO%+F43@NjM6dsE
zE2?M*z#BCH-(azfgPk8qbOmw^2sAB9+#S(vRJM=~hP&EQfcONGtrJv>EI?!iKBWv&
z7eGh75wVTMRyslJ2te2gZEflsgVf%(YPoPx(w&`Sym4SX1y(uH^yWTt5rK4Q1|C*!
z2XHt*Rp$f_WfZXQ^r&#c6+J%J-J(x`q6<2B!!TrZ&z0d&ZU+?QkhLzeY&AtON-P;7
zUZ-;ydaOe`9BJYpebokWAe&0AaorKb4Mkn1gM57_sW$$7S*Sw+BYG(gCXxheqksq8
zNkL(`wYkwS4B`wVkK(Ukf;LfrCfRYwdjt^0>F1XN`t~)f+zWq`w7`dR-y+Xdn-CIG
z<*r&i8--yLr2cm$q`^i8YGo`KS(1~IuE9n@)y}#Iu*2N>_U<BK5p*IHtzj2AT`fd|
zWDL*X3Z)*fZ=n(;6krm-wZ?Jbf-2HqYJ<60aXe+qZ&8g5P^TqSgpig8b&?)EVYuFc
z)Yn%1I5)SVot+)Gk&?tp+`<|lOiO^Fx%x-o0~6Tq{~tZzW@rA1O&O*$c40pFn$})2
zEQ#!2#f*Dv6FaaNrd!l}O~-)ZAB<AsI(?d%0>eV~$K0(tF8aM2j3`NmN$AY>ef*f_
zM~eB#9*DV%A89(_n`iZ&>ukOB^}KIRV@+fc`-B`4ehv+ZARb5s9(UMT#4m58sXP}H
z-YQ1c5^g(AsE3rTu+6QX5IJ#CRx#vbB`YLROZyA#`*&c<V1JKcTG?D<2@Wod+NZ6l
z2_kX1Paw&mgK=Nwlhe}Hxdv%5AI|N&CacjzFKTt6C-Xv=&Hc_;jw`RNBEh7Hr{HU^
z`BL!?4-Vol#Q6FKh&T@1mVz1C3sm)LcdnfB<;z(mrIoB*Kc^To9zVXFYpLZ~%`56G
zblba&JAhpb{ivq0UZKP)GVQo*Vxr7Dx!>rLE@TS-h3ALaIZYTFRez++U-x2P9Eg*~
zFcz52<zRrm4YMA)7wzs61Z?YUFia7i#<o_%V6I(^lb!;LIS-lNcF#iQvISExjAhd?
z2!TTN3>SboXK1a=&doP}J)t)YgWL)+KjI*s<+R)FtdAR&#(@fhn}f1UVPAJYz_9f_
zT7%fY$v0Pb+=U@=0H+cxsDJMmn819cv%a%Z2h3;shnKiJr(5sCd>;7*?jjL%*oO53
z<$~DPT28}|xCavwp_!V3<d{_DCnmFsdIG_#COZ){uYk@OKN27MSq^#&689Jq!Mco}
z++D$bozZ?IQfG(M1lkV5i(4PjtkKWnqR*71C-u>-O0-I11zEM;q#?*dZzk$2bM*Y&
zYUz2{Qy#UCbfE*H=gj{y|Mrxg-P)=DxRimEvW;Z-?%Y>5EuD-o<!2@hj^gCHKj6D|
zp)KR{=&n+ivfOr9rK6HCmOlH>_c!xl5M{_Jw+gL3#DsnrQ3xJ{80||XoMZe<-`t$3
zJL_|XwC|WgQZ)Fi%x=J^ft4lQ`!OGAmgR5CVW+X;xAP!=_B|W6B;@$zjZGO&T}Q{w
zM#PCzuHQqg?{#->tl%53o3L3bSngc2e>B><=jq;{oV)b9G27^wawYVe+S)yf8%v99
zZIZAw@7BQNQ<B2u`*hn*sDBnu%Cd{$hE+i+|H}Kb;Jzd)2F&7qzRTbK;_KHFyu(Ao
z_+o2j9_D(6kA6F`0;N9|boXm9F^{%4()>iWe@rcvqVnZ>*cFuWmjZUO2FPm`co$e@
zar8*))tm|xzPsP&dah+?46JiEcVD@zn(v}|2M*M<JM4H<sp3WPG3`vvtNF|)Pac@H
z9Ro_wC$(Gr;I6OOM1=FyL5q>6^f+8b{!B=1Xds~=d5^_TSOS>TMu>jRlgF@rk3K=&
zDHHkcWEC%`Suqq`Lme#2<@FI9D+N(8r(xdhV#raC3taNzI4ir{;2@!1bCX{<Y$#gA
zNa)Fa3~Te=y?dug8{K~$43!~l-OiU|Vj4D_oanvPn6{8D*%k<yH>t8<55<j_l_yx)
zHtjpVrKQaqCtF8iUM7(n*$fjaotJS=CARR0LU8{htPtL0x5HbeTtnIdEo<__iHn2q
z40f<u-~O_)9GbQ_5x<$j{*V+?u0{$zig?9!VZ__1QHQB8DMpY7&x8xB&Cl5}F3fQ-
zyZB1`GN2`u!$*Cm>sIF*>0x&AD4o&R@Z?Kwg%VwLsUEjHg}Dx5J|aG_>z*k_+k^(G
zdo4<9YrBy<RlB*u!US{D6HYEJ$-I@e<TDT|e|S&I+a_l##Qb#XVGCZjAQO0<<CeX+
zRNlFfeG*<6717i$Ub+_)#sM;<x06=GI>by<SBdnrG^u`XL$6x{nUfx$1#t?pPxZ8v
z)md2CI|Ja#fgKRO{+9?Rp&?C#a76ruqA8|5=xWTqHzY(;tfA&1Wgp}_gt7oTnnN3b
zskj$Zra9QyWPt5Ww8TFHj@F~hdc)7iSNcin!C)sbIT;z9>THlZUhem^xZ*=Tao%R;
zsS=iETapz>_L?u*H6m~?eGahNUU0VP7CWZ{MgghvE!5OPL3aGw4CFW0bIln^Ajn4h
zI)#i|ujLU7&EaCNls{AjkCaCf(s1YZRE#DFT|Sqgre<~;Q~p5f7aY0q>XmkW!_&*@
z*OZjB5)~t1lTh$OO%UXqCU7)!Mn=Pz0e0)S^-!w)Hgo23Yj$nb-4yV{5w(k*cRzi6
zEhOKifaNtL2ATse)10R`g-(l}c(d+0Iy*ZbNKsi?7rNEImFgOCXcv7>Pmdm$xJ5#x
zFI@_H{rdV~56Ib2UQy~aaKCdWm_l6#wsv%sh5(g183gKyLfA{3R9|Uu*OG*eX$eqW
zlT}b?2SvC&j!@|dGG}l^#2FmG_~;-Q{%7rjm!MlpFnnn^{&A?|j~fa16p!(r4SpJa
zjY7A(ZC=$q5>H9j>fQq0v42O;8MfA7G64}PSGE(%d`X|S3|7wp#kfySnbM(aHwRr=
zbd762Y?FRAT>YuDT&LkH8_pj*+TTYwNM4wk!ONGj6i4!7KNV*709>!R(4>Oi3+!j;
zk-^r|5S|adX+V}r2!+w7>`Xb%QNt=G$7wjy5|JbW``_#VI1TurSVPGyp$9hLR`8G`
ze-+g3EQf1@8{7bUg`c~54}d0pD%qK&Uk4-f!Vk4S2HF|(@=u3O-M2s`iSaBkuukiy
z?MfoQd2`rBJ>85Ui496pbez7@7OgTXc}F<2-A=b2$#Cx$`Z-{*1Eb5T<?gD97gq6u
zLYwqY+iUrKDTLVSpQ$_CzI@5B5_jiQnY3ZOOlnraS$vyai|!IXtbR%rRf4+S8KuI$
zgyM&L7G3QQj<#TC-wyr_bhJOmDETgr3hA)PcSXpNVii|l)lfDID~8l)9)nL~wg>2}
zd<?)~U0Yk*qfb7-1bVoE0q4uf$RWkrt4gl*KlQ@NA9+z=xv3w>$3WOig+@30*c;y*
z7)zH>(zc8TJIyye-14ZD%RJQCUSFU|xT?!;Ikvu&io)u~m<y&{jJ}}2_6@@FNO%HV
zjm96XozT$H=@y^vmoJVpLO+UwDpP{Ib6~ud|NbcYldFtaTLw9O3@dV}(}_B#A^ryh
z6B^EpQC-czlc?bmN{+QZMGuz<4+kY#3$Ot|&HV+;MKG{Ou$P07U)J)Br)+F`9*gO2
zn!#e6`M+Iv;{opm6tJQ*3P<Qd?Z5{Gb%{m=4H7A0Y(I(Al*Ky|eARQ>BCyj$)*<r>
z6oObl6+0pQsNnqtEXiO|6id>l{Jq~FBt19R_4y%w^OnnClCVzOg;|hqP+=BY_bHn|
z@<Pnjp%D61fcw}RUBQ%j{~8l(=mP<PZCuX)sit7IsCtUUm^95az53T+=UXgy=~UB=
zfV*BGEd{@L5mP)}B=<)z1~j<Xx#(cZQ)4b>Y5zb3J8-HUb7xpHWP?~(7YE)hB?(gs
zNV$MX+|Xn0A!u!^II!Yb1F}Ys>Zbvr3dP`<0x29lVOLX<3Vd3w1srRc(8L1umL_2V
z&s9f3&7BzkeJ|gt(`&AI@pdsqeBX>>AX=rsqL>87Wbk<z<TV6)R}YW`b<tHuy3})9
z=j4W@oXp>mtb%8=6C}6TKYi_;(2a>aIz@<QIxmXyX@ZCs6c8Fv_X_@FLKQZi@U%&r
z?;F;ero)RHcKIO!W-g@u%nn=>)FQ*$1K>$G-(0;%xXQ)_ctN1+@HBJA&zCg}ThHk_
zVXT;tlN1D6<n%!fRj9{-$!Bmn3zk7M3^MEDwd3bvJNUm~`kI;vFtMkhzF5B2JHrRn
zWMJ;KAo2CalMq87O~9}Od5T@Xxp^~O_%_>?Svv`ue^1K!{Z!zHLs_B07}ROGtygKl
z{SO>CfGikb1#{{cLW}Iyty`!_&aGbr-w~rbK|Ybv+NoPF62*FIsHB5a@QiKwvndx0
zo20w#6A}lHVpc<7y-=l%HvlupbjgAuXv>=5U<99@3Y6#SA|UCQfk=SdTGM4~)=*MR
z5Rs#ph;Y!iq?y)g_oCp_{s-n@TsLKIM|)SW&s7@Pc)+^{;xLk1VG5#_-2Owy$d&j7
zs6z1B{rEI-hB3*ZpaaS+=)TvsDVVV_t}4h|L7wGutS<jIEHX4S5!zWn_8Es#7^+x7
zvT9f>`L9w<ZaZb$6+O#b>r{(!Qab4SNUk6r;zgjHv_0-()aBDS(2UI=4nf^OA@|NH
z=qEu{6sXj4dWZ=N>(tcL5aN)k3}v16ygPcxJc}nrtuMz0z2>5!fgT>&4<A%-+?Y!H
z37IGea=dZESK*PH3(5uFuDNB8h*RRpTo$m`i89Wt(a%x3y9gVWr)2it!WwFF0f_->
zU8~DO0gWpq0|Pgv`pf$ZBLV|s2L=ZAj3y)`OhBzB)lvV+lP9wB@<A+6%r^r9f5dA?
zTMqn#Ux<p>ygc72s^Q@g_wzQ#;3b|1b}zY-$2b{`@ai*jTf}kkWVw{?86R?vGimGZ
z{GL|bwP?9Px(JRoy^tm7=LP@4Qc2@85YRz5$pJSh2nk#;4B*lhu|&O6M5rw!gO{~k
zlUF&+!Skfs2a?s3S|4*YCo@uhi^g0G7ta0>Nyb|!dLl*XzBu`+Jd__-S0^(YMH|5#
zirSUI@YkMxT@ZE5ynOjGT*9*m#iNMO@YSuZ;<bXL+V!oWnTsXTyDpy%9+Ww$B40uO
zDd%u&*0q}7lXqe*`(^X$yL<EGuQQc*<X*eJQ8dzReKVn$Kh-*QQXxG=C#Wc7*u*l6
zEu_aHB*W$Lu8%7l=g$j26}ozt_EqP?-0b-X%M9U#mkR^og1(axmJgw%&-3iFYP2ya
zVPcGju&&c;S|Hj-qClu0S+3K)9e4Y^5R<}2Ayq*Mn>+a`2=Y#kzNe&cf^;MY1uo~w
zT*%Il>YdDAL&2!-_t!Aq$vDLdMyL6sxXrY#+xqpc6nr=V`n$Bmlp69eC^w0e|6W?^
ze6afvA2ildYwpx56j^!3;@x4Dbqw;vP5v$p24qh|{tPJabXf@?Z%F*nhOM1qU%q;7
zPX6Q}IK4k~gnd@_xy!7x6Tf3YX*8Wkx65~$+~cxd<#TFn_fBi7e%3J0BEduXO~hg3
zxx*Sf79F$=QE=_>vSnRb$?u9Q5lc>~opZh79a$0{`A+C==dF_af~l!*DyKHunwmF@
zw=`D!XS6n-c+g%5@;+%eKH0jvxxmFv#rrQWz%2y<snFnH4%G9?c<k60T3S~>8Mo*y
z8J!%P8-Kgk`Js?K!F0nJndzrBs!XbQ^upNaW4`B@$0GlD!xRXgDWK{yrR`e27+Z2e
zCiYCX;JLmJuiM+327h~5jA`%pc+welWBp^WdDhVb+}iJ?0OP2DG0*teBV<%a*DGvk
zH6|g@jZ>bdGj1-sJ4e-*mhp%xe@RWc`tc!EuWOW<lqTDklZZh14Y}(=4ZS#O-XFXH
z(vr9y%|O#>o5#(kpZ%tU0X6w+KuxU8GZy)t)$#fJjF8ka)Gj~dYQ3FkTLQwSi05jK
zb;+CzGUg(uC-dekg;FTlwfQE`&CK5ymbuP^)4Q}uR*AZ17tzi1^eIs?-U&FX(IKA6
zc|l0^h|XbJhUB!DI%yeCxlGRQtDJQ_Ks*5=E%?{$%_jk5p!!xo+aY@TNior#GcO_|
z(?Pmc3T-}@k||q0+om9%=$viy{g^)vN<%~d!T6$mNBj8t&Ud&aR?TJ)Tb69Ravjnh
zRS5q@J62mO8@W|{(bza1BBZ4oS<?fzF)F&Ag4mnq`beZIEwUV9W(tO=&~B7msZnCV
zLDPK}ivHAEWUh~Wr(-_&iPDh|F=BjaX#GQN8?`f}K<_6GxdQONe~o7~VFPO1PKEB`
z{TC~xZ6+O#%?xB0eA&U&a+@><43&VnkTwrh)G?@usexHRT>r=bZCIFpP4%dykEogJ
z*d0%*Ax4jbsj(mLhxZ)z2hGtOoLahN9)-xc2r43u(C5A7Nv%chs!XJDr=|5@IdtHF
z|KaPI`5Uv}zOU}95?#*e=FYnD;-a{vJxLL`KZ=G!t}nug1)M;^2#dx28x55jBnA~G
z3txPn`u26O!x~Gpdr2miRHHDc4aQTX9HS%%KoJ#}!^a|5v1K8+FCTZ74+VcACJw52
zy7v1^XaT)AA)RyM!wytdT^(2QPI2jTt8N`8k&3)oa$jla<X%_hz`afF0Iy<^p=+`;
zQwx-J<T3(7e;=$?sYoOE;QSXthL3b0Wz#6?&V@F}T(<=iCFqnRKAoeI_-qn5KOdLE
zS$%4dmhuz(_8Aw-S|D;kZ_jQd8(<YX|2X(!$M(S&=k9JFeDUg!gD>R&=v!j@<GhSR
z|GbRyqjl)pw=9y9)v>+P<4-ZH-~!x?*kB#n(i*?+QFD$VV0`?68Rk?SY1lGVJoQv}
zKWlmaGKa;*pkKwb4{VL$RulR%kJNQ1x;1qt%qV<PhZ}gtfA>1aXDa<`yg&`B#Eh5<
zJyGn{Abg7UHLtqvV0UuPTXovkD+){_z{uK)AMbLQ!fregEsr}IkyO|k-+i|mlvK8_
z*GJZjO=L+(GVYL&_&Z|-y3^|frUQBoiE_E5xuD-D1PRtJWp~A;o#D8knx)MUprf<>
zvE<}Xn?{E4WS^0Y@)`<0YUL4SuRvcn(3YOe=dE}NybcOw*U%7V+TDmv)JksJtvFv-
zdYvU>!twXrnMJ&P?w;Fb6gNmQH_DhX!@=o?wRuqjJ9sB{n!?jPUGc&^$c}Qxz5#Lf
zU;b1(wc?fbo-*d;&?RF&z7crRZ^ol*8a(|Pr8_PX=k*Upi(zy$ul;WBrS#=(^5HS^
zyMtl!AaRvF=g%X-&iP@K%NMxyDF)<vN=EL>N!tp8+!7jA;uuLygZXI^`I*7)cXQpg
zzM70Q>~-(J&@Q+Muidx^GbcvJFPOP9(6YbaKHNk^u=Cr9G?`TF$Ou2P9^DQXSPlgb
z(M`f%cY4^G@9F_R)xZ%6b&Su#+F;>B$|GB_%J^JgRwijmDs-!-NTvJ8n09%W6v1)%
zoxY)W<eO~My=_^>58lS|I`uk9dfZa?cD;RDDB;fK%R(P{(brQ7hH~B3kk+Gu`%p3G
zP1f<WYG@1>$j<)i^6*@VH^R;AmF9yFVusZq5CJsUXX=`vsiBdQSw3=v9>eq=W0X?R
z%k07BCzl&*{;j|+7I-zX;sE|^`!<T&OTit0%HO@PxhVBc43uhX*XEyp2tsP!^QNxs
z75=<f^OUr-Ht@s4IVfy8>6$=^IG}I~m4yZg=yV7dch5t_5Lv0f4Mx22*=H%qKs#l%
zZ@GQ198TV0fhOWEdCb@t9H>&*cdN3I2Tu0rxfPq-{ocqI;cBRALH{}GOhdwqTr`jo
zOF-Xn!b>^}vXBrjC;N2XyvYH>4P`*9r+)RSbs;YaXDqrdKzErA_tE`t&7pj*30<xZ
z-z{=eph+mc5wHP5T*$Yvv9{C+B{ljY`3kQ0?+(BENFqfzb@|XMyD~sfs8Img_JG>8
zK<x%GV(3pUSgfn9RRuR7%Q)yH&?Em_DmO_1hud8pPgOd5Pu;-F&i&aDLUF%|KW+#N
z9T+p}R|CvB>I|Xj<HwJBtJ*M5?Ci*0iS()>=cyEc(lBy`i*le4tu7ssGBPp>CJ<C-
zh?No0#N{u?$jR*-NqScJNJwohp=U$DL7q-9!4~tOhMpr;@Rd+82*u3L+oPxnh?(0H
z`p|~Oq@AG*g)|cIIxr<2ndMH?+|?jO>O(ybun6@}gY6m2G!vU^Gg<*m;)|f1q}ueC
zLCvlepd<2GorWS2RO(y7Bi#-SG)kdVu4G3-=I_8J>Fk<HYn+OOB4%)S_(Y2U*eXUl
zk374A+?mjhO<%-^&03(%WXG!pTh54zZ=kLqj03|GAQBl?)yx06`?4O(ooV+>c82P%
z54ne~Ny5-b{OI0YwmG1t)BWwE#kN?P|Ld?F*o{4=pxR>C-y-cF0oq5zIA2ah1v(3c
znyr@{uYh=Zmiy#RWm9ch25exj5GiF$!@CPYSjACTEN!%!<P_L49BAYtQdv6yn*_f8
zuaBa@yN~NWDds`qGKeWBq8l`|*+C5#!~O{V9x|S1e}6iuTK15G0luBRt@D=BnUllr
z@FhOH?3l-;KmQs9oq(aQyD>u*!)M~}9|z40?&BL65^~s4q4n1_5K{j5$jdMeew*p;
zj*^vjp6#3aLSPYUh`?r9kH~@l=H8V!4{8#+S-S{fk^!y@7vkpaJM-q}i^8%@vT<&1
zsv5yVFK2slV9=PW!wz9(_c$z*xOzY^uCpPX-3?h>l(M~z>eUBYq=cVR>N!;=(Vu-H
zA<fRvv%doU()IxQWm$hVo*5FqIXzr5tEaWaJe@it3U7vC+&`YUe-ixU-ZA>wt4FIJ
z1v!wM#Oyxk8<l^_eBVjE1Yf@Wn}FpiFT>R;1IZrHr7`U2+$ZrL^lP5u^zIwB%qWZM
zeLA%C*yh(#yv;9Sr6+nytnJI+ekBfZ0fzgZ+0_!c8o@@1U<Qb?$?qEtw^p^lRi6yK
zg+*WiqA$1v7r5?mh~a^A6q1wG5~1H;Axk$M?Gf<yK`e{#gl-6A#zu|=__rNwo0A`&
z!g6YQAw=qV7m(9wxIOF_=q*BoB@-m3qD7q&*Y{u>Oo1pSOF&KbN<rw@FzVw{hB45y
z1ifbscT(F^e*Cxvi1PPTE5sK>IS`$o7qqq-y8vG>a6*2tU<aUSWCMdH8Vz;zIPeJ|
zN9ms3yY0pS;Kzehi$iB=1Vu(p-L<gefRMm&h^O*A3C2ghb8PQ}hYyijQ>YJ@LUA}$
zLNiqZh_XdrNg?V{fCeQ_2w2bv%Dn*)2L!V1Lo=YiM79)W22P@IkS}@M%e}Ew-nF{Y
zsCM|kpmyOi_v9T|$pJMX?iWEpaS(=}BTWiiN66|2f~?-^b2|Q7w~$}xT$OyAXt1A(
zE1zxR`!=PDFH$$)rhOC?J@YN^ySP3rO!6ry!wNWp0Vr$>;-01ME~wp69KYBETs9rJ
z2afQFnmGg9g%FYz!I<CtXJDb$qT(~4rByXgw!Chh9LtDdHkF6{>EaR+;CP1i_&Xr;
zH)NhY@(lU9pe;tw(OfD}UHL`^d{F5J_)7zxJwDe-g^gy<<Z%IfTK5`COd?asnlaft
zI&l=zWA)$BO_HByzCN(5U?5wM_6<jW(OO+~yu>#_gFe`0rK%ay_V(fJ_t##&^uIIx
zkz~OBf`DmK&PU?Or!7xajjcvb?rCENq%pe(;7#ON4Gp)VgP0%g!;^5HkOFm4<)hm>
zg)aT^z+b;WkT9mX_79wNRAuKjPU3zmwE@Tq^O33RXiWf;@^qhn(3K-s$ggr*-VU)6
zFMPa%=qrq_RLLgh9C+0nvwdIe57cQ;{SRyz`CX*)**Ua-Txp8+WtBo@%gak0yvmHO
z|Ja<JrP|&`2B1Xw&g!M-v^7(?k|HcS%SjfhO**VPB_$1L2*WZD+CAY{=5D9r<NM>U
z!@^X3Cc97XL6Nv``=qy$r}b>04843w`2555YweF6mt3h~d@!uUzh36Xqa`#EQ0o0V
zM)o*V$M|E<c?#I?%S<7npEL&0XJ*CExI9jUc#C0WzQa^@G-!Y-mfrQ8$k@JknTgqO
z`o$!BCLkMhA+y2syVcbi1ev`Gya@ZJB(@(>;UDnQ^SGFto}|uEbRSIO0bI$1uB`C2
z-T2wQ{d&v$najPh6j8jwJRVaUO6r(u*8&ErsQzZPb0l>;B>su=rT8_E-K7U0fMLfs
z-hI4AR3M~k9eep|20OYvmM^q#?zG;%>9w?TJMvM$!&u;w@7v#gkn4E`q(^L41$MqV
zI6DBu@R>G5Jc}}fZU@x11wAU@6`BCE{G3#<6zQ#p2MV4*a`U17&wv#72J-xX+tDpa
zwt^IWI~v91nEx<X;fu~-W{2TXprPd1Ga4qQ>7G2@bZBraOX>&h3H3&T-x};B?@q|=
z#4<ml(xIP>KST1L4SINR&mc7fc`A<*{{`t-56IdHqo8yYf|7a9_O~EF({TDI4GqVc
zGhgGCfLccwN<upeYMZp|(Zb`4TFMR6&q5yuSaO{V&t=FuI}22QuCx7b5DX~tKiBb4
zq(LX<Nr&#n44&|!w}QqgXhk`RDmYMM+&*Xq0IE2YeI~x=XF|xzfupjJMGNfpdnhU4
zJeN8#eF4ms4A#tY5}1vkfuSe|Kuz5Wm^!l|8T@+0y%#HCQrkY%<lj)mzb@%NH{jnA
z6U_g=gOUI6GM@#Ao9aPOLUYZ5ka@s|>mxPq#LM!;y%6wj=F5$bLj*JumF$0i7ANp}
zbYjPigOXp9Okvsi%oyKZpLM&wy@O*84EoX%q4I<EbDf5`S0Oi7`5X^W(`V+&<yV#z
z1M$l@WyE2LRxDlMjrYMi^L((YqCPSYm`&PvVHi9bO#-vB;=Wv~xn(D2Vr*A`plNIt
z?og#Kgbj!Z&_}<Q3ao$!JMf<Xt^e@dFw9KKxj4r?|K}+Cl`Fz|%>~<=jSdn{e4^a&
zFyZF4l@<RxmmG(-kC`+({<m=N2UG|G-QB~7Us?LPK;;X=_+B8cI;#j<3G*Q*gmIAV
zEXX6<Db(HUy6j#APk<TT%ob+aS~G-~h;VE8y8edIk97MsBoM=$^VGpXEo&RUy^sH2
z&soLFFJ}%;+4%(rvF@p;j9mtg*3iSKXMc}hbWF}^dmjneE<}53r{Z=fND6s5Ck>nm
zk@K=`oRg%9vXaEM_dxmV$&Y%X9tXD{3V`k5{R7=c_7D8yqX61;bX5P)J+re4eD-q}
z>;-RA#flYBUyIKOko|YUL+1@>SwuagU}T4Weuw4-)Yb_nas{FK>Zi-4aKIl^5}a0w
zMOEDS#r>bbBE40lN<rG5S4Pgg*~1M|hBvDH8PTzkj*dCd+4Kgwa4bQIje@O$VFvdw
z&%1zj;M6<O{B+G!h%`JC`4;V*aI~T&l<1(d+3L$d*cSsA5$J*QAKE~L1vV>(I-Aj9
zTyQKz>9bX1V`DhjP>v839i0i7Nx(5V=`%VT3HH8g(Y}bp^9+F{C1OgU1MTn)w38*u
zM$29E{C!^X$0jXI|IqXLZUf*77j{VK^IE%uYk?ijic=CNs{3ugQBCMXCTPfSD+2j$
zZ31DC@3-tUgxLCgjaruN_LUy0{=D`_9MU@uf+g@(RD}ZhF89{wJ;)NeDC{`)Vqp@H
zOEQR2l;A7HFtG<53`b}GTLuMv#ktnKB5-g67ayNGOv{2}7hG+2eCOD|;2P+*oQmik
z%vfAp6iKFZ@-_P5Da7W-T-_vUr&mH<9cKsJ39h`c!vUol6l#b+p@|~Xx_C+e9q)j;
zCEx&$_j13*;lQ#q%~U40pD)SNnq2g(&40sTWfnaJc4#K;?Jwj{n`5D(Z0+c9h-aTa
z22U6~f8XGKuEUMFswm+Zrmut!g@E%V5RL>}ZaR@-bZeu0D-nPy{|%?ISiJdfXvjcr
zZm{obK`*{sG&J|2b4k&W2I!m@@TK-jx37}B_&?cp_oIpn$dcm9tUaH#M;VVzmQWk%
zx61fru%nZue%3%YF}QX+;FJi9U-{5<&nV%c4WQgzI%Jf$M<;w^-H0E$-|qC4=;+ZY
zd-n*l5l&el?<=%A%u5B^9yN+|xskRY@!Uy)o}?&>0j>teGJyr1m<5LOvLq!9jbJA4
zbvvm4X1=Lgpq)Nb@8!h|=gc6}KbX}ibrX5)MqjN@8g79z{SBPAmhp(2nhg|QiYk$r
zO6|@ZJNn|(bI@|-2`1a4xuzy2MON3(9ML%w@@y6PR+Yiikfv!_+P;fVKtn@gb(qJy
zW~04yml31#(y>W8!5xblu23Z8S?Ph6>O>e+;3bNH16}}+=zg|}nl7*BS#2xZ!9hVu
zB_(>$N|M`f;sLS+!$xwn{v)@hRUx<!`~el;+5qc6;%+z9n|>7mEflEi3r=SiT))SH
zA4-9+@St-v|4rRPcm&cvUgQ&P(ouBc<9VPzmFM;uf#tLcPAjpPDu)9Fj6hSNR0&rC
z1L`^861PyJRaIBEsUOt8hoHP}JDv?mcZgkb<;l%skKRKEsnOS0=c7)yn-)08HwZ$j
z#}L>e%q#ZXB>&rtIK)iWz<E3HEo^9^;$a;4;Q&@#3mhw#0)^YA%Jvoig-{b{7%Z;x
zXPndv_k@H`h(pH;0;$`MYK_s2I`8o#4d*WoefZ}Bs*(k(A*i$4Vfeqx#RHF^R?Bhz
zeDeylSSx{`lcjw|x7@2VyB|95fNPz_p~@>@HR#(50u}^SaknHqX3IRpXbW0q1tn@F
zgAtF|g1$5soMtNV?eOzPICO~!BK;(sqSFCyNSCCG71%#_sTc;861t0C#6VRMbvSo;
zL7QkG1Tgv}PVmdaC#Zq()5D=gL+cJB&>=H4uLo7%C^&;i*RUKot{)wLiiF%wACV8P
z6naqz9MIjLkTmbO4fRJt96XD`Vnn@(nwwi8FuPZWcEzaNC?Vwz;BAs(ENIiA1sel2
zd+DNG1qYf=+^xl8tk3uTKkdDFIMsXGHol}e)z)MRg`|WC$y`X9s7#5JIZ~N257iFI
zP*R36Q!;KznN?US5;A0nSe1E}d3eu{eee6;&po{N{k*^9cO36~98Z7jVf9_#^&LLL
zb)DCFo);Z3+2hS_(|ttPL|ldSM<10KRPFgFjf18RVRic{Z%0wHM`di0sG8~T0zYk<
z4Y_sL^reD<E`CLWf-ybtwP%;-dFBYgMPxbK_2;bQ-@S;1Rg7$he?d|H<9+$RW)=VQ
zwSP#o{^ybW&m;K{lMu|lzkzE1dG7LGzIKgQaQALCefQbSCuJ11kEDtT$0qk-TNbUj
zs3>{tU`S?1ZFhK_`Rh{Q(1OZ%Cfu}%yXMa+S-7VV<s78xzd5;bMipFH$}0OEaM0>u
z*+)urYW1w<GmFUTE1Ctb1nl0b;I<Gs4k&x-nwq>SI>6*efdC;MTte8T9R{CGXJB0i
z6OGuD!;KnFe8;dO2gZS5b^&5Vr{45ZY@AXb9_>4Yn5%ZZs&q7bU5JwnA-vfl_R(qO
zhRg~3EO9y@1yTHVL*Lbo5km8EoB9+9I7$PHdKDp0!3|fAY#fIUc@qx{1d02mBlp=H
zB;MZeN!qx1bMnK>1r4W*;A9&IZ(}}amLLj6#o%heC1SG<xh|Y%vE8;OpNE`;`24}n
zid;Iv@rv)$8F*Tz;isNxeI7mpFk+!c!0}pKZ#=ZQC?NEJ`({1n3VqE6Hnt>eH!?8T
zCTD?9>isefddg332y#Bk&F8h#Xtj@zu~~k}`58!4fC{JVBl&V7XeLPr-iU|<w6XAy
zM4d!DsNqtCqEmerY}`KqeE8XH^YH=*lCT_7hmJBO6OUUBbq3_uWRHdwPp``*JU$}O
zC0GS6?`Qv)Gm-9d*$eJkJNfu30U&OKfgW+cMu2^<#K3AU@JhC8002U{xuJlK8#cso
z$T)Q(NY76k*mEw7PV_g(oVa1Sf+B=CY--GjG2H}>bKdM?OJ0LneQJ0*;>B3ltvh_X
ztLzIc`uAtk_lzLzVJ2pCxF0{ZOIfeAF-}Oxf*>ardL=62az7uRm9a1qpC}P>JMRl`
zR|$9^TR{tNSnSORTHqdY*8o_TwVFYvJM`HwfCYKVL8zP<fBEGcamm}wP!xh3#_EGu
zJu~XxVxUO-VY_98AzJ?3(9jEyWSr|hqq5D&%-q>K54Sxw@EXM-{r5#piG{e#fsE-3
zj~Z@K(HaL><okj<Km*lJ7|-1g=F)Y1J9Y>|;q+))!eKB9UKIMw*m2=AwHHysX|xuo
zmH^iz9m=KP0ad`s85J<zqG6jBN2$sesZ~8Q#8tTKA^1|UzzU2C=*Jg#Z3w-pC#>8?
zoL5A(DIGk>it%_5f1g6Tah{n9rFytr8U$Rz!~umHCcp^@2Z`Gm4Q5;-3`M)4q`E-7
z0z;s8viecZLPG#g#qc)5_07NySRBmRGnvGolB()Hb&7^5Pldap394-^<cZkG0hNJy
zbKd!&;9xd94KYMiz9?LE%v%7fuW%)PXm^a@UXqnL0Qwp$k(gWcH}3-zh2@~N^>D3Q
z65T1-T3!L6iE5FS|D!nqjw(a<B4>)TAm;J|5p!RYYPh8F#Pl>9oUs$^+yi}yhWsUB
zWF;DGqQw}B7^8}cio!o)ZD4`eoJs_KmSi&f;dXUeV<qK-gMuWQ@Qa4!fQfc9DDMu3
z75GRtvCq#!hCp7IX%p0meyBjy9|Ys0Dp7;q5VC2e{Er~-`65EkdHw~#78U4|FDy8?
z;mSAx!w3_wsB=a`aW-M75@UhFQ3pOw(oJ1k#&Jb<99PEz6C=E8YUJGL_E>lmRCFXJ
zCNd<|9`;x&%_kyKtqr5!^URq6Ydzao7w9B3;D}aZNy*8gIxjC-euG@teX>m77aq;5
ztslQo^kDxuF*(TspAg#@u)JCrxHfVfUX8eiSWxb$@XGI;ao5`^pPMm&%uh112_8zg
zpQN%nhMH}Uue{<$#Mn6_3HBtAKS_*7tZgVyT)PJMvv896=N1kQUg@aE8xcGuSK;0d
z&zVyoHp1gQ7IHjPOgsIh`YVyw+c%7xp>9jKuyxa$v4o&-f@x~grLU`-43HoJZfuxC
zvq}SZ0SJK4SakXf7S|V-z9f>o3G2B|wskx)pdnsdShj=V8qKdV)pyyQzi2m6WD;8q
zDE_*)&2x&m4tz|@4l6OZTWg-*JHAAIHgPUH+0rFT&CRXA4Xz<l)8NU!-^~DX5dw!&
zhe~}hO?M($$AqYO(JkEnCTAP$sZ=$SYEv#a_V-S|u}V%(_TATRrncD3!LK4*D!05!
z3JfJ;Xo@qG6}&)oX3}o|{%ex<{k_dEw0Bjc!Ez-H%0LUqW{GUkpHnUsh=hcKktHFo
zLavwlQx?Q5P2`?7f>-V4QP&|Z7)WoeL!EU6;Fk_gq;=g@IC_W6y6O-w3m$@f5{ntU
zt{U+Bj)jUS1=5O);6+Ma&sn!gW3mxcIDYIWJXNeakD5HsXI-Qp^zBanZcQtigpED1
zDasv{!7>6Pmnax+9%*SDp0P?6$HmmS16D&C{k}m{hN>auuA7>ecoMa1oATs7*XD-3
zUoVjRC$dZ!*+B9eToECYq(9rOb@=#kO?CAeV||DzbpcNkZ#`mUqAM691$t5>ad?6c
z_R&_i=@g!)sGMP~dvQi=4l9Qy(iu47$H4XU6ej!AbbXkOLCG;nY*s*O!me2d7bY>c
z83$q>fs%E-ko)8jvJn9WCQMm!5tw*H08G@58YlsRw2G=%5=5220_t$3c2lDlh^-w)
zZ^fQwVyV5S7ezhIO}9(F_r8=OZtHz5`DaKw1tdJ9?SQqgg$rk{rTl#IVyNu^afE`Q
zXbN=chXSERjV7eqK*A`7rPQHSn|6T|VT!@I!xa{zfpJ8D3NH$nnAG#&O%j)zQ>rrY
zxGduCT?gMYFmM%3Bk|15;n0Us^N4gHz`~bE=KgrLy0^V)tH%?@s5UV;89pZNe#Esq
z7Sfk4LVSmMq}izti+l>~d-=d@Ba;F~C)bXQxz0^?5H1I?B?z-ft%8B===Wg#vtu%t
z&!m+@B9`B04Hw~c5sXpvl6Cq89G*a+s>d(YAO<00XbN^0y}fb}Jre^2R2B)MKH3GI
ziAFpoJ>e~rNQsm+@V~?d1b>4Ss~QY|+H4zB(oTSMi9ml8WtD2;6ajTDoWyp5B0`WM
zL~38AS~M(L3-XDPq|D^Edt_CCOC<3^!JDeCaii1(G*Z1*eX;o{Lny4IC`(K^@Z@8~
zFMi&l0*QDv%wV3wql%zD#2b(>*i@_|Hjp3zHY|D;G6b%I+Z~>8vU7j7a^d8M%ewtN
zE(q(0@Pn{Ph&f8i*?b*@n!!9W@R22!{TtV<ivdd`v;|{+SKRf39}Uv^0F6Ot^mB{@
zKtdR-7*vm4NJvZ!tTn%nZaAQ;N$aigC}SqJ$1a2WXW=kEX1inci!H0Z?f_Wvb#$!h
zGHD1P={%N9+>=>6+h*?ZQwB6(zE9G2SVIgUZdTnQ|Bu)%20p8#Z32Qha6{TpnQoIg
z4_pyJS@Mv+o)gT?%v}ON!Ou9}TKbUKJ&|~X1hLRQTcF@US_eHTJ0NpD*@CrTc5J~_
z9xGlQ9H@1lOc&IA&PAXsiC8@S{L0cL2x@AfK1Hw#z(WJCo^S*gO6b@aV~I~2tXQ2u
zVt!8iD90KtIN(!Qem<GrgTIuea~1GN1w1Aa;{sO}#k9p>sf$+yW=Je!RI2f$1f4i2
zdF+Pl6y1G`MwehK(YrKvDN{#C=Pmcvt^45s`GAKwZ`A`avjwWWG-W*{b?L(4?y5|c
z%p@rwj*plwj{#7k{&W;~nVX;wZVprmJSr&doKRF$R0#IbQP<E2EzlC_j(o`(49+i+
z?MDd4?%!P}#@;e#^Zho6oD<dB_S=?MGA&=V%I35TfUwS(+5obUOY?&&9|LPD*p_^Y
zaRqgpw0a3d%-vI$?^$&P#eDKUx<szfU6vCBqw=Ti3cKs+6)zp8Ki(G_9y6>~r{+wx
z=NC>n96EV25DWynudnDJj<JBM9b_AuE*tQ4I{bQw?H&?~*7fm=UTriaPg2>VzyskL
z7*KJrXGsD|LzFTk+aQe@pv1rTZ6ZY&wKYIiwKxLi3vF$Qbgwp3B>_e?Q&X=EnjRaV
z^|{NAXtcQLPQmvER-QGVj2#333V!R8(<*q|6tKcjuP=mtb%@%D6VZrDv7S2S1o862
z_HiC<n56YQ(V$@)_zkcPaL(Bf^;)>DR*srO1$Qn)4O9d~`w(HhGe~#HdBjURx-lb%
z%~pG!CuSfBNDVF+P4M6elxo*!p%lD38rwWd2{iinTS(Y2-`P!Dzh*G^xX>GZxsR$9
z@6fe&Hi?QsyBUUwrT!!LQ6=~D1y+Ulw!b+7k*7(;qYF-Q@T$^;6%_>b1m-8=3#@-d
zo~x6J%-so-cfk&pj15=~eoGvm1iOjYwB~;s!Sa3r<FIL#&S5Pzhm*kHsOS3~#bhEz
zXk_x6w-(H99C(5^LiRVFu7c@)V$&sxN1wAZV?W};q`Hp!<kMs#AU~-{nWN+_GyJla
z;M?FVIzbTn2NS7q%5@4w01O$TJ`r>h1O^G4@+3@SF}Rr`@i}DeVm8xjD)-5Qj-C+S
z3TOK+2RFMTO&h6Gg1PnynECOPh6ZuYb}AFwJzMz7b-vpkX<P6ENkbKE=;BZ#wE!Rs
zj`Up~%jT>TfKg8YRSoBFtTlnp88+rR^jb<lOB9KwNCfoY(R72dTSv(8Slv73lcwfH
zNN=o?5@K>laN;?<dn_4tuwV$oLxyX)e)i}*{C7mpy`pU6;Yolt0aV@^;~sjoJyc1m
zq<#^LSyaUD%n+UK&lMl-FCGc}aMf-)0w&(LTfL=MRh(cdPRh%~ae-JfxQ%5*Ucjsk
zj6;M+oU=&xA|TkQoebqbTu0$hNc3UEX9zfDA|k0Cd{m1NGYPJ$orENft>!1Lis;VB
zazvCkpfT6x43J$F0m{OM$UL$05Z@nypd-2^-eE~GF$bV5(2SII)jjJB@6v_U29MG>
zU7NH}LFmdSAdo`MI0ixtv8y0{NszLH!weZarughE<&}1+cE>|tpf&e=eK}>-67N5d
zY)q>YtJjUio>xlHh$ARaHQC|H4hO=)iQu=Ak!&z{Al~lqwJY*kt6w!uj0oX;W?q|k
zA8RbBhY?g3t|P=rj68kPhY9OHc|3&cgUoh6$aV-NlkHlOgVq8^j>Qw<11uvqRTc0Y
zxgp^=5z9EtUIG}PiOFzGy=MLBEBdu&h&EvHl<+gZ0ByubG>t!jn_E@g-7B_Jw63T7
zr&ll$D}~`%{S99(?d@SIoVfX5*XbA%u+|phjZ~yrobjju*@qx3wpA&96|KtfF@<qF
zwzOy%tsVmtpc=0*2$~wyh(adQbj471Azf#P=mKHYQ>p%}*!``XgT}sgtG!T%0u|Y5
z9pm}bFUd$}!5peHrf}{)=K){ittx1WKzQay2lfKeAYL7DIvr~@ds+(%5z0IUi}Og}
z@krSe;-P`ZK{QHQ`mWo^rsbG2osDlkH~GLyY-$j_Od<G6CEj)D6;p$t5**0K#B~{_
zkxcn7V6jCUCsEdM8_#X+TIfXp2jW_ZRv|33sI8JXhU;B=)N6!@g46Cxt_UxTenrTC
zRZJe+w&~X|TC};(?mNse>tQVVs0qco0pi+B8qYyh^a&lp&N)Rc%rS%qiCSrA1*Z#w
zFa*qi3K9J&CW{^1B`;tGlM#W<%obKnPa#;!!A7fh_T|L}5pZog3XuV85@^bkfrUg?
z#72~8Cj$cmk!U_f?S^RAx*IZ{09Vo*BY`RLs0NZo3^K*s=Uh~(5sMH-ECxi`WE-!F
zfX?$u7>NArj1Aa{O&DMh+o}~5EpIX+wzHW1f{g~-Th=S=Q<ta2FhCzq!CCIS;z3qV
zu!3A>?LJ-ZvH?usc%lj0HXs9E3>h}`aS%qgr-m(=HMV)vcm-PqzG$PywmpN!Zm2hF
zm~7D;0*68Fhke~?2fT<e5S&%84_!cv(G8cbCr1`t;h^T^yw<LF0Ik3wzi{TeP=p{I
zr5j(Tz25x4EtRw0#egH;JkWBH#%o}#JZi#>wwpSO;GB50ZBs!k2M^C;je!v~+p2~E
zYsV4UNNgg&@;>*?QNnnsk`7WJJBe~(=lX<U&7%UNHFb1Rh;V3Mu~zGQ0h-|@)OBS1
z?v4$_p&VPn?_UGu_9wjf{*VHk?YL|O!Fu*C1vX}8zIVd%Phdua@P?TC06t+Ty2FFc
zMW+y%k%ftHbdg)UZHPcc30*UY-A}OV!1uD?By3LP;>DodG@V8+QG;*|p`I$~M8X{)
z&Gl|*zXaBHm1;m<#6c6o!4EAOanB^?C&>+YC@4N!brk|>#5ErU97G;>nfV!tv@dz(
zrGRA<gs$cWMF?VLdoeP?-gNRtio_ssjzvmK8sd>Ac35h5K(&W8D*=VFBr=o{RuczR
zf-(}5bJFrl6#a#e#HZA`aw2UbRyD|>q3%0Fpc2v?j@axj<TywOVG|!m+REXQnDX6F
zEhD9QIL1B&48{*6m84=}doQO#q_0BER>Nm4#UB#a9}?dKutOG-*mQZ52Y(S+qHjQU
z?fUNRL>qm)4hV+%*JRc4W80Hzcd9_%r;gMBxaKH;w$2w#m+tXkMIde8{>a!7U8*cB
z-grj3)~&<`59QVx$oN(K_s_%LJcx&EEpQ+Tj6+WqGtdo~^3l2FtoXxYu(Yj)INf~K
zom6VEzY+^x9u*|`A|sX*wMgb39^tx4RutXzSFSA2)ZdClvV+A3!nfd67jeH`GdaXy
z4XGhRks4Uq&tXImJi8hFM2`35JVGFGgcuaprRk@^DILjcE>=`|_VdX}8&PL3-fR|C
zpa@P-;Q`?7joCI&crIW|PXy4F>5hUY!WyX#ufs}8H;Dn!?`Vw~cHuytY2zQe$1s@S
zlYfnP{sAr$&@a#>qY^Q!?8H}klHdPjGZl2QM427FM`A{Ae<A+z#J>y0N%9Wy81-u8
z(0veD1Ol(SJmE6Vn(<5#@cd)(wez);g$ki2xcumjEw3^Kw!Er3>qSt4wXzfPc=IH$
z2V8W@VjdxI=VkZlhs2v3(||DOh)puj1?=!)h%c~5hBLLPHvwb>17buUZ<KHupflpb
zGKTf1ie$+^R{&EKo-A~mB?MiBtvi6s8S7)IA*mcRR!`67J9U}LO^0GUh+2O;L^g$r
zRTdTF&>CDIeL4aH>Jg@qwmmhk?bp82FibW~9#LA#stEW`MsJbeAfSndZpRQ`W;5me
z*OhUgftjExfR72T3P;oP$!GH$`|7b;J}Y=HlJc1L1rFzH)vFy4k}fQb;m!z)nQWr2
zO&Mi~eO3}}F~UueGCJ9^pr@F9##BKu`^MaZ!x)J3YEnVu4vHu?)PmcDJG!7>dc{M#
zzUp|TsMVvko(Y!0)tfu`VO4R|3_A;utM?wCim_v3iG1cw=d#H#{PMZvT<h{{OOO5r
z-XN3@2q8MUPKG4|wMb(wN5j4Ex(y72V!n1i1JYBiNW6{+Qg<P5WjUq;m>lY>?_DC4
zNUFBhJYhgUr+`2IsE_)$>?Sa;{_p&ntoh&5t`H!wiWqql;Te|MSyU+o<?mCzx-L*8
zkxGlfm6VuyVZ$VOGyo3WDe39n43f?f5h4+80ABvhD{%>BfDq(3pBt90DM6<_Nl6{D
zCu5r?eG`IZ5%mGg4dQ_d<N-vuV~S5HU+rdO^uV8M!zzysi_O`VnL0laedLaYf*3^_
ztjx~AqXjM(KLO5)+lErU-p7S`u#y=~!UsZ-A%Fp(#%i=&*(xVjFLxIGCKt$A!25-P
zqB%$D)I5%h`<2#i40aNlQBYje&~6HoKD-I@2>-U}%0Lqcl6?e{Zp*O6I*R0_3K36F
zwGh%D2+;%MV1Wnk(3;Cby9K1A9{%NIAU3^P26UO|Z2;P<0ShF}EKt`AmWA5B5tEvy
zOKyHL0Cf|x>Z%5<qMW@6{-f;Me?vE)8<?L{z0D+~dEmfxVlaWol_aWJ)|5eMqMr$X
zpqRA2Ino4^!$<n1B_$V7pIHrq0|y!G+yDO$z%briJ+xb%nb{n0HTMbsf%R<zU4&TC
z{b+$V86WT8H{(+Jf~VnPMAE8biwGlA{?om`(GF7==!K#(OP2g-q+1}bMs?jHWcc5C
z;>T3~#5?@sSn#IrZ&NyWc++i~F#qjU%TH`yARd~95#nL@lK%$lGxasfeQ4a31lv_&
zydTrCYUN5%Yyt=;k08vPLcip;YdZ-Oa0LVmDBPL@k|t({m;vyiAD(Y*`LgWv=@xtX
zd4XL?my&%HPi2%{H@d&**2%IuNm;`a>r`IeFO5t18tQ*btVz5kRL3H);`ok$)k{jL
zW{S!I>FME@PJa>>_^ioV(*Hc2S5zpD^+3b%b2kOd#cB>!MwQ9BKk9w|Y*laOqw$f+
zatYIc>1OAdcPaIgPG9Pepu3r6u~_76uPF`7yWfCIhg8*<FYJ*s-?kr77obo|c;z_>
z#8%bV$cp5~$J5hO`^Cl6C|RE`7PGUn8;a$e<>TTamNpIp&aG#BPSR2M=%g4qq6-V<
zuv5vpG&eVIZk?d+J*b<3NIna6Od6<<ek3;>syU^)G508ia#+zL!UKU@X{qd3dKOIj
zX^5`;S||Kky&+<i<vqsQ5FPo!g!|!#gNnr<GRB~F=;B&y(dqF4d)4=MrWeU)ua63-
ztf|=mNQ6VS1+L+uVq$(-Kfp}#N4%w^SPYNyb?8TCl7|bh0<Ia5l^Pq%Eh#C<#btn<
zKmAJEn>U+O10y0h#^V^Wg(yK%hB#wg-KPBfd{kL0P=-8bYj0nQDD$B1r|Rl;5BFa7
z_x48T(&-_<2-MT2;Tk^HxOow}spUU^{!B*$H56X0Q2*Obcy6c{o=w)x7C{$bW=6)y
z>E@31XfF;QBtpAk4b1@*?@&!`?R`v3iqGp`{Dq&te5vg0Jnd}*D&X|sM_hp);=;$D
zjp-<AOGhy;<rjnk>#&MZ?hqk?$AUnYhft2UEU0rAbi0R$H22>ex-mJ4-{42Nx5_>1
zG(Ej|n5HD^yFOh-;eDLM9}~;n{_0w@mA-zz1Q*x69eR3jVi}cJxY^z7vR!)ZzAIbo
zIVgOvncrEnN+>RT8D&_(!*gYAb>LMF%3CP4cGh&)GWz*(e-sn{Q9)0+$&IhRH{yEM
z;LWm!tGnAetsxnfd1_{wZ5}Hhyo0rw+C`^-xC4dEjMlxBZ}(MxRr}C8uJN_A<_|8w
zYKr5NWvq#zvjucp1{Rwon2b#?^~hTv6&=<v@X+ux;->Nl87ZAJXAF_F#I_V73pM~4
zB<wynBY_}|=DVkYZ~y(>`+vBr$uAOB&qyDePE9KFyE%6A<fo-eml^<`L^CkeVr19`
zsH&tR%4e*1LFp_2AO;m!3=QwzKfrA?c3a}6b*n$jclO+7-t1yJ=OrE>&4|ljFKD$_
z6qH`O=7$?OC+a)vQc_Zc5eAx>nR#O^E<vrq-wEgLF@y?TDfyS<9+R9KLMW?i1@GPZ
zoXO7=<Mn3b7kxglq(4un+1XjcvrYB&Ouzwr(F`L-`}^x#!t*y4W%#n!tkmZ_s%CF*
zze_Ji6n3xDKc->qfI5ewVsFz+A+%NS(PpQmqrB3XBQ)N;-fHk`=g8{SjK;$qYBu9>
z@0VAq@$2YV7l~6PRX<!XH}`>G>RQf$hK%iDVPVv~7rD7oq`(>Lrm3u~Y&&D?I)sW(
zDKe}P9scd9n3*s7kGRjj6&vKX9@&o<Ei!47fR;DL3~e4d!gasCp;DV2&5xfsb7pjG
zjEeM!hN3%JbB;XN#bWmbNRc6u2dj76f`g(CheQ>1xYyKvKbvuy?a<x5i$uxIZGK!(
zAd6?ugk8-aLfvBQuWYbaxD9(GDpq5!w4VNce>__yJw3e@NjI$|UAl5ADs-2<j#kKa
z+|npH{PveM3EM(_g%~|E|2)mSX@IylrgKTqDIBR^y}F56=K|k>T;-N7=<s$%#`o}X
zVb4!+(PWMB1}Q&knB7uoEC1>fpbOT76>V?$T)~#G>G!`(*rssr;?$IN?z+Z5f5_WF
zsoiz&oxo|+F3FWgJv?;`0#{QW?WSWjzojBhx$*`hcpHv+wE`pfL5LB(dQxo4Zlmv8
zSH9A0QM0(%v0D9MtR-!>d=I1h5Yvls-T|@@-0w5vXE4st<cQfquId@P<+kW+*Pi#B
zfArV?eRuCus+ZNeC_0b#SD$b6l$1=CR@{&GysG@bvc!feP3D|=H~zIJe!(m>w}0Ny
z#aN)|(QSnzaU5OeHP3|)O`_>v{faO3`HMF-@3N(Y{6fc?_@&C^;lrD2DC@c86*Ac@
zt~%-IUD`G3uzf?=ugfV#N4b7A-z)0)IAae}SWh?C(q%Rs)IHIaJc<<N{cA;Qt{>$!
zs84NPcU+;(@$C@r!gsjw3rBcG{&El=k;vT@H+i=%TB3I6=|%-k9WAC4`jO6Jmv1ck
z1<z=+Wp0z9rG?bpRyqmK=%*W9U9(RZE1cp#&ey>)7tP2>PoG_%=OW?DBd;K^ukq^3
z(l7>zg=4QDO=0d)7#VBZ9jGA@Rd>_~<FRd#qdI=1blP}C{4KXkB}M+kMkDi*AveeB
zxV$x4FQnZp*W(e!?0j3}lP)`Yg`VD#?TX$V;bwLCuTuCDaZZUD1#T{0RgOugto57M
zTROVnjT%SnU$jl+dEqCFzWThQ8O&3auU~t=Y^})WkQ5!u+`gN)ad-lEog{r@vH5<w
zrMhQ0AP+UaZ+b^wfbW$@&)C@W5%vG_iTh0HpFX<Z?Uc%I(jv2mceYSPc%(nH%5J}6
zJuS)bhn|3N7fq<Re7?&!wS!ky_t-qW_{nQJnND_>%17FYBraC1R?kZ9$cdp1`AZCH
zrdswmrwCnYzGBF*W)1a>R|~6b{g8vQqN0)PotE`qs)sB5+ms`oh+{ykyn5ia+2<Ix
zgahFZrM!wp?%beXM*c;AncchM!}^CKC+^rKKjm|ZiaxMfAV#?6n)Gw;gNg^1zEIk$
zW_;Gca)!otGF7HoEu*0!l<h@Tb>OOxVV@eKm$B{)QWB*cwAknR`w=Ji*QGx;z{80?
zYCy+q`7@@rO2YWtgpcRD)xyTrL5v-&PeiHRN>tt5!dSlfZkthjA56;gKGQNbq;~Cz
z5$ra=zuG^}4@8y*r_NP-+cYFSTOTIBoRKQ8P!jf*i&amL+Dlg}G;*yxq)q!AuRht@
z4(llvReYh`;lwF_3sTv|%f3k74cEYZRC2M$42Ipo5ZmRH$*uAVe!tmlqI^@XUD}Cm
zJkkWg?iNWnvHf|;)oVr~jPte%TD=j{dzQusw)jkac8?Mi3Exu|E6;n>d6}B723r(5
z4OU4E<c+kSikwmh4L-l-yPCrLtC_b`Qzzuys~A7fy9)2%>Necc&GkM(p7-!yzBb>T
zHn+22Bz&NFx@Ejz;&Z9&CbV2pG0JFg0B8M(9;-&`MXg-sq~Eq0pV+cKKL5rW{?hHY
zgO1}ey}dsjml8ntDOSXEw}JdqTj%;(sY*<G8=!W({Paky&;34grc(Z>Z%eGXp4+QQ
z7*8AB4Sd>2$wx(D;;pjmx>ZbqzQA1g$yNpcSJ4F!vEN76ew%o~fSsBzlJqvGc?2@O
zRlG0|_1pWgcU;>}WztQhmvY<&+`J9--s~)hHQLcS5YZErN_9Zjf+M=@waLF<a#3Ha
zT#i)9o-&^(T=&i<T=!92H`e76p}BUUHJpXO`_dtt@b=bzZg%2UV$7zvOX+%nZ@8?Z
zC$6LwNDha_nLgO4;Joan#tYR=J3<pzR}zD*qmsL}ppEN3xZ#DZ(^M2tX)wr5@Lb8Z
z**8Sib#&TG?%rtmGXt|fUo))R{1NW9(eMuv{jK1?E@G2s8W?d}lcA4&gJy&5={TcA
zarsEQy3^@+I&CyyW5qCk>0hV0*!YLTM;@7sWadUGzPx9y&km`mTY9*mb8oxq{HSVd
zeEc>9#~k*}R?5;qtm6e&y)PY#P;oLjEsTfm@<Y4+#neeAH9myY)9nyuIQ6y`%5iZW
zM`3QA)?#Cmgz319J$rY>#&2Wt_p0C=aDDGNw#LW8%>QSl=-udjb&O$ebxvf`NpiJw
z&b4v2<1A<M0OJ&Opnh}ujZaj$kx1#KzJPJ~^Exjpu}R)n7gpY=mZ>1&HvQ6FiUZ>^
z3**7?%9Y)7KQ7PlPEAc6*4rfCf_DmY`p4$<qlL3gg(doP-}R}ecG3doCju5ib|7Lw
zIlDaaKDF$(3N<{2HwR6`J}zr1tJ!h)d3!X`-3G$anf64=uquAo*k4RTYUxyBQmPL@
zNc~VW5fIGjj7dmJx~=Ncz8=qI%K#pT?hox#_hl+kdGLV<w1ma<5o6(8Pa%h^t7|&y
z9a-owRQhuogdiV;BFYvP7RdnVYo^im>qvQ|oE$XMoX8{G9$8UYCg+qr-lUJ6)Yw-w
zwkCwR=Rp07M21SF{UDNPGH^!PR;yUS>@%Rpe%XI3{&P&>Txnv~7hh1&e6>ZvLTE+R
z#vaAST$<e`MMW3z(fm;=q2@V`v=e`_BOfUDQ+LwaDi#A1ZYPrim9MzgCdk8{zj#4K
zA|4YRy%m+)O=!d`(T&U~EdY3Tq*=Dw=29K!TRviU$1X{SFCn=JgdGk)LQNOmaWm{$
zCk5G~xHK@>2463uf%bCi-f_tt$EEt7PlQHJhD!83|KvaIjQCL)h+rG-$B!SK0|T5R
zjj&7n47!Rg8a6p5v8{aHctqm@SKk*0$yFSn25YAqC>;%#6yxHm09JE-b`M28Ffm+l
z0yo1nGq4c71q=G2qjLbUyz#pz4&wwA{cfYb*7f@030TV&j`hEj8CTpaEivVypr9B!
zbGqi6>%RT#FAPw-cOgA25N;2Pjvo}vAFXC)X7(l%-b-$#Z9V@cpSYgv)XwH+7NNYc
z1J56|tV?1_@h@M-08e_%zWK7knX-@$q1@hQhI406ojTQv>g9w3kol^b8qpt_U%q@H
zwFdT|8OZfwXiuD-!IF~NB&h4~C~?PQTwi1E_SsqP@kqb7DfvMk4hFr8R65oYyk4|w
ztlt5Oy1vf7K0h4x89Aly)du4s5P|)|BvLY>W@;PfL*@qkIH_f6`J49^cKq-j|5lMd
zlOIgucP?0G8ewg6;r8@Xo!srUWX7OQy>Dg&ML!nQ8*b@O+|>65gJCM9u>NmJrWvB`
z{{H@n9*pi2^}wTK5Dqhu#l6MC>pB^=VTYXeMpRtKYs*3z#1Iu<E3jUuDCKfuFjUpm
ziMLCJFZrHUH&<%=<IA4dfy}C0s|pF%gS}JvdL=?;44GF|6Un^9hUHQ`1oks++-+bS
zL*h?4D+2>VF>2+^f5g=24`u+rM#j56u4j?_BTwBJ=f)Ru^x(Nv5%K~*oxXzFcqwui
z8el&Ml48Tc!9PUILbIjE%f-1ZGrhxZbDwlvVk5;h4%a>bFitMjqN%AVHb39BIGej(
zT`$vIqu?<8+oYlA_Oz~DBz1QAnWXPWrVS{zG+43tg4(*VD{>KU>i5wH!p>1=j=P(t
z^`#1~rG;c+nFnuf<jTMxD1ipR@2#qonC@ro^q6^Q=4_u=aXlv|Ckte>_SD?|_m^L*
zog;(Dzf`LI)tQ0vakEF&%t=h&#~L|w*14NWzO4+LU_2m7!_MUdx^R(PeSLi?8miH%
zj|d5zSEAgE&cAWjQ?p5;aVMtZI(-ilX^cc6q+7>jC%?6A`+?LESu=eVpuYC^QHs)O
z$vTZv%mRoqP6Vf9@W^gYG~as?y~9uBVK*x(1q<(>%is;pUFVh54!jslHE)zU^ogX<
z{_5B}fwTL9GiNt30l37EURgHnn$G!E1Bs9GZ-^#obv$?_r)y}~FYQah5c~Wdy&;f6
zohK&Fv`%S0vVOl~Tk`V5Z%5uQ93Fb~(tN^S6Ymq2zWTNP*H8h;U05i_3&!(>B_x>9
zIKe;~t*o*79>xx{_afkZb?H9#ICAAN=Jd9Rl@YDT%z=luhs!$G5cJFVVvW}m-z_Wj
zwX}-oW@A4xeRYg?^oftor{_yeWPd}ZU`c@c49x+}|2PR~VhFmw6<Tcu=j>t*&igRQ
zHWj6%&+h2f_&iT8O;k1A_PIT!L<6@^P?9|r5{45Fz|DrxDgFL$T+D^!8$CD3jwm*u
zSM+1Y1OKr#QCWBkYcY0VVChXmo#w$`j^uCuwtr;95oKrPZ9UySA^Y|iRL3<ZJ<(LZ
zj$s2sJuezGkGDalDn%8Iq~SySg~R-~ed@N$7D1b=(ogjy;s^W>8+S6@PsCh?%_<{0
zoGVGAf#nd$1jWl_2d^=*t(M<c*2EgGq}yAQlNcYHAA6EzN+}+LUIq=e7v?K!7vQWv
zaKH{@h=Sg~`ibrb=}l;Po~&Oei?SBy0Lt(Sm+0;7KG)$cXk~nd_pcYWDHs=Kddq32
zM(t1e`R0$ax}>$p5VqU#q0BT_flDwh1XthsY1yK;Qhn_AD=LL>=#tJfm+$;m75h4h
zL5bvuoG9LIfa%U{@`PS})t`<+@lrA;H`fr0xfOg7N>M^$hijkpOvS*=aWX*#>>MeM
zxK&mJyt!AEHcJLMr`#W$DdN;ANDI`^^WYdxwa@1Vn|{gxgiu4Q%R?A8BWPr0k4Qlx
z^?LFxo#LCCDP1})jZz7_^z^dRTOSN0k??pe8kyB}c6NFbLSoJaiN}9C)(K&idw6&@
z$9U1{g0yEfiXNVaPuvkZaXYIrQB04{|KqUheN_)nk&|~0)?RI)l&been_j(?;<89I
z=<apV#1|t=B@$0E$ehWSWU5<hI$7`@=_|{Qw_yo?{7spW^^q~0+FIkw@n6;6d9r@2
zsA&Gp>%jxv3F+vAnOit@h6^{!pA9naYzQ^|@t`kOWJRT#$;xN!FZ%N6s76v&yl-IB
z9c9AeL{$VtC;#DDxWsy5cC!AYfK)(T_to~JkA(60v*_=wo|?Zg>&yuppu|rRq#>}u
zF-V6Ohra8TJ}f{oPNS<FG6m)=+!bc9F1S-Ze7RK9N8rs4;_z|EWb&oqo`ZQg`1w9*
zg~cn2hc7Fe`a0MV49wAUd5uprS^K=FrkX8QNUio|OuK}vX3xi8uIXHHFx{_M{7Sw*
z^VjgDRBx@lNasUMz2ya%lQPa7x+S@A@Kt2+b=AYi1Iq7q&?S}(NiI2DbKwqvy4u5Q
z|2R+Hak}uO-$DUdB`^Q@i@-2lP_uUU5(QgUHHzmI`Aq(=dy1nSr4$s_dj05On6;$y
zF-Umb)mtv6cyD_L@4AYRyQ@C3@(dqcE4usqrb)_sYGq8#jT?-Y3&Wi=CB&A%3t_#<
z-L(|+yA(8K@LNPlP+Ii+kx`pgRKEJPfsf*$W3+M2*Iz%A|B59}D}O^p51m`4>5GTT
zyj|+Ib;vI%=bC$8K{>mcvis82bK6LriFq&nio6=%hwR$7z`%*!UAFn0C_X&Dam_gf
zlTiB2;q2N~^s9Xh9_*uR4yXK!|4_SE9$^B^<y_FM3Xy?sABXW=o5E9C!PEP`Pd`cG
zZ_n;<v8UU%ThuuxO~~NNV0gnPRxT>wSwrBVSNnV<Z286|BL-b)evF`nRb*@FLwQ_U
zIzGE+^o!kK)PP{r;_?VyAiLj`Oicx6IJ#0^Qpaa$xvThgWIR#9c^s{h2SSEi<|q1h
z769`ufe}D#W~La}2vXg%U;y$|NcJR$$ho_bkhlQKp9fKg`2txV%2;NN-vZMqPomDg
z+@ZkTyf-)L+rI!mPmq3TcIgm$eMWP_W)!)$C2a7C^E6Rq>vwHmY_>&6GtJ-3jM>%g
zebA8Wd`+dj*6GjxG^*vZJFO-u_FVbpR-5g|J(do0-Mck0A{P)M-WP83>9LYa*Ad$E
z9LrR{xmm*0K<cx~)>cl?B{*d-UAp8C1<i?inX%7?R<D;Z6{DB8AuOayb}{-0*^SxW
z6fx>At_{wiL^&CWY#(#I_fBzi^jkqLUw%?*&Q5~0Zx6RUNwO|0kO+&Rwvy@2LBI-(
zl#7dmTQV{-N~uA#2FUrRPnsNdRPE0;?Nw=Ay*gVwgof#8u6F9wCvT4LV8APq|FU?@
zJ|@uyd618)YxgO`-(gW(TSdiUya7WL%qN9H*29g8RuB|=vgO{~TyscAk(ZUI_>YLG
z$$BFfhBivVeoITss-`A30_S4mI!Msl)6a~G#lUxEpL0a5RTvIB2Ei*f$gf^KF31dU
zN-W_|C@9$cSqLQzTk;&!t-rt5{_Wg;e4nRa@TO2P5Ybp1S|L5MAQL;RCw^!9cFAw4
zbt`a^#Qc1j#-=9aN4dzMyxh-bm>fg)LCOq4GNjeqzbqxZTCS(ZxX1M1(Xb%rzV4lN
z@O?kK%WxS(Pd96}lm;mmym5-A>{i6APR_9TR+q|{O{=J=uA!%VKjg90Gk@?vqvP*W
zIXQk?pR{~~Nc4?G>!PlmPtsBr75A5}{~RK<R!!KG*LL)$62qE;J<$r@ysDU8+Pv%U
zemGbj!gxX}*!hFWu`P>t`mSK#5TYUB!6?FVjV0j<udUiS6cIeS4Gt_@LKf26l3&*V
zZe2=AyiJZI>+Vls{9H8*-GXGwBu^bDII6Lw+3Akx`(2(7?EEX=9lg+rW#fIAarh3v
z&J?w}T|Gy$;dJEj%;dD`dk}vqL1OskC4#XgpRNnF1`U8>EVO{xw^v3tTbwP*5F0*Z
z#3BVh{?pt6D=P70HG&r3gz3oXa+B|tE?!>w_azF$vm<a%*!QC%d;!o#H?)Gc8y8U=
z&-+R9zGCj=CXqQ3)pk%{eGx~SSy<c#v#=PVQLnTvBr51hR}4Y>dQeXMU_6}t>qhK2
zpd-8$u4JrxOUoMl-&FHJCsIq+PUnKn8!ICtG;3FP99}{>9was#LP1UI?^8yr50LK|
zz#_#8R4S^_Y_(7cGzdO02)QG)M2|bb+-P~};yjEhGOjVF(m~eB>U~y*bui*K3FwOL
z{;n)>SWQ5g2tNNPz~M)a9XpneDbPmy>tqXih?4njA^#nx9v9<GY$UHoA>rVeGfmz-
zzo&Q>Dt}4wv<IhH3jb8mQ;_igR~0=QBrt2o?+~|`KT?+!AR<J4UfjuDgXkX(R}7#o
zBC{|H1jIik8XQ9n6%}`#&6B{rZz4syGND3gh=D6Y#4UAh@uQ~{GeaA*g0r(T3VHz&
z_K@YQo!8aUV!|@2Mf@;I1Xoi$)_o$@g8y;B?*HsiC8TY_HYUY!D%tg^sVB7-Q!aEL
zK{-z<Cx;7V&^4RivXB4#?$6C75?Womgt7y3Lg`b(<Z%ZM>#y?t(iHWTbgX<~OlJ>C
ztzDSIK99-7r37)4k86{-S=p2$EH0LZFY`Ez-Bbzdu2BZ#9uvji%Qx$n(<SWEY@_(>
zyvG<qm+()jeE+si`Oj}#zTs~%y@i(6Q~sC-Wc8<b7<P60Ab6BOiL$vqSjyk~t<hmE
zrpzO2$!wOCYF+VB0rwp4brL00k7r|VR$b_bu3VVTna|_Gy$+Db>)8t@J4Hj<TN^*W
z>=Ct<3k!irsP{#21FElgPENChxW%Ha@rD}_>h&6nEZXy30cWTDpA=#L`OW=DrQNT)
z|9&YBdI<E8q9*uH->keNVwu5H;o$?nc5Gd?=352*D^zm6J~`{K6|XZ>#bfEj(+3Zp
zk5FoO$;V_vzqr{kHC-5$UQfiba`*q?%`(5+xtZei&S<d-#sAp<w<i#lXltib{)k2R
z!pZ*3q@<dSCoIV*rN|fHbL^^{CGJS|_mr_KD;kN!C*AIjK}?sx`nQ`=C?Mf>twY<*
z=Oeor$qN(H(W$N|T8P4Hweuxo(yhK_lN#h(*dpSqp+^<P*@O)bD(dO=i#-&(c4y|<
z%4HNa4eTYMZ^#buAJ0n}&;K;+0^{NJ@n4k*&;``77#JzbhYufs-fTh<^cElwe<HF?
zy^T2Qe!^iKt+aNO@t@a@fy+`T#s@rP4}dj!6m;2%tnBQ4kXtD0jYGIyvV=+Z7c|ZF
zLlKuLLlLKX`-}gItnS1l1LV0@)hgRdpiX0kWK4mijj^d^CK-3ZwBU+KOA~?;Pp$Xl
zKF|5d$oX3t6_QCc##er|E))W<eW%?D<CMZe{WDsg?`I2>bkzID70=jTw7Y>n@q#gr
z2(v34e2<AI?o&^I)g!J@RGeu88XZzkQuI*M2$nR^B827afv^mHxa_v4u*CX6^S$A3
z1?uACY0}w3-%DklUi7@OIXWdpz{%O!5KK#ys$X}awu?Q_H;Z`Y@SKq<qfnN+sVMGO
zFOs|5oF$A5`&6$8co>XW?bY=4Ie`V;hGlZ;t5@>P@X!~&+>)oAZEd_u`?oEi^RLsN
zc*A6X`04}>1;N5ZdiX7OydWFJgNre$w6djzqpz=zyMoYji{#QkLv=7Kb+EIw{T44i
zTq?6ewf~5*&HVd&(<j}U@4Yu6){Z>8iAgRpkR`)t_PSekG<t0bDCkU_Q<Yu@sBA0e
zk15Mdoe<r-w{&2xe_&Jkn?|d=lm_?Vdy&Q1512p?u!7$GOwN<#{`RV#(g<@;Xla=~
zc4GNdcYm#=o?~^3#&)_KdBpW>(3SbLw6vfSQYAbeI~glC6-%~QA8oxp=EktuTiYYN
z!B{9E<O1+~xKkZbY#}~JL;CYS^{vVTUKT@C(X~$F3>(!(6)Yb2*`Hb9=C)&l72JQ{
zH~2-Wv{}#ldt<3iD=L>fJ746P)#k~@#ntY~cFDVNI=7H*B<Heg&zT|MVkJ;vpDrIZ
zr#ieak()Uy05DAaa?9q&OE*+F^s6WeUUY?bgg-`l_F@f@vyC~!Q7QbYjk(?1<eA<c
z5z61t@yO;lMo90+xr=$|Nfet3-!Zjjr=FA{Q};2`62UYRUsH>*qO_-JX#pf=M{}gw
zn7Y$PaNz|I3zZTiQ_KCQ)@?nGThHyj??0mz@!Y<)OyuhJuQxbn&XHCMu@z_y?^J9B
zRsbMOznp&XR0GDw#>_5V64drnyMX%LhQj&j!pp8dMPgEJys2!_5Tu)0EDbS~vg?4%
zfTJp`5B35wdGtBg{_-QZ6~(4Hf=WaqPm@MY1`$T->l@2x0P19ZFdwJ}38@4Z<`ppB
z+pr$7jN#c&U_*hldNmJ0gfNta2EjUe@S4e!cYs`#qzjCULwNpF5RJ&VOh5Pfx<T*}
zv!0#T@HQ?CB!1qqWN4ZD;IbRwkg#mVA6@XWzFI=2P~KBG8ZWduRXaU~IleXsMLLT?
zpw;X%uh+Wmo`6As(ctw}g3K=X9cA4N@RMG*PNlT6E}R2&;IEMEobowoL3QYV*XA}E
z?Cv<&imJ?Q?X5fh^gx+A7fqI%%C1MxRHH{|ML;-QC8n|?q2AsTKlT=~!0GAKt)wJ&
zugf1~{}L>Dy)<cEh?8beOESru>P{kTGNRKxo4$zcqfE3l2HB0@?=CoeIpHsHK+f<l
zsY|{^0E7)e<0*0ntA`(imYCF3K`=On+Irmj>fHqJ30PftlrZ}0%^G~ODi%u!H1s+#
z-E>scol3XM7fBFolq9lkg%H!RFfw8Rj(xF57O|PuLP=BAZTOy5ljX~ePjx(QeQJ3<
zw4(CLBEh`DHxgJUX$Zijf7lUZesp}C=_zn#93r*YjM)pST*C>gAIh@BcNqulu3n{r
zO5I>MH-`YR9j&?qjc!pKfgjV7gzV_lR(Nt;@c;g-Oz^h1cbaRTMQlr%kqBAkJ|p`S
zz9z3!i68E|<|!?qlXetWHuH6voHvwDTM4SSPJQVe{YI%@qP7a90(1Wiz3@yaxB&^{
z-QC^os&n5|7fOeuu)MgI?pwgvn#E>sA68UUd`96nWFQ<5zzWtIR8%HTe_N=?DW@c{
zVL0rvV}w062>|~cb?iUi9RC~p=>Hu0KTqKQohR^mewnP0;K5I#0?$EOrl=?#S4=%{
H?)v`%N(v2P

literal 0
HcmV?d00001

diff --git a/examples_skyline/accuracy_vs_fpr-skyline-order.png b/examples_skyline/accuracy_vs_fpr-skyline-order.png
new file mode 100644
index 0000000000000000000000000000000000000000..ba45fb6549543282de4c380b51f379ea94c80510
GIT binary patch
literal 43619
zcmeFZ1yq(@yDfYJDi$^d0v4!%fGDA~iGYMONT`Sc(hY)OU{Ojdse+`SBAu@RN(hqD
zA|)UoEq&(VeZSiK>~r@1$N0xNXN>)h@qNJKbJx1px~@6rHLrC~QC^B+In#2Aq8Ox4
z9ao|#+6szV9JzD}KC!Q~slty%)<>mPmf}C>r5Eqw?`0OJG_5IW#X9mon#$&DGx+e3
z%?S+~W%Elm_UEk(DAV&c7AEF4CPo*w+Zk9{8=0FO<T=2@w{N?ljg5scFYiD8oyXkj
zGVilt;RcG@PDvj>qT&$xv)NHw&0ubRNXX;q(i>-a5}(|bp*7YpW;f{d)JwH#`t<tI
z_wLj6)o+SC^;bmuMJta7KjmW?+Zbx47h@AueflzUe}eBN_U%7{FYw$=H>n$3o!8>X
zwUqvPNavWdsB=?jpLTa>xl5^s{bk#Og_~(`_pxj2)i>cMRn5N05I+iNY5DQv91RT<
ze!O&}81O?w(v6JDzU`89__1+KBsp@$Qj8bnw)p>!!+2ggPg*Try!ca5kyN(*U{YFt
z)br<@`}bE2TU9-mtD2o?<&`cjE%o>D@wtEhetcFA=UOJFc(1KOSy?M7hNXB~?QcCX
z5w)R*t^Gqn5?IBYGO|LiEMK)s&CD!GR8&-3XE{~99M5X&&3E#{hYv|S?O7KuUfjEL
z=ToJUjT<+rSqj)5Q*|A@owbcdTY{F~O(sO#B}>@u=UzQMJ=?xoxso7W`_k!eH&<7u
zUQsW~nV;#F@QvCf<!#vU@u6|mb1$t_<LDyRtEYWO*y()z!^3%mgoLC%-t+QOJ$m%U
z)L7phb8~Z-`PqrbLgwN6ThE_A?}R_Xde3HBML&J^tR~0N+V<zyGg}2sDx|i#MCs-_
z9Ui;d_afb9Fn1zwQq2D6*C%%U4Tp8!l5sk5@}!Kf-9S^^S+mR&Wg$Z9jcKN1*{!ol
zrM5ab4qq$}A6HTFC-)xLao|>+o~urCmaVeKM()vyW2--zrOQtB9Pr$7@KUT!ww=?D
zCwG#x2D5DYs$ZNHnw*;A+`e7%;N{QzY^TlhXVdWcbMCWgv8Ih_@~uYouMIHxS;mhl
z=XmCR-I0jZ%fB`}HdYgP<fhGVhqTl@Z~lyu%k+3aet!Pbr%(MKKi2A+dl03XpvJLt
z=h3;jx%*yTHMpPt;7xdWxP!e14k(p97EnHa{>erj?brto)}$KO$lSYkPri?nhsSQl
zzCK=+%WKQQlQHsPR#U9j-Jex0EM5-eO%GhDT}&AqpcJKD=cilc?CjDwI5`h+aBvhh
zHa0SgIaSZh4!R1l2;1xzu^-^a`|hfEa%{JlSo4OB8y|oCcxvSRdWGPoCT;7sw~JE@
zN;c|y^105mZ<Ev34%KhIvwEAL@985)uJaWJ&&^JUL`H5;F>e)DR#ZGUJTij&y;7Rq
zdhN=64-YwG<9MTnH*xi;SAy_m77@GPuCkC21@X!E8yFcMB_{408LQLVn%!)s^~Ugg
zWo0FE<9%;$!ELTHLjw6%G&F)IJNy)wtDc?oI*^<{mxWU_nq`;m(|B31S-&*Xy1R6|
zzp<p*-naD0u{+FFJ9g~Yy>sW=i<d4v4h?0lN;lW@d+^|#C*P%!uCl^+@7{?{|4b<~
zOApD($w{cKOVSY;ne2|(s_%K&w%1=@!lWgqe$k>uoAo^%d!mC*`>=|Yu9ld=;m@zs
zC0)V{@wRQUZ*{rURr;uKb9cTE-tpVFZ%hBUNiV0X8{X5JpYJ%3+3m?!c+tS%Q6Tr(
z$EjE99zT7$bL7*Wh+zCJoaPvPlYTW5=J~=$Gc(Qy^$S;4z0#Mm9{PTw=-R#e_YXXJ
z^k_FPZ~4#G{8s)ewWZGE^_R&V?}>1(bZ~G8R*;yj#*_c7Pbqtm-eT;{eE7moyX<>D
zK0@t<Ue;5r>-Wg>XJ%$b4@gn5OR-2f*E2F2s3mHgU&6q2@ltuXOU(>#-tV_>%$gh4
zzUNVYNX_1nm}z_5m_N~Sm8H=O=TQ(HIvzK>TOr&;*Lmm!Ui<WT|3S=5+=ZTX^Ri{j
zx}Mx&Jy5vWdi<wmdw;)aWmT0IUtyDF5zD3C>bS1=_tw{G7j??4sBoPu(POT{&<S8I
z7Y6&TX^K6w3)9K{K;5Pd8!on79Tc<fD%}z@j`bO`ibXh>JmY9zU3sXuO9P+z)M(G8
zx3`uRZq~Q5vO0|rSN7$Zlowwo{SS3r#}BW<@h0k4vWk9v^?7<Qzcudmk()~%O^$R2
zX-$;y=9`@KVeJp*!Hg#}JVeOc?q?8Yi*Q%|Xr=6>pI={j_h4>U;fAE#1}dXH>N6~#
z>Ccf1#KgpK?A>eO>@%1*om4uzlw$mYotfe^{o}@BOw2&6+nL$fR800TQAZ^yZ)QvF
z#}=80a(mg>j)#Yb2Q=s^Dn6(hMLe@@G)<2)C<z>%n5aXPU=CX6uX>YB?Dw~uS~(7{
zJo&UTES?m4ZrR1ln|#IWDjroU&Eye}ZcZ$YiMw1LzI@rThiqkGBI>QqBVJiHJx_y!
zH^0zK>B`F2J9X-=H>+50uo=SYV}7HZ(oX%Ubsg>PyS8oHCKoQM>HEf}N9o1cba^a@
zUd+oh^VR{4@zI_t85x<%6}&myB_t%Irjas|YZBD^W+q3brd#KeL<TaB^*5#`E57J&
zNVT5%+28*HgCxyfXXiTlMap}AH+gc~J~b@j(`VC6EUR65rUsu$`3R3cW6R9W=GwdW
zRDFGY^gs~#Oit)Z?dca<saB^iU%Y7e{_a|DQ~Uk~RaRD3;cGJvaeZ`d#j#E1d=!K3
z$&<HP9Xr;VHoTE`nHqg~=gu93djroi7)OM$?1aE&%a<!%y7YYciWSMrzmQ;v)uum{
zKR-LDT9TcadH7nM!_d=fLmqmDcu859$TfJc&6pnt3J;~JYBBiQkT+r3UfA&3z+)q?
zZsG@zO+nKx*@)bTr!VyKbv-um*4|pSo_R#reju@<<09jxO;-@gnFNgeG5M+!HB+qW
zu9CNW=-TJfQk6TaSY&1gu#TRRB@$zxf8oL#t4<ll4I2zUhaP4+c*#@Wg<I=Q#nTfW
z{Fh2MbL-`4&rXf`u`OP-C{`gtoV;fe%=PA{>H>SY<?x`p<wIjvi~o+So(#G%&W<@-
z(UN<WrK)f<%VDeIG&D5cX(jnru@J+O9)2z<QI?nYt{Rt^`E{(wK1bWhX-bt0=eMQE
z$%kF%a+;c&s-Mekm9ETj9C?G8DqZ<*wM62hgO`mF@(vXKXf)4P#h^P)47@<}>_Z5w
z!@@C0Z%j396qHu7v3VmtKVxlh=~8d$qk~SfV|B@yvxrqloM-To9%6;mChH}W5Y9KR
zo<9|UfF2_kA{dvEp@H`=Z1K%ae~xaoXe3GYUQ7a&{pqc`PX<NLzg;vtS?*ff9pRdU
zl_t`#iCZfQ@#<r9vku~muZ(w;*4RMyASV}>EV7x>g$qv$y;;sa=|oB~NU66$tPn7+
zK80wJZ#@0#)vMR_CQOH|qLAf#3|+rnryq@xn3cXdF;MgI;V$kv=V)1})cGMF*Vx$D
zQ+C-padC0Q#l@Vd^TB-jZ|wTD$aDjmavCWU>g?={&U2X#Y;diPJ)@O!$t%sQIUWxz
z7baZOb8nyD9tAZZlmiD2uu1Qc53MK;;7XppN>k7Yc*q%jHr>prH!h|+MqwL<PYu_%
zWggoq=JaOWx^?_`WB`U<OyjCJ<(Q7PwkKr0k?}{8rlq4xz4&RJbY;PHIsv<%8o(^^
z>dCrH!Ztqy(g*wc5|BdA>KA&R0WNvtFm!=on|KpnzWrbe0Z8G_!_t$JlX0%|bKZ*V
z(ip~SJVI#r%$e|^_M*O&a`7sxfETqd&Wbb$nS71Bdw1#N@83Vt^JiZuDJfOIF{)ag
zJmI@+{cgjG$Rk);%0=ERV+aCyZ%cTzGZddZd1Bm>lSyK9y-8Za)Jo<<ircRJesY^(
zYnKCO@!hq%n1jR|u3kNxG%0iB$Pq1IG~KKH0;xt-C-J9KyO&V0^Ure*4jCKqjyaT{
z#{wW~<vOLRCF{N@^x7KKAOW~TBJA$d!I7AR2`4-@%Jbfz>^FB!XnS|(3T}$PCe!rh
z=ZH4i6BD#<C7F+8yG3XsX)-O^ZkT1XEt&0>m_O6~IZQ{#`{=P_V|Dp+suNAt3SQ>f
z`bFLZJL<ePrD8n}(K19HnJUqnL;MWZS{tro)KeL4Wo_O5=p#sg&{|qU3mk6X`Q<(%
zS7^hoojXH$Mlw1O8`S}IqD{Y=2+rouPs}G7uU@w73&73`Ms2wZ`uYKZfzRDnY*Ooe
zY)To%X51pY;icQLRsR<;hW{sM&41tI|C@C5Kl|Q#6Q%uj_e|u^=cbv|KmGN~LOqC%
z;u#x>nAdilZfw=oIT;POUJO!u;>3w0p2e|iX!(0{2o#}T&1z%2rN#`{J*zJ%yQ8~1
zrlO+4YT6u;TPMrrVS0M{Kvo}*!NrRzt`8~gOGvR!zy6?4zQXbsAJ#j;7{JZXud-Fh
z%;q#bJw2C@kUH>3O`d5a=bFef$&4E}W@X)y;dn7{sqAr2Q2tDGi$lAYRgi?^aMVd}
z<~ZQY6G%jr-04kEU+EXANySmdr*v4#-KOqRUc&kl1I@klDKXA2E+l%__uEM;f$Siv
zv&<r(2mslWh_0D(iF4n+>L9&qaR|`Gt*v@z&YY2vld~Q50o5CHoxeKP->5DwF3u!q
z5)81hVbi7<Bvh{5yHED?^mxzdIS;vCxNw0yJ@B8)(1%UFQS7pTKLXF1S;jloCTcQ?
zJ7=)<eE;C#G}RNs$jqFWX4+^x-0>+3VY12|+3t2j>Xn!~tWH1T2V1l4&E-N5odF>4
z97q7pB&-u?P0eq&9KUG;zsp#SI`X&V2RlJGt+wJJlfK#)b<8DmQ+4^m&k5eJ1%xBW
z=Gz}P!#*GFZ_U@EqhF;`63AVXW_mVvyx~g2%kwu$Cg9e5eH;%($nYgoDIeCFUv2Md
zUK_~T_Et-8bN+Q*zXd6X-?)0O7mJ8mUS1v<$-JpvF6l}Th=50r;?BRjO_UDTsj7&-
zw1T2}HvjYiKK;TU*XdawBqt~D+_`fU12~;Ee(>BQnqD`cxX0hVog1k=YyJr7d*qAM
zwk-+@3WNd*nzxwH($X@!96NgSG^qcl(o%j$N5|b<TqSQz>MIj<vTF$)C76c%J@f0`
zY7md9i9vIS1crID<7cg>MvbZ<Y*b}hUmk&u^7zIL+Jlzgj~_IuWCKY$eE4u-YilbZ
zrW-bF2nNF3>gwuh4bE>pH{;me-mafF*|k*)Q_0FhT3%kTaPtDe##255QzcyLpv@1(
zk(rTl(9!xF$CRO8zxJ1lPba==><1rAE_E3q7k%*H!IgpJqHV5Cr^7|-<G4X~oo9xM
z7+ZTN?SbK;#GU%}^xK>X<Na`-@j3F~Zf@=}azlhO0sj|)JJzRMmdm#5F9o<OLk1&%
zmtlGudGdAnxXwJg=i!l{XlmbNd1rDg+0_(i;GErHi>X$+nW>;@Lx+YB@ZaQ*XFjK|
zUCVPsqCahD82hQb+}=5vaMM>w@9wPLD*Ze%GV*BO`0(%n!vEae-MfG?h0<G``o;!~
z3)Xo{K9m1FT*9lWs2I5?+~FueiLqy%ZCJlv`O-X|P8%S?7^(q54?B+CHvITthn2m3
z9e|Q;33s|7q2hC^#YP_x1qW1S7E+Ts0E}(9*u=-GDrL;*+JWY5c4=H(6$n89@5#~0
z$#~}XP%p?HKnOQDc=Of4Tp4-!KsLbbQ}nBtD=<YtJCu}s5e}lk*Nf}w)CtH3U}6@?
z$j<JI4$}1v3JQ8GXzHXhO`^%QYuDsW@qeHVO|~yKNtR{vhpS%Pm)(&&(NYDT6`7=y
zU5pzdXBq!l|IE{KB^caE(5otC1+nPea*sda8W@&@=*OV)Paz^1ff5>5M%@Dk8pABS
z0!rg8p6fJ5#s~0p*X)3OzF)Ih9p0EGdmm<Iqrk<FYaAUM48Om>=l!fvQf>zZ_mI<Q
zEarPhS67$eie<|<Qd~hY+PiUm7f?AcQu~?uEZYQc7Ll$WRam-3<pA)x#==5Eb!UnK
z8X1-?I|Idt%?Pa2o5}_FBy?(ce7yGF`aS+ZK`}&qGyYv)e(gDs=b>5z+^qsemB)QY
z8?GdiWs9Z%Sk%$#gMIb<(9qBqOtG+r&*5TRkP1usXuoL%6{g(xGzfyexBy9?=^=z*
z>~T#p47e>_bvZ;0TjrG+9mB7F4QECB(k;4v{kpKzs9`{2He@;Ph}_&<aF)!I%mVij
zj-NnPEB^dBkgxDd;G;*Hm*z`u+WkeWs$vvG=0YW0b;~8@b2^2AuPth-_8mBQka5kL
z?b1w#EF$GY4?PjPI==VtVa@fsPd{KwHEY(wfYbo)u?RpgEnHigoH+Smlb#B+Qeek1
z0AtOhgIIk%cT2&Y#xOhxrEiC`sEpd^SAP_1Y}~LRYNFM(HEs|~tQtZTxJD0ziu|s$
z;;JXd=);6<i+d7lJN2KaSg}p|e*b>Kz`!8x*qv3KW-&YamS}sCM*>$I9v!U)UwDCN
z1C{YKw7SHK*En%LLDEE!<K*P@Z8lR!=6U`4^}aejOa>Ql8X_?O))Rse9UX1-qvDBg
z)JZQUqtPE<oVp%ew&|_rU;w69`t&78iq-Kr0$<d%-);2LN_YixU`;9xjnD*NH}-Sx
z-TQ#=BnTAPM)Wuf--TO%>oL)?LB(e2eVsviY8Nj?L7ccSDMA;^lIUpwwp81foS!s1
zJt6B0HvG&-eCm`hKY|;P4N@+b*CICB4m2I^|N82f(x>{2(NP=FZM~F0h=?Q=^5*=y
z0ZD6a!gXF9a;B}*q86+oMCAj>*Tz^y!^6XsAO>Au<u!NY{e7%J?+LH&DvwCU(l!3%
z%MNmQ(LIo1`SMCIwlPf1IK0y3_;agy&Y1z;yOnRCq+Y2_{1RfG7pL|@t4G0g$`1+k
z86izmr-Sz&0dT1P@#R@360u<f5~;9lZxrN5IZ)6#v56-Ap2?aB*SSo@gLp)#y$25}
zlSu=umw5y{;$liYTY&be*N|e^+1O&ia2j6e->Dj_rFvPp^y|T)Mw_h5jhZqn3Dq*o
zo4iD%2;UAcKSWZM1odP!gz0MJ8`G8?O<#UYeQ9=mo$_$8WK9JttJe_TYA}<$HuD|j
z%^mfH^vGH?gP5i+8@O+QGg`DAC36PB)t4>z*V`3N$nKqkgUL_=v~-2O8Jl$tTYsmh
zNcpUh&mtt)^dYfY*mB@`9+Q-ZME|iSnyR_63w4lcOXSO`x#>Y$#QdzR4UtE-O9owd
zcbj4P>ecE%N&L`m61s@iQ{YB(-_ujY#Ka^zDS9!buAi6y>exmPHB=t6qK?P`(S3QN
z{E=MqCiSnQWzy-Y1+?E<x<e{$wyBbv8m&@b+`Ks!aEE=*o&*qVZ*daQp-+#Yy3iXj
zKSK~w3>H?_OJnNg?N_TPPM=*B>R&UhW5_Fd`EtL2S(63?(JH+8u&Jo%=pN__fo#pT
zbvo*0^llB_3x`lH?mvH;#N5fJU$R74EJw!|iaeQT(v`!VC1i?rc63x5hEWz`oBOE9
zVbPMo8(d|fhn0QA#)HW`6B*1=hBh<&>sLm+9?}N>W0Rhq7*w~gu#gMViUE`>X%3r@
z$=*2run*_@(ecSIiRanL{k5vw=D34=TspKEj1m)8(vC7BR)@BP`v(Nr)-?JcKz%#J
z>4zc>S!S)R^YOMHii)_z#dUo*u(F!<Jj!P{gY^y)s-tuAMWMJ=E~*?KYS>kl6$;mV
zZ?&5IDwU+FSD4lNcD_Mp2Y@8>s!FixfQX1>fBy|cM~Ex`<hO@qb&@O8oiT9ZO2%kf
z$!NY>8p+PJk;&hBdg2aScNzZr_D63+Y63FSGw279Lr|x9;j1$2`k}3@Evx5wh|SDz
zOU&(zjK9BzqOUb;<gWz!RM~dwZ_8{U(3#X9@YuB%5)(H6_1K=ab%g>IZitDCA<7_r
z*?$^`#59UhK=#E(206gd4VVE4Seck<g_%x?ewzo4m+R|vqB6GvfsQPaiV8(2-8H|*
z%fm$W=;^gGwVEB%_dEk_3}0{%IKafF?qj>IT}SR*VuEPW^(FgWHuf9W#0QjIdlGpm
zBQrA&Ps1ZNwIpMmzRtS>4Ryywh6OvXJ>g$DBlhe`d#mKTf>vu>@49edS(%)I*qFb*
z6O@0lI6B9EHfO7U`SJyduo^NKSrVMW!dgUrU{ss-tFBgIU=ddF<#(NP790PnUzhtC
z@{chlq`_ysR~Ot-=HP7FCfWZ5HLb#|`F9%X(-sW;ibW2Nj--O*)aJga8j7(@at!38
zhLp=bY(|w)(h!Eoh^VR>)ZRt*pL{%`o~ZEzb&q%~5ysxtk>lqQr<dvL^uJqc7|B#`
zqGTA`F`zqal%g{)EG8z#xNTc9$cT;rub7w)vU&Q>UAqL%6;ROmWH=|Ihkm#FB$&76
zA!$?*1lBdBudmPI!y^nb1UkbWz%0E5YTLE(l!!QlWhc}nSW29SKkezQit%w8{}~5K
zk3b^dH|O6ivF)imfqZI{^JPE_n2)>Jd890BvMWU3>bQy2G*@yCA%e}<ru&&BuH{_%
z9NP0+Hz==Uq+BA|tT~GVB`*ZPH+jFmuYiCoQ!YB<Nzf2-?BTI_$UH=>L)w$cy*g0~
zQNQ@bYRR6>eVhB3n7k$FYTjS+``#X%ljGat>FH}EecvNH&sJjU$1}tv8>X-9M#cUd
zgeES866m7U?LHDoj9hA`e1STiLo@FM$^@r!i1sc(m4%dA1A+P8zI!KNP`p9Fv_T~~
zZz^8UtSJ^Kk$^QP)F4NHrj++Slw$oNF+Zt5C=QT!ZdVor?`X%7E+arw$gZFzs`GQR
zsu~)xSl7J?$$3hsppl^?Xo@?;&+l1hNl+gy4}?+GQx!vEHI{9V(;E&62|b*$#+O|d
z82K5pXuMI?^S}vus!+Hv>-N1}6_Ntdr+)WrDrR?dWbdy_z*U`yBs$(`E`nl3p3VZS
zA=D3Z#kMpX(p$&CKoUl~=l4*1K*N9=4IQDFK%}wwM)Ajw6@;<AdBZ=}SEqoPKGoPd
z_Z;w%<&cG>P>aJeAl`kb-!{9>&(>*AAY%$#c)wHtl0RXxm?wnn1I_jTD5+vKll6N3
z-BOxkGJo>|+}26`8U6U$Sb`pz@_?3kU1_UWMPq?32=`U`G&efh;Xtdn#naZ)mSHUo
zQ|#_}`;)KU_N$2vy<h)W#6IbBgoNIcIbB^{?}+3vS|u<&;0W1DRUg4DXBO~$&~AUe
zli;V-;bK?cTrOuNKxm*TQ!TI9b_B5MbC?M9<h=xtOH9|78{14v9xK#Px8B@myqcC@
zQSQdg7pZ3nXGM9n(1;pDWduxsO<FKSj<BQ7vD>$A6Gh$ErmT;iH?u#j@eTBIm+Ag=
z->8x9@;<<dGeJB$V6RCSZh2q-&`{^HF;XbgLJR<ku>mtAwE=QA)Qe)E4VRdulSKqn
zl4)Fz;!D_J>yyWBul#}v#AuDWp0qF22HD&eUX<%GsjBkl#(1UKc}@22+qZ8bn>K?<
zpG9Bi*m1Wd@zNKb+H}?nm=F4n#}8L4EA$}uc=y=1I7&fIQ3hrs1qjU6^1!-D<l8=8
z*GUhj(I0JZ23~`hpTzywVPY_D*%Bx2GW7#<l5i3vxKWAu5eY)pi2@tbNz!0Jfn-s{
zR<n#Fc+X0MExEQ>5WQ`8wyB|jN@gpjFc+3IxmZo^ReRvb!!P=pGObBGLkzOTfZ>Q1
zUp9CZ<jr%2Zj<}?-ML#Df7fEh^JCpX8#4M<a6RV~70X7Z{E|<RC!2s3l6aofHi9iD
z$nI6w(s~>{=dp3i`gIpPL(YEcetM!VeHKeaw{>}MEJD^6)Ok@I>8VKwLN!|5%Ow3x
z{veiK&BwGp&cdwwasBOK2}Q#COkEB|8tV2I?1vS-zJE`e=dn2j4kIHgOK8{4)#9<3
zPFY!ORM7dv1SU^gTY4HM$rtJ1Na{fpRpb3CY2X{v|I%pE`v8>xgp?E)QU<7?vZSP&
zzOaNuzD#-m<_e~brZ0>E@rdlY6T+>K;JGzkMwU*Zf-&{ogO_BES${vjb_WANeHP<g
z;cscGcOXty;<nhN!NmOF013Y~Wwc3M)e>ktuWJVR2;zS%;s8k#*GLkCNVFSdtWdaG
zhWYxE_=l#+a!*o@b<X)_AH961IIT{tvoV<8$hYc>p^LLKTfoW+6kJF-jMO1BG7fFg
zXOjjvRYE+&tmS2371h9ew0;8-4>kiC%y+ZQMtOO8HHc)kpd)>m-Qm`6kf_y#nr-|@
z<%zHk0;HmRD<rJDPD@u3mh)HiN1{}Aa>^pr2bB9{5q?PJPI(@F3pM0eAK<j=l*<Y+
z>83Omd?9kn83LM`WyMWNq!`;O6?xtGLgN+XnO=uZ#tv!U1D`MMy=I%k#*uWjEXVR3
zZm1#OH6Iu&^4jc_5qu$8W3^{VY=5Dhsf4P~7fd<-2M;1q`d1_x19JW|g7lDvN(Jg%
zMt__E^tXnl76iw_6I7jzAhq;1XQ%5rb~2Kh#6w6^FkqNr&g!JjKy<PF{p&oya}Rzc
zwP38oFmI@BZ$5pI;Ng)cr1Yfc<{jD1N#DPJk8jqbIE(&FVZN{Aw`^grQu4~xaeuW+
zz>8u~@w+9r%=?$KwC<7cb%N~G8g7s|(jhsrOJk5^6cmC`pC<h5Ivu@?FT^(#g|3E)
zyJ*EIMAU}aw{S>5K4ke6)jPeeta+$A9D;&sq@<^E;D_tnWVde=2rrQZeWQ@gNdU_<
zo`e?!Ot6R4uv{jZ?aKx&fs73<U%rBxu5T0)hDJRD6|a2stwP(RFVgNf)A9;W+gPJT
zhLMjihJSift*eQ;xc9CNIE%c{S74mO*T#SSQiP+$lkfiByA`<Kuv*A#J)AERM4{N!
zkOCoQ6&D}h8Q`B6U;61{cRn)kG%`q4UxyjL9#}43`la5>=<(v^3=B13hiuXSrpTs$
z2u!1V+y@EKTaOdqlDugYtw|z9W|#3b>C}Gr6&k1u&<g=J7$HREL_9UgIT0@7kFL!O
zNfDlj8j(8pS#9k^6n%drWe-%s4n!)Don2i-7J{gh9iNXEO^SFIJ}j561gj0O?_A4E
zC9Tyi3|3RS%Fb8&L~sWH37e(l)10l2BB_o-!)o8%-g7p8-sL0u6yn}#NWl$xF4^$Q
z*p<zZa0ld-*d2aV9{&JrX(uXg)RN=q1j&Yxg{41ZS+_3;CI>uULwf+?plj6W*f&SV
z#_k>Kt&v6HI~hD;6fq(mRl?r7<Ol$_&hGB+H%WwAAyxOb<YptfJ|O@PC08Oj2lDC-
zWW4IKTCs^|L1@71CM+Dqb=QP*gESdt_wM<ok7(>6nS}Gb#vQE>|54<fvwyhk5#xoX
z=b4WlXzsZ5Wu7N1sXqcRFe@;b0|N#>9x({rUd3VzhTir4K4;9I0XQ?#H|yrBqzlH=
z4wr2c5B_rq@7vnv8Mn0^S?%2i0f3zQeXQ#GLMv-`6Nw*%L<m3vs9;)|hten(efUuG
zmEmS8An>~rZc7w9dBn;$P`w8MkUXyZ&Cus|qU-)vc}0n#3D4W!u5=7^uH^y8+?MP!
zM0$hzzS*Mpu)ZgLakxjrv=2vKf0Z<xv4rX6_Wk=6H8nMbEx+HAO*2n#-15Wt6!IOj
zxgZ@A(_cIT`GA?0|J)Inm23z}Hkn4q?)~!YPD$G8$nzFs8hiL!p4hdoBQ?VRS6ztz
z*|HBF@&B~L(rt<Hzmq56X~1GuBFF^L21V)2NEs^D1(o~?ijk4A2PK0VKrvF>g$)FN
z@$i@<R1*QXIQm<1&%!=8Rc3yz{i!icq#f3_B;6b9Zg~RuQBg5&%6Nq-856*zE<9!l
z2oKehUHbCp&@<rQ+GjYaYT@uC9*R;Nd7Y91uSiGms+B9N!8$^xPKOA}-CV*@i9GWR
z(CZ2kpn%S9s#*qiOvyN3KR<}8Y+_^8$|PD4uL#7>PtDDf9R`1s9VF&{KT}dx)=@NV
z_6-RsxyDI9y)<<TOrK1VZf+D@B=7mc0s^A(dyF`+Pz?(CYzuZ9U>ZmPw(l5(4vl&E
z_^3NOBUJ)kBH!1XbryT@_;K-HC#xgRO8^~I${uLt#pUJdGBWpXuiiFLz30#&jSxXo
zUp5yA3Q!G+-;=9&v744V?6nLJP%<EC6i&GW1bAYQLG6a$wpWGlJKxc+5c6IDD@7D$
z3VpUELr^U#D=YgDK<CzHxs0CTS0oVl!H#304Z7DR4T{R4|3kEHG|t*b#g3uAkID95
zG4N(>fQeKQ<^&=ZfoJInNI<%?Z?+90u@M0fx#II;Ra&f4#-$bNdpS8Pp<>2^32||A
z%YFLv30Tm&cLWU1xHOorA@^!Jc)@5$>rCN^8XOe)91_aQ|H8bWC~Y@&N4*^)B4<(e
z@a_SihB*2zuAG#}3@f(hLY@-_as}zss4^h8bz*5n06^qIX*Dh;yb>vt)Hq<?QWzNc
zOh4-9>+A3D&n7)JRJ1MN@#Eeb#(myk6XRzNf0d%B3|5+{<q)oftvc@N3&U-L-{oDp
zK8GbRY!y7^i%h%=U<A;y_x4sZi{4ET4u1c3kXFKoL32t-|1HcTAQ0f9vTA%SMI{vp
zRH)y*ckc_}stte+@(mX|`w3Km?0eUv8i|A@?VC3@bp-`d*h2~#JIN5;{@<!hY`M@2
zrl9wsE*A|;%aysAN$_H8%TB!Q*k#&p&)om3G9+&bQ|;^aZh%Ju=LB>%;v2>^&;s!D
z{~s&2WamH0E16k4DJniJCSe7oD230RbNqwU4}?8DbAVgg(lQAjyBQa*HKYRk9{f^v
z$YbM8j$)zQoIELpNUH+}U!>AeiVBdmROBsQrrJ?d?=sjnmy29$dfwl7EZ}i(Qon0i
z%}|7p)xfuYZuZ?-&ky10VCR-}XO=Ll`s3zMmIL20-!CW1nOU40<cf;Ye4Co09<1Tu
zxVLpF4K>E9&_*flXmyQ<{C=T3WQj-f$k?p8OSNEUlt(-*71?FE2xrpH(U>Z`Za{8D
zZ!K=+ZjgG95Pe|HDoWCYq|zPd&)>7MpRhBwwx3Yb*et|AOEGOEw`tTUDBsp5Jv_Yp
zVaIOX<Td@)u8Qfq_k5gXw|D=%D!`OCd3Xs`?b>vFY2@36G%w}GG<Jphm;I?dk7|aB
zx%Oth+TVWq_ILH5oMMqA5ryxNne{+N6gifKtYao{@`fn8$#>Nj%54hhH&W?9x&Pd#
z*XsrvDwdbHuh_pPkH(n$FWl!5+$YC9x^<n+A1C9OTfA}t4!OlxVLXM1SN&ceYfRL6
z>E%N!9)>h+_`c~W{%De1O2tk%ET?yq@#tffI=6)4KZL7e*8KZE0M7DDen*e7+fVo^
zoSjS_PI`}bUd>6K%gRKMLtM9bIzN5wrHlTjWtFshgf^;Oz|t^g!5oYYy>%<X_t%0h
zqnf^ys$LZ;{?34ylB^DtI2hP#_6M9@L}~j}UCyADT9^e}l7I3f<XU-9h<;Nm)zOj=
z8~XnEr^ZRumf>+8qvWg;RKeIGm)l$mlSVrf3*G(JDKj(n2x;j$(J;+>fBSc6if+Xo
z9x0*MxLz%#GSn_S9TBKl7gNr3Xkl$=uZStPSiJD7{~cU@B`%L=CtVNm{{My#1_d!t
zhQZ$3(v7Ca6{DYLUBM3)iO&uidKSy*-8l9C#<%>N4E-#39~(-L)fHS*7p5FRF?qKa
znF>4X*ZuH;y}j*&#};9Smls7%8Uu%3Cbh${$m#cz<BzrByq~x}D!<*WD@aA2pPZXs
zB;}A7m2pU+)wSF~V;gzUN2Ld&^Hm}f?Wovf9^HK|O<V~qwkxm<0+x+)%|_E=@=CgN
zcB)QgsH=(elF7;GcWC>W7O&9On*QdGXYbKD)KhLt-jdg7dp3V%v*2erA(I)S!0hbw
zw{JJE?xQ1ztgjQ3mwxAq39?ftljA+_ar}{SeU0S1b14BiulDacW_Hrps%fwDCF#54
z!^_V&III&|Q@dW&olY_U&=gasVm~|Eq5OffzC|aVnRy(IQm@icULU`BVQAxzUfLqG
zG_b~7*n>2kC^FtC$vE;d(XLvPbeOosKM3CIDAIb08&KP>n(NQc)@<6t&Mv)Uhc6V3
z%SuYpBE5`eI~O4Wa#~t0>RRGPN;Cg)i%P-0vfb{tWd;V=BLqVd)Kx}$EqW+7rUK<d
z)S<74R5J?ZPs={;eH6H_MMc+uE8fvI!-kS%Xj{rp-NCfI(Y`QklQhqiCA^EKTL?RD
z$1Xij9q-lqmyZy+{_RH>o_6&2H+A*vR1PRU6#033C5NL|zks^R_urNwY38d{jpMWN
zqh_c*DOW%F=rl!DL?ZGWdjCOMT3YGUDd{<eWgL&}nndk;R%Lsx{1hB9@G@KUc%w;W
z>zvww`E^1<J+S&@tr*1VR<U~4VPg4u^5~XUARzKEeCZsZqa<%ixh?s|#v?DI5Om)|
zRy^#6{)vH?TrYh1Dzj|WP5Y(?*Q;p%Y)W64W`hkP<4KvTDgM1_xq>8gP0J1}4`bnM
z)D6FafJ||5THOhA&;Es}mus%sAFK6mA;dgP{wlD}LaDZ`k`wv_GYB9=(qAjXAoSCN
zhi3{Ot^APez32E%ilN}f{igGm+jvBIlPjk&CE}G;RHD$=qS9BFOq48OK7i1!>dPod
zk_rHE55PxBdTDs#H(Gc>2Po#U5SI;9e-Lv^PPFdzlU;}t8GoKM=jF@enb{%}whPhP
z{-<hsn23G6a*RANdJ=4D)^rGJ$LrELfEh^e&J1cxGjzKiRhHilsPVpy+l&1E*YlE~
z?ce}jf0QlrQwi5`1bgGgKo9SPvnUkbxN(EI3RQX3-FZ^YVJTj<O*|3Wa`y9LI%D^9
z`F}f~x+C4o-8@M%*}a^V-Scluw@RH$>I7yY{vS9O{DmbrIPlt~eWTF#aR#JGKtrb)
zO-g{VQl?*}%-!>&Jr8c-6#U>VXfI^%{@dNZ+OF5%Ka9JxTh?nU;EPg8U=r76f!hRZ
zDnd?!iQs$i=i*`|NlD2#7iFL`Qvb_L0*pZ=P1YA`KI7J{XQNYVrEs;MF;>-qE*BBQ
z-acPs5dKL8+QK?|HsgyE4<*EEM}}ioIbI!UDp_)4;q<3Zd!kt(3IK--jX+=oD76o%
z&aPrU6iMzLWl+l_0d9RCY<-lqCzNQ1Qd#??-I69iA}OZr5UI-^9h5>Bh#Hh*LMu)j
zKW>P=2{cCqJktZ527}m>&egha8Nba&*7|!>cd)`MBpJ3GI1UhOX?%6EGbrpi?AB^G
z8EI@!Sy+5()R4U*m0JXWp<4L&9MzcGB481h5)#tUqdBZOaFkM%)o{4Dr*|94NbA&F
z6#NK039vf+Eg;O}KIh9dRQ1lF!q{84d`kpx+?0KLP+$FK0XgM&t%)j?SnJFcG?Luw
zH|>dzkwqMjxYfOcOcEK5tcd41Ato6B>53i%INW<Z5W<m~`}fak8Yy^Jj3LYiwXT&6
zcxfXcXB1YVv((^x4xQU$Rdc#gsBNUgEr!k<;v7#Xq3O-d$nf1{;eN|sRB;zu2NQ2c
zR!$0=j?;-zflWs>HxE*gm&j=izw9KG46cIeC}}_9zh-A=?}4tI#KRDqzOPxqIxdAz
z>HPUmdxY9lmbbG|)rn49BiZ9n8X-a^VQMI;7)=g$5;?oYv71FbXbGiye`pnDET79z
z0Y#}2HTn%(w<f@R(chG*jHOwpuICah)aqPS(@7&4@Yh4=MUfi67&?_mAw8@C#uNej
z0qsqyiz!tfaym7f4uXX;4E!urn|QP<fM*%e-$g^WB;(})Dpt$Wo>ENd?Pq63ojWjl
zT0yV;-MeG>-R}GeN>S`@kF0~r5pG-LIeZC;Ft}|}TSV2|{M!YHbQ!Ojz+A<po}>cb
zo{q*^N@aDSNTeC?7Ke>pY?Gdq`j4NA_2UiNI_o&&T6F1Vp|Em5f+RYuCm*p!+*rIq
z8~9n}Fpe8f!EsPdhnEktNpHD)U_-tz3ea_7-K}*3V}g=sxI2R?B(YMnOQY(;leCnI
zr*CIboMtzXvAESOi{t!Xa$Ki$=<&L^i5H9bX`tA!m?9L}{`F{X4csHApGVtOn~jY#
z{Qwoc0_%oTcbNS}Pmpn!DAb(PgD{ZZL#twEZ_M7{PJL#VkWMq@!JAisyRCkDV#O@{
zD%GGsV`W0E8H5vBdD=_%2Q6g(oUjIHZ(NYzJUHKu#ir9_!H()H)AHUHLM~}1AhNZk
zDy2BKdC;waH|c7q?`?LH_|2@8@9N#fAC|M$EUM<o*I6N!$w!F4xE$_y5~3EuQwV8W
z`TNan{hn2{Hn{Je)0RjN4-ZmVot&QT+m!#?9f}ucb~6u*YM5cwjz};VA;qporx#Bz
z#+~iVU!#4OyhYvIoLleE3<1YXtbRjDNl7KI9~jwQuyNRhDj^>yKb}WB94R$J?j;5I
zj=@1ac3ZmC=IHL=OC-lwiWd0?(PVrc^R}m`zFq^jz$nZzL&%v23N^Hn&<jD%PWm@c
z+k?FVjJ7PU9OWfaAnQlHQ~CMnFKD1hcx@9|n^kLlx=lEMWsp)?{|}j+`C_bjuc_NL
zIP3-BNrAFW))egeswgd$=$Jw3LqPLGYg~1Pr9LU@U`FmsafM|C>|QS*nl5!?Ip2#g
zPLURx*d>A?8@BNpTs&3i?_5~4{%PMMEK<}7V&I4ZmFz<a7Ny2ec*o8TSfRTFlho=o
z{2H7hBH7h>TefTwcK*H4?#P+2HMY#b+o)fqerT1Xz^PLYhC|D%ISAPrJDHnvPtg}m
zEh9>egE=FgIU7$R(xR5mzIwHfQa2D5!nm~WC`!Sc0s^VUX*%1W^S<^eZtHa!Rr`Im
z_>kX|3(ARsj!FfG<18Pl{2Yj@HYkY-<W6u7iIJDCuUER3SDo={!x}~Ikhm)23psR(
zM;;l3J|qxT32UDewc0O(Q<F4|7LO@Z2^`BhA=cV-d|X>(qox*!-(~tke0`si%6#cR
z`#qAre$`mf8rdc-CuguN;F;jhU5&aScBQ{x?4B2sDqGGI(gs?{E^Vngaok(>@=;!i
zU}c7oxRF0c_m4l=gG~Kvy`7%5EcpRVT|{mh8j47ZUuMaN_wQi~J4OZ&SNF%Yp<(OQ
z45HXabb}hy3^)_E#IC@+<}kZ*L*+rcXnQ3`G3V*DI^p#)ZgbKGMIuzqcD`4m^CJo!
zUN^_2b90wW$p8;bP4y+0^mqTSTqIVbveMFDGA!F+QB(^vpi5O|2z%Ob+OSi_2BXLD
z{kx33-(RdIR4~E5w}0&qz@I~JO)V{75iEpamzrVyzbF7zqtHv@qvRvy<DV$|hc0^&
z+_wcY_vU<gOLNf))$ZACt~VX-JXrczD02J5y`0||j33(@X+HPA>e$%Pp!z9I^Qvj1
zSK3UiEw_>oZ=H4{cj=oKnzm=lLK^m^?6u`CDa$=w_kP_Pe%awwYmXetm)N~qDpTZ;
z#qCFRBSXLVN*}#@w7O`fG@G~8I!|!x_4*P=Q?~)sj7Rz)y2(8{Z~_9~>z6wVJgTdb
z_fE<gmQ^5=UBj1H;ekK}peyItNh6~;Ks1=23@r`ZHtpS}DJ^Haa;=bYhf~OIA+MaT
zCY9bS-mTXH1BF{IHSjO7<B`gz+jk*d7ByL=xz%;IZn;aBi>aUYd)>&wePB<~&{-_<
zy5{(+Ea4xnbW8fcDb`&cVJ|KLF$o5Dqw^=+f?m}#vxx7ncxc9+&b#?sx@>aJaC@7q
zEK9kjPXXiXs+VrfJTS`ov}sy7&$iCHobZb`Gi9vRHFC%il-$I~h+fH$lcMPSa}Uz{
zy!u;j_0SEPL@C9Cqap>gadF;TtM!tiI_~dQ)!}_sCm^c3HS#QqX0Ee+dh4A_P%iy*
zUx3fua9OA!dT=K${dLsVIb`L)+cOKNP!)JuqMMKu^rer9H93rqI8o(b%!6b1!THTo
z&<)>QDq~70q0NzHY2AG@+oF0UZduZRA$}&6>8DoS*4(7Fp$00<-#-dIfFEcK_3zMF
z_7Yyfjp~aX9tr2kEwOKl-7>{;?)n@4BIX|@F_!#h&UnVZUGeP@9NDB;O=Ntc4gw0y
z!YKGUYhPsTPdlSQ-A}E0_wNrs@Q^L#!SIRPPi|ZNhtnc~V|X#=QlxqFe`buF%zU{Y
zMMAE}Dz>?vx^=%COx<*0RXlq1XlVIaG{vLfrm3-3@oJ~K;igSjzcDrlhWH1t2Lv2u
zO;}!@no*kb{LX{{26U%jh|Tv~&tkURZts)%>tru+G9GU3fb*Mq%00NUv^E-2Ip3F?
zCjZ_cx_)*Q)f-gz`Fb8Ha6&+y{drl!k1hB2caJQX74{`9k7Y^DINHc{ZE5-Od^)$O
z&eUBMD>m+{kP~W-+NA4ngg7x_Pw9CL1o}Y5j@|BN*5ALq@4|4_Du>m{3>By&krq?p
zox4)UlRG~*-CO%2q{e_IRVnXVYcx|6V}tI{F^An4WSYR$qei+y?(JwFAb=O9%V=0f
zuE5;n8-@1Ag73v{4P5!b9n*s>>W22c9UsFOdUyH_Ur<(NsS;%BSly=i!`{t3J}c>F
zUe{O-k0j|R1(vQQ4o=twr7O41{i63spiK%d+ih_1P*R22OA|qHji$TT=r!*42|GWo
zlWWsRHw$Y>Hz&@TE08)HV1q8H>7l-HlT&9Ty^C?wamxdaO9gHvZWjRkI&@cz!M50o
zxSiSJGNF@DLPw=PX{FitJ!@4&mn}_}HhT1IP>h2sN3OW6Of^|I_szLKXfZRX^k?|q
zxk<^0u2h;SB?#V1$UW#VG`8iS6k~qaONw2<KX$HP*p9+t)1Tcyv<rXjzfN=lD)`UM
zDNg^nImK?MA^31TWEwF4Wke>R3KsqSFvIQtn3h(e^b|GbLox7EcG>>EI~h00-riww
zStNY@Q`9eh+7+F9U$au?>qG-%>LHoA{1Nr)?|bVfD*as+)4;)?q#XVBXq~CMyLUfD
zSr^l@zfQR#vG;uI74_oh8w?^o2ktmlACnUBn!|lf|E{GKl5`LJ^vy)5NUMR({->HB
z$G=@ycXLDF22NY9U0d_8%?3XeCj6*z6%`vQvV`4AUOmU!ps`J7;lRM%i<e}XtypPe
zV!pqLuWb$0RKRj`-4hzd!KOI&za5QL5zZrEv~zM&`r$+GZo!b6fSLy)yP0X-LbUrM
zZ#DCTIqs6<v=ckEK3_80%kb2$t+Cv+{1uWAI?BV9G#~i~)VRG(d9V6_5|x>ZJbv+F
z{}OgQqi?gIk#d#|r_zC$DCLLC=eFylg)_t*ewId?$V0iU{>FKVVlBLsFe`a0#kN27
z@Z8L|tN97ZD;VfN%z9bXkI??w`Gz0wq*^RFWBEFY+0Y2@$s)gGFk^>K^Q~jlukGq0
zcbFKqSx3@js86y874tIf%WpsFl-1rVMg5{xKS=%&=W6m0F{7v)W*R29tCmF!lv@xO
zy#+6i0{yui>D*4my>YRa6~Ac4rVbZ$_E7u2r=3vH6l1Kx)pV0uB<M3lS=>%E*)3aA
zXBEi=D`Sb~@ZobQhX*pgY_DUXhgY(H-O8j+Nr9ymZb9wVlv}bzBl~9UlIzB}ciy6I
ziqdOKlii;YNVjlC?{XR{@2zRjISX$3_&;QBxdm|zP9UHgZ4DDX{5B<9f*xXEUthAa
zZQD`r$AbM)KYsA1nwH#QKp075q}4QDqlz07k4*aWizU3uDr#LBT0?K$O<X1o?z63Z
zN<-=1;b5kc8tU|7gx^_bzJ7hwWmX{FnU-?c1}QGVuI=cD^wC%o5z%*W`+}|~V1<`O
zNaVkDnHx{fC%mOO+nyA_Rm&M^2R?kiVX^7C1(jhe7h+fGzM`Jd?9OFQZtfZs?%!bl
z3h?Z`0C^BRuq%c2fFD|n{<7u~I}EWqui#peBtQ97sdM53ubzs`h4_P|`iqzfR6V@q
z{qpnsUWm`k(lN4>y$}~AmIG1<`|-n&bUhHuJL+lL<C=)-mr8@5Ag)RKCgqN6amg%h
zy76Ign><#j`U}&uAs6(kYTg({u}KJ+GR1xUYQwXA^c(<m@7Gt)MjVeHzXeyA_V{6k
zUy;NLN4zC4Q#ZgR>JqdpgTX`QkFemkGzHuZX{l?2b4!Wk0FMEN1hi_59+`%0`($Jl
z?QvNJ1$7u`*of;3UcQ%+4Y#8sBi%++80;gSb%*=i_egGPiMsb*Z*l2PNjkUH@Nm~X
z+$9xSjsgoKE9<~(1>6g|pOQg06HqQBT{GyH2=MpscuhnKrE9^bneax?zEtBl%b;`e
z<dW_<rU7v&N5>@VNcN(qN<BU6Tp9O!;01rX3EP-I?h}DD0WIRh*A6OBA`UZJjP;^l
zFlz|ofuWmm$8OAU+Q+lh{=y<A%9-dmfjm0#XP%ve4^f4<8?vzN4T`G7ri#X$T6Dw^
zT?hN1$dd*l^gJiP`kf$4Z|sFJ=fRi{7^K(DX0=&(z|L|7tq(knsB-nqG?bSsAYfOb
zKfgDdpZaA;zWUBfjfN6;JB*chXX^`E%9+ASoVIwTiXcE%bcB4szGj_zRA`vCBic?#
z%UU;Sj455ePQs%D(FiBnP9{*)wgR0)RK_MDt;LkX;gkRKSpBX2ryU2U%<zQ0XhXl6
z`Xb8ha{JDs1$SNz3sO`UA1%Z0{c0^sr~*-1{%`Lz9Jf-`2$t<?{Z3PgDn(cj_Y6p<
zq4p!P|LuzdZ)tuXS~!OJpRQT(>}!8zOKx~(!FmscBY4#mwOT2!K-u!u#my2tzdDv3
zNzF<7X;PlRFa}9FQl3wXu3)8t?zc~VSB}ko-+7d>y&e6QCb;v{3N-VW#_D-OJv!)F
zOYJZ5+vG6Z(V&=*EY!R5nE+Kf+_5-3KHjUxIL{p~z4vsCJYof-DZSf$e}%!RxvNFA
z3WyxE@q12Bb0+T)sNq|qs2lR$MnDhl3&qQomD&<2Y~@>|X}@Knp0ev0Cs#_AFPG|@
z-zq99omDkUD_ZW6)2@_tr~9?|wEBn`xfSJojH?P7`E<%A_LCy8=>0v($2Hd4>-^Se
z-L%uHO}`(K6y;X*t0Q13SG#wQ@mOy0js#9!7LqJ~ZDtso8zM-nGvgN@SIJCu>eO83
z)?UZ)4J4(0<8vOnk%LcM#zeDGX22_6si)JQoabt$Cssai)D2}K)^~pDI}>HcX04}t
z9qkq}_pi;Jn>qxHzMq}&NgkZ;vW}Qbc;`ZHLTU&%(Wm6nDikq2TE~ib|If+WW-%HR
z6G8OA$qZ5GWE>blr@ow}r6qCfpvyJMDQOLkP&>bB)Bx3I(p+j<|N6w*wQI4XhZ2?(
ztYe|mXrx6ei6dz<A)771E?4{ZKEQ%v#OCXHrT>MXMM?@wD}p{@s|n~MVddy%LDyIJ
z>vB}^ZGU|`!ob9Tif+lu3Iyi9?9;e0jawYdsTSD825xfBUAvCKn9p?J;@$Cq<^*i9
z5j2Zl6?hwYq>1SY8!7CIO>5h`ckef5)tlCs{A0M*vj1ubB34{tN)$BF2&-#-fHq(Y
z@Axx~))&)d<KmR$z4af}zTqF3T}DZo^0~JkfV_;|v8sXkRN?9(_Oio=>kT5{xTre3
z+~NECWh|l&icnQHs;p2cHw#iyiu!5tX+N_M2M1Z-Sc#(sV98`(6~r1D;y^&)B7W8~
zND~m`{MpFouz?`GRKXoF21_^`$fe(luiuY;F5~W=)VO>FOCzR&<LS280<;wo@g&F;
z(be`^Kdjg24>J1l^u(AB(>X19?@U}Y(ioFmH%WPqH4Ymv0Pq4no1FiwcgEw{#nuw$
ztlQocgG1&UevbVWwr9^p>H~b4)lW14lAQoDGx(@D-eVGNBD3STkOct*s7*U%ZP>yA
z11@V-RpttMs`^g#i#X1Gmbq)9Uq&g$oy}Ymy+eN$|2H}lLE3|q#*S{A51T?&(pVP2
z4n^5UDVN;mJrrjm!nA~}AxReLtq-3U6eewYx=3)9O2bl~<3=ABaH;Kk^aPH+>bY;S
zg+pSJ<F!FlWzqAJT0ciyhAjX>sLDQg5#v?x(_r&Lz4cFMtNqq2=;>27SJ$FvVP^%~
z51ekp%=&%3nI}5xl)FSz)?tyxA?@#Wk-yh2EFXzea<hD*L0tQ6Oz4ZrN~f+pQ2sQv
zJHYgg`9i3hiBx9q3&HK@-6J^q*G-I6g>Cct#Y=?K=sFkfQ{rWP)Tj-|0j1#)YtiR1
z)-*e`EjO*g^q2i=vgneE7bNh1LDzUGocrYNmh7iTM2I|qUzab09>pzzyN_eZ;M)P`
zjgG1;>?-$<4+{>ky1KB+u766aeS9g(%V%oF9NXGzYdW9K9Q~tOrRa^9d>C`vqZLbC
zn8>3QtYhe!5)F|<YfN!5v(K%|5vXKQRN%KMQGOK?oz5*wDnE7%mHB`DFhmZeuARHs
zB_xXUorIM=+a@#clBt)TWhc*_eD@AjJsi3~ExuW?=D%H$_<qB}v#O@PT`?lJeIapZ
z3j{Ojo?|4hw;)=hZ81TEFsR&T)Kz5omk-^108CO~2w?;gf@Mc<XbRm{HRuFo-@Tg?
z-c8b!h7#DubMBi|K+Tax|3f%dZw?Bmm1Prm0^h%Y?_bBj|2je(#wfP{8q`EduO}ND
zq3ZS$3a{ZB4E490KuguwJga*I-AUNQMdwv3Y`LV*g=~QW5pzs)C*`mYppfF^M~glY
zVMv1>q~ll+wI1y8k$)i{{fE&pF|xk$;i7<2o=!<`+8}Ct6D6TZ;UtE58=9^L4m=G-
zKNh-Hf?ObMhU&G#P3iRg-X2a-wFNw+5zP%RE&OKIu7Bw6-nHu{u-50bqFYdDEucqX
zf&&gC?3$(q`HFOI!dw>QLb`8foKTMgVLl6lN3q^pcx>2l$XGz1jIOLyB5lef3t;oE
z7o_n8J*sH%`vPQ!_SHx<L7|dY+54KNU>AUVfs_}km=>uV)C-~n(Om~or3%{s^=9v*
zemN{GDC$4QBQaI}ztGQLigwu>7}+PMMOgdo61ZroZ6}Tu8UVb|ziy27-@qYq_Q1*_
zeVO~CvlD9SODOT@uWz5eI}WpO0o~uDWOR<`m0lVuLRIsayj<#YRtK=#Z`ZA?I9qhJ
zwexK$xAKI<9qMT;;-JS4k@8`jEiF;fv$a5P$**r~*W3MG_<m(&RDooj>GkVM4jBSt
zI4C}GM@+P)@@i2+qa+vD<cX6q*RFXNAp-DM_ZCwHMJlxqOa&9TZn_ilC;7o7jceNa
z<Zbq7%FR3>cE`IIKr;-?f0mH{V)~cCxFPgioUCrHul-6g_f<50IE07wJMdU;y_~UW
zVcBy<J)hegUc$H$@o4~E92BLyV^2|&U1S=GECpOPy%)_y6wfZ;6UM(kpLf5T$O6z&
z|98Qx$p`C>qLR<2tPMUh@z1fA`rpV1h;G9Fdh_lXt4YXL1#)FS!MV>ywEVN3Xi!Ga
z8y^DK5f|g)ByM+G5xL-(G`U5$_>PX?##PA+u`5<4M!qZUUEe~Sj`#fs;lHD>5UW)_
zT=J`|ivD;CTL}yWhksstOWbqe-c^b&?b`p2EDVgmKV**o+4NLDhuMw|6jc#3(3s--
zp^ucpU`zB54vu~Q{&+6-e%q+B42N=AY|Sq54-7O${}yR2fEy8hxhSkknA9_W3PCL6
z*2$`V%88_Mj`3t9ZwjO>V6(lbyvRM~{{mW-iZ(YQ?@(@cwoB?^iv|T$B5lNd?%q|@
z)oo#%exZ}ChPDyX;D&0nQUp7`T5uGp-yK_t7^4Y|>~BD}4?{$f8_b`dW1L19;)1Q)
z*FP+btn^y@IEG@B17fi4f`-m*sdi>fUhG%VyG~R~SC=(6Y02+&ZL!Y}Q-9d1jvQS1
z?q$|eh1YN$lKL`=9NuiiI_?{VP4!;Cot2f9B?U(~#?>90aPD4^C2R9ny0@)W#1N)f
z=DliHcE>q|?s7Qr+(LUdh%)veB$WZ686t$CQk4DE`sY`^?mA!{m2|V_-oN9%pk+&!
zR#ZmIl4dG+7J5m`7rHD+uOD%egK=xUzIcyx%s~iuN6%Deaex~9MDV~JbxZm;NJz}Q
zMYOxx`>~9)w56<Ri|4<;zb=xGsND-0n1m=Dsl55Qs|y{5b5nf_%_ev~0Rvaz)(7dw
z?wzdhMay3(H!Jm!(kp7kgu0^=Io<KXCSF9Uvo=YG3$?i<l{*GMYy+Ovsru;NxqTn!
zCCyE1cbz1C-MdbD#=hp8=@QB(KGfb<!hQa{0dwfCyMhiFB4v&bwhq27N8fM__TwZQ
zO8AHl-$iGm(V5iR*iVOjqtJ-dizXOle4jgZoke7ws;92Hm=g=r0a_cDdhD9=TrOm{
zH0drS?HO&?porHZjgkE>sxLbR#(h%i2@209Mo83lhoj@B1a#!*gz_i7aFIB)B5EqK
zf0Im3*qoc>G%C!0anTQ9D>Ub#_fOZi8XH}*NL-t_?}BFAQEc5|j7=EFbixgcU#m<-
zPh-nVvgIJ`yvcRKFI4=jN0VJ=WwEv47t*QS+DXIn!<O-{VexGIs^bo&pqKQ9T(}CO
zN{Kbb4n18+z7cbydr0~LE3sAEm7}1k`H1xQS<YKlZ*^ixluW+4XP2a&J7Mybbq#5z
z38=rv!O4tWJc6N?G;odAqk|j9-scj~*s<||BC;hsCh?)izE-%SuLCLlb2q33<cegA
z2qb;7B^}CCY8o0EE<whRdOwq$CdRN&gW=bg=Y6Bh^CnMX`ZZv4$EorDv#p(P=_~q5
z|9q4sWs+eIKVKDjLIv7#`}}5vj)Sb&b(C|0bm(H28PBP`WlG50CsEJz!rr3|8Q4%~
z?Y3>^&e(&8XL=vv1K9xM(&pRe6SY!{gZH1m!5FRee0TQ`B^aay#AipVoD$dkUYnbA
zVKl4vOJBtUKKd;0QuH%n@5C?YbWX-VD5P8}gJ~M=C|MBBxsyH%2k`{7V4qpEYJSO7
znL?Ilmqyo31oprqI141facr;Z31cg7O5w;qUUwpal~i!SKS8|zFnwQVGOmt&xNiOW
zFjq86`#*g61nn3Eun|%-cx^&N+N!<<4RWM&d87JjO5j*BU3`(uElCCe^A_!^<3BCy
z17Rfx73@R79p;U05?>M3WPGis0R%}y7^(V_de>7MDwfZQMNtgcQV8_M9f~{gk7t^g
zW6uNJLqJf_<ueA<H}O$nVCL$tI6f@Dl3B>Ci7^?SD}5-*k{&Xhv<PgsiI@addCDbV
zIhA^n+|-6uk+(UgQlGP-)l?2f6JNAHk<DAs6~zS>3OBGa*}3oFLDJ){gkHOpNMmaM
z9bI}ANGICS!;`K7rV@y9OIB`<#LgxdgfBOi(|p6*a@)q|P8X|>f-MYKDGOS@ERCD+
zvE-Y)@coaWx0ZEb8)9r=oAB2^B>(y|3gbmYJ~>Bn;(z!GkXHUf0sp@b`*$^s{~nV6
z9umxf|1D(wKfc~{c~A-0bDd4CF`#1g)0^!=NK<G+No)W*e$kHzA-lR)$&G2~Z;|3$
zF80|VE)24PZ||rJs;B^|yhgBj+D?f}>YQK1B;HA?kHg}gPbrJ^sa`!QUdrucuE+iR
zm2cfw5VA|Qy8(SB8v(JA{a*e@dv6}k^}4nV|D<R@sgyB=N-B|*nO250AsH%DNr{kI
zW~)`nSQe2?86rZ;kXdCYWQqt8l6juz=eVr3_g;J7d%y4fdGF`*{PBLC{#h2k@%z25
z^E%JtJdWcO6cmIkQVpOWJc2I_jLgr^r@{sZg5v}Tp6p>4rlzjGOObeX!W2512yp>h
z1q)sCf!EivX)D-9XNh5Y0rmQBm;<ZjDS+Ea+}6s4Dv6UPWRt|RBhMLrTiv853b*A%
zcy)C{#8c(6IK~Oq(kWs+jItGZUw88oUS(Ou2+<WjZf`N0nbntJkW7pr>R!7hl}X|j
zh4Kd%V2#QjUaKLk8L-qPDrrcI5Qc5O-0ZH}L2Nbf9SOgUbhHD<Eb+^0p+Z;7tB|ZD
z!ToJws8v_jX(GU~6{)iZ074^J%xrTUK1!YnNaAdo3oWHr9#W$%`!pj!2u(Ar{-p{1
z;8bZkw!`I}t$kn#ynvR3*n<$FAjDds$wXNL+TUvkDv8sb<XE{7K|jK7{bV8j0g&yS
zIDS0tRXQqfQlP^=)dP~4R|aJv$s`}|DDkZ)i6`_h@b&F$U0TfGxek(4>Bo=N4JrB%
zgGJ+aQ^X4;JUo2NWn;YfB;dt~&*6*b-oAZHuA>wPk_lp!uGeXlE1{bAj*jMmb9Cj0
z%E}{IHUmcUaMnFcy2`)Zxx<3o@i6X@7fk;?FiY6oOe^{Q3i)#6!T37f@Mmh#M6c=+
z^W7APSEJBkDFzSI%g=9P)XSGu)13!gu6uEae<6>Dg_Ay7XvIOo#_9n>BX4OrnE`Yj
zB-FR4xVR7aCjbqsH|P~egqYF(7D~$e$*K>kBVn_cBMI*Sd@Tga6E8?inPDy=jFc{9
zP8vSg#aQ^>&?R+Xg~RGs+zT>QZdj15rl%Jswjoj2E_TR%yn%9bKp!p)_Lmp#ahD)g
zwh_t0@Rr5M_Pt}nof~=g?-zzAgxQUz^iQ4hz{17)=<q^PGe0&(MMoFDfrq3Q$bTf+
zYjv6mWn^c+L45zv6%>Ri^bMrJB9%K`XScGjT<788VTPlh+9d6J0WS!f#FlcCbn{!@
z0Ezd4g%KOl!eSV>88qK~PftDq9F@E<e1HYt@DTY3#4>CehAQ-UdMmn>cRyWCP%!~h
z;jr`KSU4L<I7f)ZauoJbF|eP|5|mH^IxY*=QE%Qf*dZJAhNhSNJYxx%zv$Kr^v>A)
zaDV(*IkEv=6<;*VHPFH!1KgEI)%h+<36gcLdsZ6OW<$N!0iy_ODWtL`-r=Bp6vK<o
zSYmOyqhbK2;=%y2z3^%pw<HvcaY+Xp7m4Q}QRQD^_Mv<uaGEi!$mI3(BFwS3y3(<q
z!?9h?1q`Wdi&(dIZAcS5d-cjg#p?y=6a)||;>Xg!Dg0pAl%A+^;X-fo;aj6uI*8FQ
z{?CYb*f*f|VHy&G8BPb<bv9pZP}4A@G(SXcC&B?MaR~B$_wJWKr)OtEjZTG2eG{?x
zar{nKR}IApw@Or(%z5xad*r<UT}Ob1jv(^2?~aKtUX;7Ly91xx0&kZ?5Hg!G%hxTu
z70{v{ho6K|4kBsA0=gyMXj8C3yju*zZ4GWg!IXDX3D;dKK7JI0HryMd97Hl;Gf)<U
z+G|WC@NOeOw<tKw6@z~IV?L8_U6rutB$&V<Bnng-A|Ff>rXB=zsu_t#RU}hWQxhLU
zvh@!htjUzb5rBAFT2_`YX-6;tXATyYo`Vw|HTBTYi$h!<4_`N!yQ{#(-86SXov7;T
z@)W=~2AS29HMbJY$~xQ!&_}cT!(p}ui#=juO<LVdLk|HCU_ivT^B3`S__`;NA_`>z
z@vYXeYg<JsJ{U)5gO9#laTY4>1MdZDlXa!7hg!6ule37CR*>sjtO+hd%?8W5$i}f1
z5vBo4*)QUywfLbx1eI*5qxk@AQ}15l1*(BM@cGN7Aal@VQM>brh%{s-=9FmbWbokg
z?Zpeoe;zfm=r5boR8e8HEL2HeT{X;%0{G_RD3~IO5;>rRL{8(7B*cbeeK-cKyQ|*L
zG*?zu&S3Q?lCd5nxDuaLcqCcDW%HW*i&7#<g;fh{t`fAfge2Le)jzs`2@Z8x*2b`z
zDK{82&qTy9LJ+NJJQFGwgFec2+#0`V1heb#WFkc+0%JV%ohL0EgZANIBsQc3`-MYh
zmFQ6BGr|Hw^y#vEFQ64h{Zjsx+4Cf6t$18edxRG4NlDP;Xc*Yz@ILDH#$_kxG}VZ0
z1i%^{>?CW`y+h`m`~zj!Euj2ZPnO5G#O^Agc_Ht5w<PGGfd{585L3Z!9MiM#T1X@=
z1AKh(jYFi|vxYM$@pi`v8d4@_6-bOBK=7UT!0k)?*#mfhXZOId;kk-;{=$G6=~)pw
z9a3<?sGhGm8OGmhAs;3NvS`cmqS^t_RfAYa%npbPT>flbvK8(cAQ>Z+u*43QxKe;7
zu1Me^><cwWPW~Vd^iGd-5a(7{;=}@nK<WKK^xXnjoZ|0Iu`b~Z)OB!HdlOk#q@@QX
zY#&0lJuD*#=Q>D?#x+Y-&E-uqaeXJOiq<?Qzi5C>eJFDSdB}JY;#PyFSYw(<D5#s^
zj-&T^p5q(g$g`LUU-Ga5o19qn{H`SR9H#)Nh6~}?(MeE~G-Xzzo`WT$7%BH*@~Vzg
zhVZMq=FV+tH@3J@V6@WE&WoTDL<{S(Q13!eg<0rxQ|%{uS)GPBY@46L#)cO)F8K-w
zLU|{qv|sAE!C3bYUP#;WlQ1zY%9|f8GioJ!pKwBOueU$>^o%g!@SDV&t?;@q#9Q0L
zu!bx{GRcAXinlkr08b$z0>tOdK-q8_#J_lPiYOC63t3K@0P+!zN0`&(`GA0c=lCo5
z2WPRi$>WLB2#K;g6~moppA*^ztPp{?5z`b^D=RDS3hhjs669jmZ&H$Y3X%hH-o<xR
z!^b(zK`Zku4!hSBPJ+Rrmd`M{<Arfjro5#6comk?enrgq`GPPDcI+<V!H4LPD>eO{
zTG4`GEwY8@@QP6T^4h45?@tFx)+44N&QsZ~na%YxKH`H_;HK6B)XK#Lg$ZmZ{=84o
z9>!l2@D19FsTbtXbHYv77yT;@wzU$fbR<8|Ur!-7sDf$NfT}7hZu4eFM&faT2JA&`
z23I3GC|Zb*r85kEFKD00-Tz^Q<{hBCXUO6t8yYX${rFng!%NH+x|u76weH-xLwLn?
z5R0JVc3eS~wE}ZJ)NkB)LGGf<;{0Bn%8H6FM6!XT7|fKJ+PG*Yf+oR>cJ0!e&q9?F
zs%;L<Hccj`PMk{VfJzZit&P5Ar2$}##+oLsS?Ymz#C}O0V`z1rOC-L@IQ*Bh)xMOy
zeYwbqE&Jvu^z^DC{Hhl&#Pe%s86woc#6%f7e8OggHBOflytb;~yG#s~2_2HSJ?S`2
zsI5z&p;&fXQ<NRNWOT1W{)4Ne0>u~xctjH21DZ$tbC~<0OAce6>(j$+8aWJ<FdrRS
zN}7Ue-YHDIDErikR(z7?2ksp^jv{)3%{fxcCgw2t0{Bp0+8N3l)^RaPTrMSHouNB%
zctQqv94BopKudJ7lF(ApPSFK32>tk-ipbF*#0c%puBC~^SvtPp>BQ7k%8G-5*j&^Z
z)ULDF{?^`l^V7|+7i}kVlMesUrnfyWx#!4{DUZ5kqeAuobjzwp-)-YYi=q^$y1*po
zQUna2Se<*<3Dlsw*aTOtS?E{bwXtT?vQ}n9)Svftc`y9_)~mZ2<5^knmTI*B6XuN0
ze3y4$7yKR()apIXQbl}OzJ@Ei%aE}VVZ*X55oUP+z=?-nOt7RwI*Jb7k0UPck^U3v
zk`JeJ5<HMR|9%6Si<82JjXwpqX;=%FWWJI|yqfrWE&ROp-IB=qI`*Zz#&^fA(A#U*
z1yR!n;oe5#n1DURd2J7Pj6Nrmt<(eC(y%d}0fdB_t^q}zy+ZOB$9PsF;Gtb0wd173
ziKv46@;dbq5+HJN*0AP)s7rhkOilSua=+hQ-RDT1n>4xnE(pfh*k#hn${DEGFoo$S
zW0aXIQ@3@Vn~sJ98e}|D%8sxr8MdBh8>XL^tY-j8Qx6X%h^|=p0ZeVcI2}Q!5ZK}o
zG!HYPNEwLG9{_j`(J?EfkrONlLwFYWMZ=_`7sz;4uRf7hV1$-VOmZ^pM`2%?JhDiU
zpv@gr7#FL2&u$puM&>sm{>0vcjc5GNRiNLi0tI}?^UbnSl6W~l)iIQGu&)R-=>$h7
z*szmdf{<dADov?8nmostTQP<%dF_sDZ4{343mm>Sb9J0Tf2ilkA(CvjA1(xB*|&Fe
z5F<I5WgfS1G@NhWnqLJu0?cm`^WjJnu+0UyZpFo|p(5tL;NtqRbAw&*QOSxpjv+RX
z=-4(-CIe|4X})=z*i2!<MO;LLj9==lv8%+q92dtX$xtHz8Pjv*1BBL%<iEdp5guv8
z4T*T$Jv;R{66T^}E=i-|2FNMU15t%7#x{q3!3Fr%*8-HCKypv~iAbecwn)M@Nfdtj
zZ6(5p5|V?&bSVljbKGBgc=K_jklyF>msKMYgu>v5i5lXQD5{8w)$(u&t#p{HzbD{7
zPTyz5+Btu^=s>FTg1ur5fYnr|IU7Pn29}-1>IR70msAhLix3Oke!A@*$)Y3);EWqd
z6j3k?lOY5*a#|n(>Lb%E&=kV~cCIr8@jelThut`85nKs(8v><xoU29vNN3^HpNKke
zV({BVB4wOufzWI@$UvNj#YT*S@BH=~oRTx&RX>AB2kkIL*wv{Z01~%fq}orUI4RxM
z<6^Cas}CtFoK<z1m)WDC+1=orMk+(n#Yg|z{A(TpFfnBHizeQfh(#6kgE?boNLC3#
zUkQ9b>gIf)SRmy-P;%mn54Tn)lr_ZFi69rGr8io-7wXe5@Vg=^LT&h&V#j^PsiWPQ
zl7Z|PRDY5kqtd8Ravt48aAI+r{wk6Y04UyrO^S3cx80%4I18p_7>gygiiSpgMu#Zd
zNPdViThV8XWkUh12-(KP>_ZQrU1|N}UMuexyX^C1WDm(wh@*<(67HQ?ev%PY4f<R-
z<rQm~G7e{rlX983Lm4pNB_6*-V};#ACW)ju&R7UZm#VEp{|w};Lzs(1OfQHZS>MD0
zm9Bhpf;9FZj178^lkVvg(!xWO%Q`fS?`s6lCqEvXr~u-hNSOlAfKB^;MYRgfeMgYf
z5PN0h7k)gzPR?MI!Z{o`_)sx$uRP%g=i_)jUm?3;Jz_bGla*+!F#Ld?7`YK68hjO$
ze9=%lkWZxHXa=WBB7P&bVC!9`Z!e;-IO+H6?b~}phPHwTii_CuSxGwJhh|(G(!9FG
z7nCH~<f!@?E;uPFpjBPwlD&nwyJ3)@$j@N6-R8JhBE6`xSN(-DwiY=aiL2T(-<bw|
z-=0dB#n;H(IxYH3i5Vd|!2I{9l4G6N48p2i84H5=Dk8nnCN7|9^*Vi{PNdrl-s)H-
z{x`v4FjE1ci1gd6p;QCwROfG0xsHzI(OlT31RE+#K?F=d!-p3JyJY(V$l_6id-*Gp
zt>Ge<&9v=OBwb&``*FA=@5R{*EJ$eG8cN#ocjTd-h&&m{uO3*n-4V{B>Ve8J012Kn
zE~VAeaUZM^n4f3&P0sz9cL5uuBk%UDH#iPh0UIDR3!xvP-w$|4rzImYe+WrxZJEQ7
zbK_aTV?Ue~l3&w6z@oQHRQVo4M*dn3A{{`dF&P;$X&d9KrxPQ>i26)gJP?)eupQCX
zB=S>HdfCZS8U8`}`8VJw{c(8slO_xf2#~AMViOPFx^-*zL@pXbwNVP|70=@MMvvUG
zuEY_hwBuw(MoqoY!Gq)E78TD?51$zqvl9dP^w4(HyvtU#JwynP!F?oLdM3nY(r&Yh
zC?(Iq7=mc5+b@PLl$QHCv`Qri4}pK-!J4{w;--}turwn5RrH~eywM=w%_*&e<U8Qg
zRlpN|(I9*q11%+%vxFUkRVSLUn}>%NcLX+ij%h==NMk2(K9SrP^9@3q0<k-hrja>}
z!5CMGDHuBxYXHxr!W>(%2A~&-D?kBBSi(PIy*rMHct>%1qC@$lY!Q2?27Si>I06E?
z%YKo*s^Bt-1HOP}<M#d`tJoqIb`7eElN0If1;o4#Sq`ysXFX*2zy@O}vDN$Fhe<f%
zWG)v0452!=hmv~&Wlk&M`rM^siG(RzPN_bpr2RQS616CKNGti9McMuf`Odk-A2c@1
z{scyb<r5h3iD{N1X7On_5@?ke43g6T$sDW`+_%|%rvlh4st^o7u(!GO>T6ONle8Y7
z;5l{mpDn+yim)LeniC|G-V;W>zMQ<Bgq_&eb1Vi|3JL849>U_nsSFsGxK3KZ#POsb
zipDzPC@5wPI!9;?vPvFBq>y!FANHh3FN~`a<d--K{xF>7bQ*uiL!9T4h6L~s*%rc3
zPWv2&VUTfNkofjih!4FO?GhpR>*(kqxDqIy#m+7dp`lpc*V%nvr^^`f5N!Z88M#Cu
z8N$68H46XM2Bg-tGuFgs4U|)syH9bkjIo>eOUNL{i+%AGz{g?`X^x{NK8q$biq?zC
zL+cApgDVkPe`_~51lBQxS7}V#Dgl>1;xTD{qXJk@ZIBb;3`ar5%v0pDk+_w6P~+e#
z5bG_Tk2;^&i{TsDL!5&ViqD}608UK4C8A3WGNd@-n}iT}0pvBN){E-u&)^_vh-Cf(
z>P-?t5mQ&u(h^~-Mn5O+Br?cl-*+pN#!Na=?W+oq!`p|hC+E!0{ioN`S$$iMobtr(
zkGui045CJdJA0hP5{~q4U;x>K1%N-`riq!wrUKOc6eM^Ii@3t`W~s@yEH(7xE2Oc&
zgQZ>a;MkQ!Qk9<13dDLAW(X5~x=R-jEkDD%p7<@HTFhck0PCKpQU7!(?jcfUs0#>_
zjwA#X>x}s)hzAz*+8(~XWH`sN0K2W9>W$IzNqJED0GSM6y@viB8A@WkMva`@nPzqa
z!tdR^4$*<8*kPQUnfM6n{mp>ppbe&=0O2EGEi*F66dcB$d*eer7W3v1C6Df;N09p|
zk=a_^XLMJmD;rsJ?an8OE&P!qJji^jQ6I!2Fp?cprz1ig;kKNA(SgO^6wnGMz#Nv`
z3R^)!+*}TYL07bF-mb~n^v@Z!z*Bf+BSVjEvlD$U;2$1d$wf1K9?1;26_QvsyJVpw
zfkQ7rkn7;Q80Y>>;~ox%pITq(hK2DsH08lB*ZT#uzU5N2>;Vsw?l>Gj>VZ@R>(E(_
zM`!u}XC2L9mH$$Tqq6}PG~oikaifoOY5E?xpsro0j{eRACGOP!%%4FDk%(Q@3n%dU
z-ns<h(S70Eeu^FJd??cqI0*K@rV}J>Vn$404n&(b{T(1rV+9@eBfAUM9GbA&vkO59
z42P!toE!;E4X&P%z@Q@nI+aP0r9y1!r7`AWIqku^ndRR^k=O{9zw%S;=AJNz!D{08
zhSZ4wj3oX+Xg^z4_Rm~Vx^?j;cE~J6G=vLo{kj_)_TMOJ2q!2F#fM@I{gLJftWQ$A
z)ay7TmN`Ro4w}<Lo}%JnBh)q47jpsunS9r(qjaqabB7S+KT2m1fNM~Z?No%PvI|Mn
z2t`bVpJKOp`9dN81byus;^SE`wGP1<j-bEDI=(FPNtWjFmzIY=;n*Z{f6#UT(_VwK
znw})cvaY_Y7+**RsJ=1*23wKb(cYelqjr{m9Yx|_l?(rm`NjX|o^6mp#z~ZS`S~5y
zrWsjWwNcA=lVYg+&*Vd%C!eyil+BiTjRM@l6D7l1NiVBXmWiPt?#UAx{jeC}B2t9J
ztPf^z=-eYja6S18!JC=8v-S)8<FR)??Si+-!-wd<bqGp^f{DG_wvFu4GSCm_1pcC_
z`8S%&f7;DS1*n1kAy-@SsJFDRIOI5elX$d(5jcyy_tMb-Jm=*Ks@a3^n4Dc2_~vv|
ztMk^8#gp%tH>?!+xl{Jl4R$wu@eNcqYu@{_h7>**v)e9Qd^PsiFPnt#mq|Qbvxeie
z*1L0pbo6d|tM|37QHqV16N^q$yv-(1_uJWxM-B|GzpUkdfB%huz+bdz^$WK5O_gx5
ze0`OxaG|F^Y{BtrP+g0|l_W{(DQBmlg;yBEVTUqOA3-Yxr9lI9`45a2XOox2P#m-8
z&vzb(di`2R3ac#^+mX_7+O?$w8A;Uh=X;(hg_EJfUYJQ%h%I|hOI1~MQ~unyAug`V
z0DVXDrg9H>pQWN4cl&+yRCQAm8&({PV@p-l214?>r)3G}$-)w;$&@!8KH}}j+&$5D
zvC1i1N!h-0#rSGatN5f);uqueDY66V*rBxeL#Z9JuzVUDpHk1xI7*GqILgZ2Xw8$%
z%E~$$m;S21IrFm<pP$1jO3#LfDHUB^PBa}k9P?27#U&?mT6{voWh>lByxrXqXQau@
zH#_8F9dOpzW}$6tY@C{&E-zb%`us8&l=>GPX{@ZRGnM2lk}2Cce##8#V^dUr`^JHg
z=Yxb);o7xp`?TKN{2=V@?Ts@l%iY}$6L#9YkF>b|enxDMw*?i2`PzzcDYVTtphWRn
zwQfrxa#Q`0_98)4f|_}bw?Hx4#l@93;%!!lZXFxvY}LKXK(@?LxG0=HeY!M&nj*06
zR2#EAc$}%JsdU$~aRAf0grO%Z0XO^(VtHt=O#b2_bbXoS5%>SoAC*&meB7^p<}xx~
z=gYxy9V)dBvtJrlQ39SuaYv>JU#?GjHx?H54&-|+{@g5zUJi>JjYrLw^dl1M8`2W0
z7$|kfE<Hw@Ek8C=ZBKw5G_SY!-+vq!V6?VA*3{d(c7URMn*1?SOIC_g&^ocr34!|$
zE`X8j>C<h3;Tf(QTzmDmGDH|=h7{0I$X?t|MI>1pi4>ar$js5@K&@=Bk87oT#Y_w3
z+z#%lf)AptY_5DfGSu8%j%Jkj2NZg*kg?pojesga?qjL0t=$a7>9cdPuERiDb#pU2
zEJw|YS*{g=mJt3+fU^9HneRBs1uWi&4skH1SjMh@bg;g@ek%Z9Po!ZYz!;g)8}R_u
zjUn?V+&{10Xn48f@QsEI^;(vv9osjBAAf#<mzvsG%HE!9c5W(Zrm9|TCq8SY+gR>y
zrEQIirF*3`OPl}wK3Z?LfW(q4f)33c{w&5)c3bh(P%ZxUApJ?Rskdjp*gH7rgGPM2
zl8u!$G9$yb!Pvpx{xWEc=4f8|YTQzG#uw1Op-<}-w*JwnzW#nM;2=ep?r|LM@B5cs
zQ=F1jng%##Dt@(rWw>d+<mIWsg(hrj-g~pm8n_1b-rjd+T;HduKI-l3(+BW;yE4P1
zY0JdK#Cx5luU}ap<?NzbdhzAU7el9OOGB7jUkK5`i3`T+*4R24a=cIGYMaEZTIbx~
zu|9B=ie&HA*LwVp%B$t3=bAn&0sk1V2<2^UZ83sf6fg!yXlQ84&9-+7NlmHORmtA2
zkY1YK9RJ(FgD-#T_tv@x2fx#o8mfxp+bKgG)pa|j9pjpIs3&Q$z`wv9{g+~7;Q9b6
zQygbTHBnN0w<fQ6<)dHR^XG~w7kD>iY85`(&*(6@dE5G`5^IS%#?PitwxsDMcwf&r
z?dcP{FRp@BRm8RCL+8q$Yd_c2Y}~(p|FK5XoRQ)km1b|+?wY??&8J(U%27S5Y}{|3
zAW#=|O+hJD>ReE}zql&rV^GGaC?}##w43zow!Q32)UHmiNRmBInC|``EXD7(I(5qo
zoZE#p93ji<%jP`w6b2}du?wuk#JHZF_KZv1FClh+L)^iEv8JfO;iuUf>o|0_4vY<)
z@ZSkQg_VQjj^tSfvaHtm`i#-UpBgOJ`f=c6No~h7`O~$9<UfmkIjeqSA-m>KW8>Xz
z0vl4MM9i`+DQY~nSI_8Ghya$nE>eLVf{ZN2JVHO4ql;dq80FSKe%mHu=9)YCEuecW
z9LMopP1)7{J{>t!lY=J?y?mJ!ySu%qwPkL$fM1SrwtTGt2#*o|cbftXb*ZyAt31T1
z>w1>YW@<<FX0?6fi~4(fbNaw9$8-|92Hu!9o1Kk;+8%tt*p|Ld8qpWs7_QXa%--lU
zXlb*jbOWgW`etvKdekGR&tKJuJSDV(-Yd6FojgqD&c+D06x)b^pZKG)+e$;deh$N!
zOr~zix$W2@k1uao#h`_mdzvC+rsNS%5f%LQURC*?S3f?r^%pA_mrJYc9kN#)rkvx!
zMX)H^x0Zb{*ljq+z;l>RrU)<o!{hvB!?x`d7A&rhOs{#X-W}a;EN`*BcC~X!rfR2>
zD~%u9U{Ise)>T_9>1AS=HS4QCs|NWTemU8p*JHCXU&zmezT@wY@{QU5lhyTvpda;9
z`C;*j(0DqVHj4bSAbleagLPMmj(@TC`fj_A^En;XjH`!VDSO+o#6ISvIK~&NQXHcX
zLir(6blk*e`0W>K_PDr5j|z>Cl1C%aQx-h?gxm3@H3uzU?>*^LIa>-xAC@Bi>z6s7
zl35cYrXUiymwjd0waO~$rwr<~ZqC<MF0qy(1@hvU6QtT{6Omc4J47JQn(k7&MI9q9
zs=oBDh{DDu?p<e^^A~LMPmT=z+{LfKfvYM$slVdHzCcgAN{2}^m%hZ!OQxoOKB>d_
zv>x63WXAgp-@4XM;$gk+WD7MW*A=HWJ(I&8naf|4Hc^>--gUV#(Yty?IeR$795Kbl
zF>#HT+sJ(Otv~nCcz^f(#ctm)4oU2)K0SA-+P*8l<HLq-bWXf3v|g{kZt+`=iT1^Q
zd*kj4KH6pj9_iL5a=B*?i<Yu6U)(8D;h-nv>o(y|{p1$i{D_P-%k)xSVxoXaR7%Q*
zu#!k=JJXWkv0rE!*H4Ijy|R*(*WvIOWl~tjVds%;N9J=LxJ@3zYpto0`TF`X?Sn^?
zGXpBN(^WHE*7xe==H@3Ob#_0&2JQb^PJR63zuRt(KPt+Z$V)RE5-=4LlRtbdCD*B&
zy?gbOTa=F7wpRxu+@_V2e+LuV&0H{8H8)g+A|r~cZZ6n($F(C(Z#9Jhg1+KAcU&n~
z;F|5VZXFfXhqbu&ThC0;Y~Op3ErQd&`zg3IN{t3GHajIQI_T~AF8sUrWy@2Zd2!xb
zeS^1T^5zR=?j2~jxxi5{U3$aS88qc~bZLDt-K8uAyrI-Gw;L>Nk61Ii-OX}aSU*SZ
zHj0w=xwZ4HI(x^JL)mwt<y>e5^6*wmgF}n_&NF4Y#h3}cIQ?U{Gz)+&=O1G;cCbVL
z%umnPdOV{x&J}MLQMk}O6n8aX(wp(?gRc}0Msu+0i@{{M3}ArR=OEwhkS-SO3fTm-
zL7_3`r~zBgBUYxDObV;~MfNRl)WvPg7vy5YyVk!rzbR1L=6v~+BL~3$O4Q7Ec9KA2
zWF&lPCLGC;Blo~)#4B&Bliq6^Jfc>dX!c$7F;ctV%KVW2F?|)lcs(tg<s)e26rr)O
z719?AS=rjUx*}wJffm-vZ606GMn!SA7j9aoK>MY3v$N^-J#&d7eC8P&@G;`g<_kuS
z8XT5s1*>bsA%Du@b|tu+6Ac&vBr97UE-f7sznz=GI^9M=p|DOvA!oF-rS(>4fcuc+
z?)LbE(%DmkCnOGQxhSN?7n*r}xLZy`42L9)jHmK)-!q8s<IkTT-Msf#FVTjtjV30H
zK%JJv&2pJW-g7=$c-$OdbK}ilzYWCOHm5>F%!g+{1k|6p$=NN_!_hyd+M4(%<gbt0
z9cpzJ>EqSm8D!rbg`1R4rEtY@(atT_+9vFb&5sIJYdj)<<=9o~aWq9Zz<{d9eh@^9
zYa=jhOAH{CydJo!?rE-MyEb>^Z3gLP+XQ_X7ut^v;=$Fy4{WIPI|22(%wq-IM>&kV
zSg?fscMCaXIhwn!i<L`Xvme|X{o&qZc0}S3&A6>{j$3XNZF)XigSgl)Q?ettsoY$P
zAb5xpB}fq_FA6rK<4neN;PWe+ImD_kE6!W-cGBmIfy#^WGi}qbC}LYLF)TyIBvrSz
zaxM+8U)q4_W%!WF7I<t}Zq%MTxUw!wQ!c*w3r+dnp!5~#F4XB3p`XQqp&B<D4XFG>
z78yLQ4H=%;@tSHDlo~JhcZynoT9-JGx!Fm;2yTzYDrNh>zaBi2e)yeiCof9aik_Ag
zqPvZZjM|05XHNa?<&FqcX8t{#d~(@!|CEo+(AU?yp%UT*bn7@j!ntHVK0aQE-bp(M
zuKRkY&~jk|HP!vuvuAH$);H;}fGL#TWG1VrsgDrOU?14_pUqzt$3A)cM4jQBc%1yC
z&(+j}ED@+e^f4CecBQXKySvEXn+}iJZ<vw{(&l-xAyUxR(o8c_y8+6_2jN!elJvnI
zNCD~L&Cg(Cq@L5VihcJy1YdHVQ&N(3S!!8`_^sPqD0p#D`SvPC#@4x(smp+bN%`>x
z9ki2Yg=AvNuDJTR2T8j)OMuscOCR1n1tk(E$`3&_0)s@%FYi15iU;CVr-D9>#0yO5
zcyO3xDnDjO=iJ<Jm1jX)@b;>2-@buR>R|_AY&!@@YJ_YN!_N`HMRpaInmOvM5=I4f
z#*%FgH*&fphwkgcd-6ek{a@?C+}vDbm7hOvA_W%sT5qmcACpVrJa?X1(M6+N?=7j2
zND_`4FDF}wZiB?tA*u+!D>jU6n5icAbIGToqCzv(P=(#5;nMrOu*xp(@Ja>QLO4W5
zB_tS^8^>E=jH*?*$GI(2y^~Gc$T#zy+6&!NjT5-Idxq|^?VE<fm)MVBx5Q!oH{XBl
z?(6eJ-)V;w03%;;FCu(m#UgPINn~_n8cHMyHg;tw44&GQ@OmE>8-m@)%K>nP+K##;
z=jBNOfft*01aSm)sC0iCvQTpX-`j>OjZ93uzt~pLh6?J-Bi;>dQ2Sh0*D1yHeBYp&
z*-({`IrrMRqn%@n*Xz~g<QrQ#N`bL+;QChA)iL4aJTYBZAk4n^`NEbnZ2OoWznVUI
za~1E@+;~&H3)_JKyFE#wox$#^YfTO;^u|mtrVo{@Q3!otYVP$In+M|}4TihW>*6MN
z81rUKC7|`8UC8}Q7LG)p)clK!Lrc#2eZZGCgCy;Xf;s<3+OZ-mtUqtZ(PPJsg&_Yo
z2Yyg;JvMa9wGeZ8D_h$VumSFAU3;l_rL_IPTc(KSq0b~AsPtJ-S`4vou5B2-v%f=k
zLkAZw8i)w2qmC#mM(!MNY<rM@<kGv}G&3y>Fim^=?CfkiP-Y)=y-NV4>7$26Dg$?3
zFR24qkJzQf(av$qs0>41Jsh_!qT8Zn-xEVnjOp4~DJDXBlmK;_tYN?U^#T2v9Pg87
zM3lwJS(TfN5pJYd_sB|*iP?GK!iBr31KF{?K<bp^YbYn;gBFv5Fn}2ayRY+j6v(#f
zWKEfazgh^UT}h9&i~&}FdH3q^rHn(Dztu;%DCnP>OFGc`iVt5a*f@!WoudUOHmM9N
za02=#tl-UkIB9#2bsq|m`{+^#g8}-bW(yrlja0tm1x_Px<TZT1b=@2Ew(IM5Z^#yQ
z{%u33FitnXmMrtPFpoLKs;364a^2Y%tMzVl4{evE&NfL+H1-U$*5#zDq^3zP-Wu_G
z=W8l^m`Rxbc&NZ&DoDy!v?DwKkofPV=8C*|;c3ymolo;`TXj~H%ORPmU6d=lS;F#_
zRoda8!-G<^lsZ5*<8*|eG8x01M;~t9yMjh^MNTv#{5Dp$-_ukx5!qXToQV7k(x(aM
zv@91)F>d9)=Vm4H=c5{%)?IANzo@PKH1x*D8?I8Az}P`BobO;z%y*gBswl*0G75ZP
z{Jt|L|MuvEGNy`$!D7OZ&z=njRXz!QB=<O!+NsGTqX;90w+%V8&PC?uOJi5vu|rN%
z!@tg**D?I->!r8Ud0p%*3mpFIFLr3$EBBFsb?6bYSPAQzu#J6S*Sf0$z2(%hUj;u`
zGYXnqkTo%BOL-ByiL=|G;?X0<{QDmEOFN(CQ#0%0GGirQ%fTG`k+<ue1*E3mt}g~B
zvI8??PGs+6-o%u#%k0gNXmS=``_e)_WA1$qhkra3ay~P5(5`P+?miv7nf#vz)7ysU
zAB1oHqu6wreox-eacO7tpKs{o<b+@={bL3_EY%mQi_h%(^259R^p*An-(MFQ;uVCI
zH1_Vs2Lpm$(&4c;$sMfOA`Ay5(7?z&m+5;r8TGj~Ir#eg<4MQT{(KTS;<dn&HwVSI
zgq;7l8b2T0rrCb-V9HcCZ;R}%mL=Zqp$M^ab5>~<eqSlt7ca6QKIvc}&3+Oa5S=y4
zLqvrJUg@2>a(Sp)CFlOBoF$d2)}z*OlRZYAJeG4Z(%4>PzmWqe1fk!&h)F|MwlIIO
zFaMbqi&x^Qen~NpBM<LSg+IU-^4vdOrtnifLVDP3eIcU%7HM4M_SVIL)^_Zd`y^2s
zcttsJ#)Y;nrZF}l;P>)Xq`@_34svI)+*a+);WOpeE<>t#Og)zR${oDo$v?jBx~f1O
z`|DOVHpS!sj`ZaBz^LZ^ItT{S&0SJkrDv;sYxvhkTp71L7iH91Y|^@a{7#qE6@HoZ
zB4llr?oB3o30@Eew_FQ%vT4T1QgUgZaY>z%SGgjWlkbl-cT2i?l-8BQGWkuAb3~@4
z2?K=>wAmxDi?98d!Dc!2!nwVhX2oBXA2#*(j9*>#z?x~~=|K;!z3mHB`W5&koP-WW
z;1${H=)KNnhKd;(ouAkkv4gsF;Fy7?PcM!9^N@6Np|lrSYpNblb&fx^koix)JQ1Pw
z$0H2}dClU~nausuqO#n)ZoaUwMzKtJ@Zh;Y-D#BC$3N7*>fHt4!AmByyky&+reW_8
zv=%68SM}KP+(Kn((%U@aa`IP4zlC;GCLo2rwVRj?s1+Y}jDph?MeTWT7e&0y+6o1I
zMDMvdmFDy|Hk5>vII%h@+J-+K)f4}7NZGA6h+1o=X*fpFm8M2@p!MmB=Pa*93ySF0
zGK<97tvK``%xH;$>G-Fhaw`6-$^`{ScL8m%cf7YY`pT)Fl&wO%df&Hi3XxqM9h;IO
zFndr-ynwl^^Mu4uM$l0zYTCz8R37(r@^{_0=j!D&`XHgBu6OC;gC_P&s;qbw{pUm0
z(q1VqnDuwtV?*7i%2j$O@X@^ZIahzWaRU>RJ9}7wV(HBfN+ksu^E#|Jbfx_5m)P!b
zs$xyS+bd`o_`>$EKM?#9ODSl;9npWGMCF>YlS~8+UqR77>3Y{IC1n-GYrp5}9k-YN
z25-2(Adr+eUM9;%zQ+b&Tqq0j4W_Cc{D=#xEXZKf$ma#@juOXw=3)%{ayfsqk*||@
zI`ZB9tiFd9fX-R<5^DXA*Ax7y0etM#x0Vpt7LtCV5%xdie<*c$@QLDHJ4!qpFZ{jg
zMkwc~saSbg5<1*Yy%Gz?_^Jnt92|_h!!$JmbW~NRMHM-9Jt}GO=W-TJj&GkgRk!Bb
zTBmN%uIqnC#ZpzxXg*O#>JJkKR4+<S@xx<JSKQcRTlrFK+Xw&Ki9TPxs_o~-0yHr>
z_I}JYqlM3umM;4mh26ETn(i1byNIT=&MD5Jv8Q5iIXDc*J4R%Q=9qdgFf|LR87HYp
z4jnn705>_uF7Z~yrzhg&zMc_E7Yn`w|2mec2F)NF^N4!+**PiBR|(0<2heZc0>S+8
zLx&C(B5%+KpIQ*4vK{r>2qq~Oou&Tq(HGcnu%35vvbUC#y~H%0?dkba3}!hGRsl)g
zw0p1>6UK`sMr<nMq+rwTi;O!+%E=xDl;ER%=Qo1=b0&LV{JHZ=+^_K=aJ{GylSv5w
zefN;b6A97w?AHBBzEd;GPX7w4=J!E7_Ex0t#p}yBP*V_nNA^ABJ3IOLgGHwM{)=3b
z{d**O*%Qd(VT}5=@NST$zt8U^|9y=3Ehd>RSWQY=)^%zlAT|xyYD<OtI!bXeGLmE}
zDJjW3DfJ+l`JZ#`O9nF5=WnoYmz}yl3efu8QzSk+xJyZ@H|w<P$r<N*4OR&3V>`e&
zAARYfPWlviLYJ%oWeY+}U9&ahsJ6EDV4b+V!@cmUA(aa3mR@tRGw3x(to?*TF)KS;
zAK-8b7_!~Kg28VDYCY2x&N+Vz;uk@r7bE<?^Ic+P2`FBDNQNm`;L<NIv@#DuB)0)a
z6c&s|2`CHtu*(<3NHfhhH!nH%U2OG$7NZb?g!bPyopGNV%9kE^^-<vShd7=Uhhzeb
zu3x|Z=+UDcCdf29zE^_l<qqDVm+g6qK_xLBI!^Y0D@eTm<;UV8&jU7vzbl>bOiF4*
z*XR03yC=u>z?Bs4%WUF$<+;<B73S+*(u%_*@8}n|IKJMEvu5q?`cMlGZ(XXmeSla5
zr=GN?(%<c(LW$$iI{$2cj0Z*U2f)H|5ZWy7H&R7}tf?vp5^r}emK~rNBt~&xZcKG=
zY<#DGPne?)wm-$q<2?f^7cU-(PE1edf2L%n^qk9;vA5gZ(8RmD=YIFl0g3eVwbg3N
z58V^{<8rZ=BfEKFt$W;kZl549V1Ml2zLHPr;>E(Gr^aGvmi!{F_MV@;;EhP9Ke`y_
z^12va+H}YaT=4_x-Sx8YHobvjx5ke=1S-`e-H~JUR<a~}iU&YT*}ON<^FtT+voqdt
z&-Y0D9J^x^f0smY#UB<cECWcuSKf6)P@4r2gW~Vzw*MKe%}2I|%VO_bZ8M2yjWP}X
zDoQZfgL<AlrQ{|3sV<(<0Ed_oLDz6O%Yp=bgMuxe$uvX?^Ye{E;(=vyH@<F6xu(>n
zV#lwPo<dDwUNcugA_IlOeW^7w9`Go|>o=#a_(v2;FDMJJ6on$u5;TP~{iWoYmZ`k0
zfZjbw<_^)8Xa&;>%JbJlT6G6$?K!cgQb^Yk32Nlir#oQF$Ogi{7k~_UozG~lk;Sq*
z%#I8Emd!-9>A^%zd2VRNDMl;yF3(c!-Rg0ve_x$S{s=}x{|JdFn9m3%6JP;RG$36M
zeYH(8T2nRGJuu)6D{(q0Gebl7mX?+s^`t%+fn8his;ycZ;2?ZYA4J#=%->UH=oi-r
zK;O*`vJr9YAk>e0NrHg-?vez?@tK2-x;~R1<t6sDyE}wT3K|+3YwOe>0V~P+vt@|Z
z-@kTEjwM~ko5N)I00W4>A_J+P+r5M;+y@5-gHW12=(}+7V$?+F!uD$jt=P0|W3zj;
zH8cX}&hbpGrPN$nTOvg+7^$`)S!jR>9Z7qjGeSGHCnk`=XFeB7mgTPmO_6Hc7K-c0
zE;KGl^?+ci`BLvFN$N;%7o-3mOdT%)g#`^;)H;H<Mt_h(N6V^O?r(*TY7#qML2qq|
z07|6Dm>7fOP3UflOj{!CHv^|L2mji~_Bw?UT}})r{-=${|JPS|8UC-G$xeOJne2@D
zx1GrdEieS`8X7mTh0*pmsu|K!5C=FbB?Dh@P~tIHjIO#N=<h$aY*W!VlA4&5ga+p7
zssbrQc}J<se``6JQSG46y<A88GF<D79*ryYvLDnMvUeUVTfY5IejKX#M|l#(J>bE6
zWK#`4yd<=&tYlNu<?+ewT0X1VZDV{`7NfDXvj~^EK)kVW5<xE8hN<xq`2fQ{U12A`
zoUz`B{!<J2TA{r-Oei@~xss7driP>(<DelQcj!YW*XWG<b5(95qqV*>=oQf!!1&J9
z_G9PDs|*$YM{eW4-sfX1D0zbp?~hDRA8x3AXoERszoo9D+^w%>tZrmYG1h3irKHpb
zn_T`t9c_y8fS1g%$3}m-t2NyCh5caWl>th@^oEG1r-TphGNz>zqvK%^|Cb9uIS%&^
z1HONbs9QFykZDAEWUw!AQxH`|Ga;4y+q3Dv`CHeG|MF%0Ysi<OeDK5C{qn_KJ9}-V
zgZ)}qYg%NE_eNSRM{ebdO)bL`dgQS9N*ek8RhE@Ijblg29otz;Zf7aU$^Dfb9Ave>
zRi$TT{q2Fu{97MZDx3;sRZ&&7EAQIBy?i6F&+^yg-z{+MF>%lO&!_Fbe53A5kVXHu
zvN{T-<~z?jr}2r2B1qlaA1Z9S4?qPW%oKzO0r2aEe`M`Vdc%dr?BlawyhIoM$FLo0
za~_{|4V677;6FiwQvbJI^z<nsW&;$;nmdZJN6*oNJCfaMO+fw->o4b1467p>PzKvQ
zT1}DhAR+oBe<lz7f>{;V&1DI!ZC<CQQG^0Bz18|}?QVk35@~jJtp`-DW76g|io0X9
z<>hWB*QCG!DfJ|;G2VSPi)+T^jmV&?*tj*V#%B!j3b$eG`~jfRi@<=3ow|!VxwxS2
z1gFwvCeTH+e}ZI4@7C-tgi@4YwyWZ?&~E3+*Wm%_4RjXa9$h*rKJ&KcC8pv<1~+wn
z*zDs3@$`QyAKOH6^$&WN;tLT=AxK*K2+!4kaxjC)x;#u$9Lj@j$lLBCmn+W5n@B{X
z`!wl;pAl*M?G1!ypZVD+uI5}a>!bb#ZhZq{b6YxDGv)G4BZU;4GAGYo{F13uWSPvX
zu$a9s%t4KP>b?a!SlF7&+;N0*--iTC8L9SlK6C~Q-|21zX;x&f(%;hv%ZD7^i}RIB
zj_i9+)j#z)=>necGE9LCTm(|Sczf~CW8V^}$bGA=t<_AdcvDl)5xkfrGAqV+pCTzw
zk_Cb~6UY#_)A~jhM|29j0dL-9PdHg$VBQ0H<V^tGg3#A7!F-JU4+0N4+(|dX9L5>N
z8hN7u{_&?Z1gH48+97}U23RTya&gY)_~_uCK87<c8thIS0rIZfxKRrDBqU|=K}MI=
z8$TWEG8=m-c@8wXZ3ptEFFW?@bFx*Ao>eP1?-|N-friEd8NZzHnfe2hhBL`%VvGbN
zJMQ$682#{HUz)r?2q0NT#u4N`M_}$&v_Sl-l^jPrb-nUup5(I({q3c%;w&~tlcQ*F
za?Zgd>qoaQ57r%U9`VGQHlDBMT&T7Sq@B$=R@KU+psNs*;{A=+@dr2OeUf<OPvYN>
zN@F7<pk-I{`-nZTAyRE@J;8*BH=6K<^F_MzcXB_lsLqPjJWGyhUR@P643*fn;WX7V
zqC3Md>s<udO9#?_U!+I3@{PkBM?cyjnITA*2XdEUpivz``aYuF?C<u0HRX$uxC~OQ
zu)e6RuV3$Q8lZ2^!ekS)Nj^amw?^#(gR^j#u`q?hoM(H2j+p7l%H|BeTSuT;vMj-O
zx|UP6GiuzLJ2ro>Po9s*=gOezn!vo;eDn)2LSk_drNOu^1evtCi`lIUrb~z@#ZjrX
zfJZMb#D>~jSTb`Sd$I^b>-Ex$*A=}kvpuo*v|b~fS(Ki8E!{VSf4V@RmH-%rfnUG(
zr%rAY=lomF@ud=%`4SQJ<7uDkIYRzec;2E_`aghe`ojE#hpcKlj9Bj=zi@Xao(UtQ
z0npi7GcnjvgKCEb;;3yB{7wjHxVL1<V0dY=ESPR2`%x1-D;#opkSSVG*Y@UYc6)bU
z+@UjEur*#uV)}pPVr}ku`JTZO-JrKF=}5Fuy^#eWhoYGzXL*xp$#SxE*Ov-tr8el!
z)&!nS+7^LLBXVdr<1Y_=#CqdE8GZjC9rCEcp{2#4lJ(BRx9zYQA~Q2Z`7?j`xx2re
zgw$8E0?r|C`d^i`|GJL<c$AfEWKZAsDjc2n*`3C#%?C5PiedNGw_`wo!}=U4+Re<&
zP6SZ-`{XgKG~{iBRfyDdz)yof?ex|@%!WC!+zG^@UQa~c4rJHf*{UO=6SKlAD35{R
z(`E+Vnwz@D^0LHn!W=?(1rrk!Ep$;yuc^DY*8{CG390d?>*c`qaco<s>plgB;GT2z
z54`a7)2??_Sc8^qBPKsRIZ^^0Mw@1iojE>Ltn!Y6?w&rq1F9l=UZFOP*Ud3!x&+$u
z5Su*|eR*94VaA9{?2ZV!-UKN9Bbrk&3(bnf%bj+?zeYKDSH?hvi1XxsPrCL$Px$`t
oEBxPk@xP&N_&>3X3YOMuzpkDeT19;Y#AS-2+!@)VU-W+cUnM$6?EnA(

literal 0
HcmV?d00001

diff --git a/examples_skyline/accuracy_vs_fpr-skyline_formula.png b/examples_skyline/accuracy_vs_fpr-skyline_formula.png
new file mode 100644
index 0000000000000000000000000000000000000000..172bfe9e4b8dbca7a624a56a758a5f5fe1fe0d92
GIT binary patch
literal 43487
zcmeFZXH=Ehwk`URQ3OnYBEbNXL6jgFBa(BJpdvYE$(Tz8Q6wl5mLeGzIY&W3at6sD
zk|ig}Z%nH8uCw<!_uPAbyw>i0?NzI)1&bBtH@`W@=%e>O<}(!~8Or_i`!Niol)Z6P
z4Z{eVF^mL9P72SswK})K1F_R(Sq*ad$BX>#AMpD=`y0AW7)Egz{YTK+@3RF@UUa^u
z<E(CH>g;yg(FC)(?QCyt=WK0p=cKENqmzZ5?FHWRy!_lJ&7Ga?Mfv#t{yVRoqZwb!
zqG%6>oy26XUedT1@vGlmKg48bcVX?)LDqvTif?9qKDkWD`&KW`Wa>an=*#@q78kyD
z4Q4ASzQ1OitD_aF*lqm0%DVN;#Zg<!*fS=kA|vKTA>2G7<f3Ci`wzdc-`*LT<~aK3
z#Qn-e4@o9*sx=q<W~tOTDc`_Aa>FrWVi-P7PQ1?m9<k<E2XO@ONI?$sfccRSo5Mpn
zA)x>~+#(>Lhlg}Oj1nGnr2Wt&ai5eXf(P3FGv3zzV6-lbi-qNCX=!OshOta)DjN@v
z;^)twL&x8XIK|nPtZ2!qnVQCVd3$T>>+83T7BGl7UhOjSE!o=IS}rG|Ztw9(I(_EM
z6`}WvW=&BpR*V=uud*8*h6S`dq?Sx@UmBYndvE9J#u2igl2X&w);1M?Mnpu{rYE~)
zvMt^utJ~7W08{DxPDF_t;IB&f^yy2#iT>gEw#0AWzF9W(9$WJlja0gJ^{NJ!xNy}B
z0q*u0X`+PYU)@<9iRx_Y>+2k+PhYMJ<8v!n>%3`~=(#qFU)x!V-n|h-#U!4<Z(NmN
z)0<nkwOHq}*6$|I=~?UNJbZik^FuN&4i4EaL(fz%FR$h8wLz=T4@mGX&cD8^+Lf-G
zzT42k>nFZ_do0JYLoQO(rFCOvdS$?KF41jnz+|HYK8~*c$ku3>QIhxOs)^EVol>uY
zj~_p>$oA&is7FhBw<l?G<ydto`EImDbLwkdx^#cq`HzVSbC;=h*`T|5y(@iAO=#e$
zH}g7Y@&@r+TVAqK(N7k(mYR~`6AQOjd--%`dvdI34jnorB_%~dLSoSvAu@{J?a(&T
zyn6MKkljFv>r79`x<XwjkGiw-GRac#TGns4@D!`G5-lGeA9^9{w=WsR#-?_^(ixeT
z`w{FXCs%QH&TegO<?Le$5^$N)*xsJ=?V{i-K51lRWb-}!5`0VHT<K=}hbv?*6U~=k
z!lL)@-w&_8VAiPN<yDLp1I&1$Od!+7YOfu-h{@+%c$FJB{*c_5IMXFM<60zS{e380
za(8VmjB`_h$dBR}mQXIWvq1eOGScSxix-NPmPrc>3;a*17|-wiu9i}OAL?^$dKF}4
zpLOTiMb~`$b`F-2psK3sSn1~Mg@S^DvG{18^Y4Z2gsP4$eL151_R%Bau1so%i?>+b
zef)UZ*q{F)DdqW7r%su!&GyG7Bz!wcOB*g|)%kXNbM1Xl(BU_sp{9Dd))u*TgSP(s
zV|9FG0;zgA=b62iUvu#C3hC<Vj!mbP3Bh&6xlDJSm5}Jq!NZMT+S!~lTF!F)^}WSw
zxkUy)nPFTVJ|5-Sc+6O#%xAm4Xs*=CpZ_cyTeWq6VNR<JTg|WDye2_LT3S=s4T`X1
zM0~dLkDt5cCo3m+zUo+(Mcdi76x|C;^Oe-^;dZ|X3v>It*pp{xC%L_>4p+gZ6k%`k
zRPnvgI~?vL{H`py{%Nv1D}H_WDYHL+mGj&{e~Lj-XO90(+k&t4?=DmyQ(qld9ccP!
zT>TO@SSi1|ySr&K?$X$687BU!V{bx2ObCgH8pxQvZo#}a$S`}~h2IUPQ@#arsOv{}
z9W&n98kc%+g)gXr9sOks#uewgyEC>>&75ZMDZcRKNL6odZ;Dabz)>2ScQtR=P89Xq
zM!An2SJ%-o4|?+^px>4+t3+z&w{h3<SIN1rK1~=-{MUU_l@Gsv-$ToXiHV6`%BQ%{
zdGaG{!AU#c&G$?`zi)r7s;V{3N^}};+)qK#2DjdMqWM(G?;kG;e`VISf4oLv;MVV=
zX*k%Ps7`bIIES*bGF;fCaM^l*sLK?~`-Zqn4@oV)r{C@=cF$t5*x6d&=%4e6|8V7L
zZDr+^OP4N9b*8>AE-4Z8*}BK0mwh8B+pLMn<nCSbFMk}OGg$lb>_|vZ&}rG)nwl2a
zUV;ukuF3|5g@vuo4W_|F)^v4gZ*Qz5H;%RUyA7tkdUcS5l$4W$qb5gEc5!~b`Hiw9
zhn<~Wj&=9q1NV~R;^3$#d|SM7GOV1bj^wau31k1A<#?&dm_W(aIK}tvxi&XtThV;?
z4#PKZz#I!&w0`33@YtNmOSbC5TXtv0Rt21hj}&v03!>%G`;?!5@$1*GHE=0co>FN-
zB*}K2(Ub437`8XB@;_eq<2e`lObNI<dinOl*Yc{%?-L1n{kC`a@Zgk`G=!bmu{_x}
zIyxG=@5otp2+M5BQP~<#KfTzAB9L~66veqbTQtXc_Uv^60u&LNXZwp>etqX>6u;N(
zVp+ilk1Drs-(Fl!8Ca=-O=S|-m3lMYXM2O=+&L9^i(Zjy>a@m#=a#SJ9*+F@`ohG&
zrQ$J#pl#npcTZ0v=l(Zm6<ae+>NHeTa3*!3iP~v~aF^d`(07(?uik`NQq0K6(91Gy
zVD&9cPd~>f>ikWxA1)}zyybGx@~^C>TDSp_iNb|!!dWaPTVv7pRa8_6m^Yua=`ZXM
zoP&iJ5)qLC%UNS*X9vG})2>7_Sv#%s-37DdK;Nw^L}XO$A=-wzv3k9Ec5@p8)z#H5
zuRpX+wtLQ%jBi8&C}}o}_KvH3N`1437J`qX&0uYAwa83Qj+6IbripxxMceg;a6u<U
z`J~RTujo(=TZBv<TsP@F)hIUCUhKZ)y!G25Qrull+sHe|pMU?peJyT--fo}8VY+JT
z>$%rHrq%HlDZr}5PwwulmC-G)kCK=|Y~7f#^G$~Rs3P-{p#`pkhK`QS*x$-vwBw{K
zEIRe2v4$M?B~u>5Vx2c^%J}7_>sPO`b8{<9cc!+%RM-7VoN0}Hp{Ay`TuCi80o%Og
zdxo*gSp7S>F!$1jWYmSFP4)E}a80$_zXByz-#R)ug}i^CEIe@UO3=|0Cz3<Bw5Ogj
z`!w_9^?h_0sVdmWv+dWj>P%@t8Hnd*Zhp_h2M^lIe0||(1qwRelvdXS1&Hlum7`K%
zK!v4{fnp$M-_%orFZt;#YS2d4%haJ7GANw5ICueLrzW76E`S%EdEYQ5%6JPhYEOx0
z9?RrQvDxMb5vOeLP1obb*}l6zB9TffGrii{hVHR(%F%d8FYHoMgESL&?%Y|PjF0BL
za6vs@DYDh6Ny4pVg^rds35S!07!nd6AK#gJQ+w=m#vshBUiz*3LbiPg5adnF%&g`H
zO7LrTMKePE-fR7fi;GE%7A$Bv8tp97d2dX{KMxFa%i)8N0GlM_>C>l#goG?1QFm^2
zK={X}q;N-Gbhr#Fv-S7FNV0&@?b{z1sGcfCin8+Y@c{~$day#xB(6jC^ontne+z^y
z{h7Re*Y}3^hDl)U$itHv`b+DLGaWmy$pI;)K#Y%#jpg9ts%v#Ts}zx77Ug~&ZBNJ}
z%kApQn)hBCxqSKZg<Vik^V+p*8*62|scVD2Wqm#&+kb7<T1fsM$e8X)LnhNMa-GSx
z>f$x<oXH82YYI@3lWYAX`{w8Pcs%5TYmhutVC>l*E0!~ZWkw%9eArJ#)fvPp&ms$P
zqysLAhK43~pu|(#Ln$*lI@-{CMKj=puoA)_u(Sj%+u6G|0eEygAfa&H-CnC395jNo
z%9$nOwD_|I-NjfHA&8aUi#2S83somN;uKG4ZvrSnAsW!b_ClbqhNY!t!8nJuDGUsO
z4^h`&2{3<?D4{GC!+Hwtx7AM8N?ooxVfSs=UccFFqWJ>~go2JgZ*&=`UJqpA0od0P
zBU1;P*0Eun0(Mqy2-gaPg(SH9XrsO0D>YxweHbXUlfSV%8IzE}zB18rrNm<;Gh2dl
zX}L{#8uGF_Y?u{53L5uj-Au~NFcqfv!xw4%u1^4vXr<BC|F8>*Zwj_jQQhl)+pfW#
zRpEZ;Hbs}xSv1@R;X&uRLINEB9clr;>yv4|+sSu~ji;~ObB&Lc4};V-3FCKm<G`BD
z=h^MBfS3KYVf@#*v{KmHn2sOU9TNz28t__DWMyH|(R*4s+`PNhw5#9DL`xed;kl~A
zl^=QlmeJ(oB*gFAIgQ=95b^T*Rv|Q643*!{DfPh;xFRZ0L7`L5!2l2c*Hsk%SCl>f
z{~-UbELZ7M64m3UAfj}QHN2-`Vw#<G#L82Bwx^Az=X`gIis}+V>+2QPH#U;;^S}4U
zVw#uzu0IsEACmT7Y7m~~jp>B4Q&zao{^8d4wn<(u*G!qi$aMruBqb#aho78)q-Iri
z1GCMt?H5&icYZV|l;3Ggq2FUVRZv6q>#x7c8>+oR77ta1*7P)~B_RQQKu>p;S--*M
zoM7!WlQ4t0!R=W$kA<&vT#>twqv<c0`bYU}SfC8!y|LT^;n*rh19SQG;iZ|h@l_Zi
zBn7IKsSbHmw*h8eTv}q4wXm>&Anv-7)jQ*{vp$XjIKmZB9d*zdd3_)tB4s~$Qo5z3
zMNVGcX@yz*UOJQSmILLnbJ75y=tUeqy=0ccL$QQPSU_F1-@e(e>_UM_bK(S-udgrQ
z@TsLHsW28E-OQ^nDh~Cfz5*wd31rovUS;x{zx4RYlO%&8R|Lj<Ac3Bb9$p`h*4|v5
zae=69G`|i%B?FF02H+fGn+$l5<NSFwsO~IVK0Y>Wj7Ur?-H?NVJhi8aRD~}>D$0Ac
z+l-E}Zf7;GtRQaCXVb=IZomMgp!wm-I)=wRS!RJxs2G#V%cW86+$6b~(wSzI25?PF
zK9u`)T3Xu2m>532UcmYSFib$^Nl-<=y~)#`Ylsk0hhd;hHwrn?u*73BIjuB^MKMZT
z_uad9XgYwwSgD}kL!embeZxxhquNjjz^(nRy*4o4>UZyc+JSo5sMVE}`+_u)ii*mu
z%U7;6K)q-U5Vir}qnWsbgrb^SP<OV)n|=HDJFZYtP+WldKYQUq+t1O_cYq;|96M$%
z`Fn^+U0>fCmBi?GpWnk|e^sf#B%I$|n-d05F3?}(CTvjX%+Kt*mBGQuc^(j!5TG8x
zs$))oFOM8Ncx%9W%|Km4<1Q_qp*4V&dRQb-_nG;_M65&kZIA;<zxvasPsfb?3!rST
z|MvP!!@@{)13<KfKMtL#hd&4KS6TOD*9cg3){8@hcNS0<j8rq9M-u9`Beb-xEi*22
zC2QdpZE+zMFZ}%l=E`<`I)p{Byv@;jS|J&~Qu78-@p|X${x+an-b6*&K7anazD-%m
zW@By64(e(6k~jbkaJxnJ3iWv9Xdxr7g~$9=Z`q>mRj{zIXe8#|Cn5{qUEAG>UI%nn
z53Gd3dm)>8D0~d+7M7O<U<6Sd+}r|urN2LB^v=Ykr#G(@pb>UuQvFruf8^wWYP8pD
z{xjJaa6hJy{l~(gc5PRr)Tv-H7a&L92}5%p^j@o}tCNSdplNMw{dgNb+wk$DM+vYx
z^|H-h|M>CaD8O!NYHCVGQ4O>ip$atEwkz9qMcfRO6dNmPzNrWZ#J#r-0>JrWWF!Z6
zr)(=AcO0)lp*aAe#pUI8sOLITZwjXxmgu42u|D>GudsxQK6-VwKUp>Add;zzpFi(5
zM2czOzkeT=SetR6L|l8jhBhUX#vXvAMrUSHLwRl*wyxGTG=!9u`Mwi^jXC=Br+%v=
zL3@eQh^}T;rqjT^5lV@TiI}N@k|J1vE8mR+Eyf!o9YrIdkP3kBOG6kDYTZ`=?C=F+
z^zwqJio2O>-2(eN4XR3ZDEHj@9G;>Dr5z0@(CXLs3_!@O(L2j*uwAABRwcYPF@nMy
zr~-8Wk^=6Fce@5`aQK6`mXi#k&Nra2&g{0v=ccEp@85j-=uv*Dn`op5OzzQx2ZuH`
z0HQhnp1-Xf4V+X#sa<#GC`lVoH+a!TBRd_Px0zX4tg?W`9Z4p>zMz{9y=nV8Ffi~X
zlSEQ+vG~etf3Id`n$HRsql8C$P6S}<q-!r|gFc~+520*vWu+tAqAjWP!%iS9eWhqg
zgSZ)3b}*w^#oXAei>-Kat&ZzUsPUM0f6Hi>LHsyw92^;$3N=v@Y^gAfHV7H|VnmJW
zqdLFq3S(OtiGko0b)HZO;4Xs2i=XrM@(PZQZW8&CpiT^n2^A(VW)pvh91~Mht?#>G
zm&AY~g)Q0%@y@`=tTCbmM!?h60v}8<Yx?@#yF_4ZWLuj*Tya^RP@U;3Q0L&_2p6yj
zVo3x_2bKA&zq&_~hJn?<-}3Qk_}BWZ(lPS&>(?f$-=Ipm4l6hkC?p;}3O~=t6DLyq
zO`x#!fc#arybaq&;8r;X3wmO6uB>dLkP|MA4D!mydeu~;mygvZqkcNc$;kzWgv3(8
z0xyh8;yz(7OejGAXlHw~uB}ZSVsaAflWgnm^MI*S<(38@fa;Za<VbGKUxA90OG?TJ
zGEZ%7ZKl4AzJ5e0)a|GuMQI1>2}F1tI&=L-P;qhbQD)|}r%WC<p!jHp3ug_P^)6Uy
z6mx<4VH)bpS1eGKO+%@gvREh&LyC@01HKOK4Ez8kz2hd_f4w5Nxz5&D`4D(6KFUK5
zXg|c?j5Zy>`J|r@g{f0)l30s?xjQ2Z^iMEQuoAMZ5ZuseD_N}N;$&x6QBrzYy0s`T
z8!Iz%{``3jH8sZ;y~~#`qlrTMrnaf6Yj<aE*AloQ6WjZFP=%rb7hMonTJdS1b5^Fi
znsTC^U-r@rrOZD;Ku>uYmEtf%C2%?W4;*NR3l}i0f6Li$;_%^*sEqD5i`E3Hmj-@?
zT%6G=b#Sx*NZ}mQ2KtiaR(ZrXqg-PSOoC-wTtIRY)N0wcD@ngrSHn$c>)1g=FC?$2
zuC!>q9Lw1EwuLpV+sRTu1;Y2LIXE}~lb`|AuAuezb6uOTI2<oB1j#9r=@vRC1_cGR
z7rIz^uaAbAH0F^u&+8nogW@SwhROZvk+TXaP-8i-bZA@k7rF$o0D0`awNQ=jFbWkY
zfVIUb27LmOPBzxqAkDCZS5H|^?m5gH;(Y_)o*?GYws7)n*QW5GS5_#u?nfxPt%fTe
z1C9C)?mLTYpyWmZq5?^XiOnJKhlWw+zi-%De0veY%wOTlc0BqWsu>Gsa(nv>o+BUt
z1*}!e$7{2^BndP?5n{Oeb#--ZQ0}HB;|ps6rzsK=6f#VArD@;0moMrzn*wOcXucG7
z^s7@>fPRbxzFvD;Uk|evt}3hoLexhEu!kU!LWnFBEd1r<MkKmx1$RxbkO}u;_~^?e
zi1LVF+xR^aNW;L8^8LF`?59ucr%zXOumMkY=DL3~&~nR7zB}p^wA{_WVfE_>Vmg&N
zWLQTv)5MZ>Z>}Pca$40d89R#IofO}myTql9=Wb(@kT3v3C5C14$qC#1A#W^U?$|Vj
z+aPZquW4-Lby`{*8X7t;Y!~(k*nYIkg0j8V9U-Jo05WmWW5e|KSBYW?l1;TmLtai}
z^(}~IbQ^G&L5j))pkGpSvnV!L+1R*1BiZ9F0fcqBBvQ%&K`t)-TdMzSdO+VCd>EA1
zZvpQf@ps92>RT6Le4;>Goqjs$A?6*0!{N>=E5jBp9Ga8?j?I21q^6;v1NIrs!Grdn
zGY=m*67u#fyX<~y>OAMB!8mqMd>~f3x*cDk@BOw7xpidruqCl^r00TNfur@AHA`+|
z1DoM8*YRfo0Uf2@o6bY!1T?g?oWNbgTo2%E8wZe!KhY(%*5`!S*zuQY1h}4m%nM|A
z#F_~JONsE_NL2vBTCklyXeMjPa&8+rz)df7ok=<)^O7B?)&@-?KS6&NOvSse3jfkx
zC(DrmpUm**xMF;FHwOX758O#frr+Vwerfa5(BFlwTBO;J1XvWfWHle3(wm)gb8~5F
zWjj4Q)3%n7_CE{_-2y(Ypy(*BvA{Fs&9S+u{l*3>!0S<DW@ZA3!dBGJ`@TVIe7dQo
z=5=(mQDj<PB7YUSXYi3fTE3YF^+|Nf`n$Q@&%ke7B%@D131KqK$fuxp2Zyu$ZY^Zs
zvp#Cuze$>Z#Mq#%Tu0lR*u4Do4o-k_E25=+!479JG_Ef}Y>vC1(QWN9be~MW4c&Zz
zb3A-zU3i1%&z`9&DEI?G@6w^E%>nruC<%2+t<9|LY=MAd{6LNZ`8Wa8ZRt;w#<%=e
zP6$~i%x}|hX|YR4l+<n$v@?UKaVKR{%HL&$dR&@_(myXR53ofXB1)m?NnCaGSf2Q>
z$1}$%MJ7QEHwO?PsA(E`Zz?gKcVp){4VS>3FXTvROO*24c$hJ3hCa&^cVF+P{SDWm
zpTnvfQ}+5B=OrX0VD%&vIE@?58(zQu#I|_hI?TPrw^xY%4A$6?8G$g};YUE!4tiqA
z?#?C+ukZZOXZ%e%Mln|f*CGSRQo*3iO=y>GgdlkU_VahaCKxFo@J%3d4S}lDwzb8x
zSO6`G@`1<7)TqbD{;$OlR|_|$l3gIYO#{p8kOL934VH12OP4|Xvm<Pl5RLFnF0)w*
zfD#TJIs}=n6{(O-QoH&P7F(}SOMC)x2M9tpla1YJquo{*=21q*R9MuUY;5tZ8(45)
z#>)(R;%#aH6?q0K4t~|zI<81pQTcdQj#I;}q265vZUEPl>dSd$VYvs!DZKm0p&mB@
z9BOCD+CVbg5|AF5B~ze$QGnv7BSAGLDTmAy5F5y=PeGZ0vK8++TOb3`q5VRWcnswU
zp^3~o-Zli*U`gpxOKqjk7B4AngP5lQZv!Rc_X}Q62zZF_M49fmv2%l9mpO<%qMoZM
zp}Yo3sM3OafHVk?wOIqxfsyJ!tM$dvo+7t2YoYf_QR0byepmp#U{YU+;^W7UQIUxP
zJA}nZNpC$EN_MdXcCM4ZQ&15Cw56<CYm=`^V@(~GO_>I)tX7-6{y4;UD{Mk801bmh
zw)JPtn-$1|1p`H+p*MdH4}bXM4{}DRv-GAJ&)@w63hWLLb6`1r1Lhn6rizD0A--f)
z@9UQ@6R1{?P>M%584`dTltHf>OxvK=@Mtmdg(8t&!Xulv=+}8D<Spyp@*}Ay*Sh;7
zs9d^#;oR#l8Qeb?k6UhR&xt0-Yy86t;F|_Gzoxo+{NwZnh+@v8A=;cmLRxB*tINxG
zeF;?<{ayTB6b}*5C$R3iUCpSS(SrEv9WLdIxNY^=oq>S?{2bpZrUtT8NO16tR+2oS
zjnVT^>BnW6HPs%A0la64Ze@W}d}~{qpzn?s7xQQks4tcP2k6CIQ#w<0V~n<^v|S2)
za2&=d$zrgH$k6wF_|Vytiois^!$|FcAtKTPDxih3)PS-B#9wHbRbFonKoi-*iS>u$
z%hO%T2*AUBV3kE^GjGr**|xus1BzFK4_P=^Sw}zC9lvOg(@N3#04O>VkR!<8U7!cK
z%;a=g0-FH92kjylk<;g5K;sBNLH#4(HmmOq8wg>8GRLMh3fMlwl~1AU*_KzEBs$R;
zC2@1ojKuy_m1tFA+#eWadysC{r=>vr`tRV}u3%JbKvr~8EL-idLMsw<?`-?w%iE)T
zz8S^!NZ;@UKBfca>wF#l229Hegf8<RU!K8cRNLLz&T*SFM0!t6P0dDwolgv^)j>sO
zzc45^R~!S8x~mJ3v_SME*$tK!LYe1W&FGc@c<ni4AJ|R^5E$)lbWOFyJcVUe$IZpY
zCIcB=MMEPPUo<O<<oTdapnpw(@~zPnC1K#($C3y}1ij3=&t?n8nXbvmSONDeegjK5
zi!SnP4Nl2<rl+HywcQg6Q%ARbA#%#fdF{R}r;RHr{4=`cjOcg@vsXZl+i0b2mWFf=
zHMfY<nEzydQC@ecw`XKZUU4*d14^dv_EXXr8+5d<Ej0EUNAfE=6&t5mUA*{)A>rD-
zePKQ{qMyEexf&ZEk8rknoPsI<us(ysenz)al8^hlP{LA?F3G#MaK|kS!*U=H4G%H=
zw}uk>SFqK&AYt(H^IHy<A`GG520L7^5by7QzyNM^(5F9h$d90~b)c>%5aO8MhUdJ2
zRnnZL#yN<BQa$~;+utQ2FrR{~*-sym8Cu~i79cL6@(V)f#8^_2ZBMoZowpRA1;j9&
zyCog6L1Z4mTejtB)17$@xGVUCFfekopRVS8`YY}G=)$MHhwq~a3oZ(gqB-@(=HA}k
ztRYm*05>`TG@uF^nDTGO&dBzZdKZyW(tkqb>F@brQLo?W$r7TXdO)yvY_D{28MpyT
zML7(39lb#D#Xpk00nktN<RotZ4FyX_9d>fwr`BdEvqlElR(Mqs1SFK0Djw|v@~ioT
zUGWD<T1fi@Hvw-Oh~3+TQ01fW4{I>Qy@wVPcDl$?WUx5T?(MP{3~f;`d}bVW<yrEP
z5ba<#S;r_xP6wBdEF2$X`o$e?!A3x=6_k=VWV=9HR<KJy28DNkiyE-_bMDq0W@Tla
zMEL>fY{(C@F`d>0naH`tgc(X}bS+>`z$1_iy0%`%oyX{kK&;T)U2pP55)TYr73P7i
zUdnUTn(?B&G{h-U_r*`Z3JE}RKkpe1I@rWi^s84mAb+fuCtA?nmz9-mu;s=&M(l4q
zd!}YMslyhZ<*b*K6`y>r^8@}6B=(?e{u-L7rW&qV_13CSq7=<g3~Sokr?Vgga{T!5
zzXW1IsFY=)D9OMVFUle16Zot4l`bPb*>G|9Oc=NJaON(=4-3fh5C>B~wO-l<iYSDI
zl3w5ja36cshBt7AboI=zcY%}acdX+ngtWoCA>!!hh-!6^edL1xs@*oZfF&r(%1S*Q
zcUoMO>z!tozVMIHx^uRvdAdNbEnHOzhJOmmkwsy}8wd;M)nB^DjgKBbo(OzmsBR;a
z=~mhJTErJ^fGm^)!Wh(I@}Qz6wXF~(oO(y&Z$XozeZ-h?>PRC>cnQ7<v8z*=y)$SX
zvRHH^X>xLM7MkzQ&)XwfVnrr4A>oGS>MvDkX+N0uWypN8K@1@gmEZI1jF2pb@Hnt1
zKmvUPhXMy=ooKJcXFgjCGLaIVTJmAMCAS7ZsYKKaBA}yucie#vl!Lg8*heL$Mw0FA
zt*xU_tL#AtFx~{a?J)6SvxoFPztvLp6y)#xAbCofDCchHE8WF~MUJ+yk<l34E2bS2
z9tbX=F+6|yvJ-BHOE+-P*}hN?t$>SV8kAyHF#dqJs-LHhBKB<IG%uv#eBKHXGl1iW
zW%foE2`IA>363@d?80eCpjs*_ub4c4@gubsnWcaM-;21(`T07Kt&`DyjFRx20>GyU
z69ZW}5$t3|yeaFkk16P;kXZ?WGU<V1IHZINrmtZ7nYvjl(V51PUS4v&RtI7JX&#EU
z<XCCDg{jQJTCVA5)!+mZj@s_E-J)-g8YnO1z7skkBwWwEF?cO3dk5SWMy*>eP$5-=
z6Gv6`^>O|?k946heb1E?wde!mUUR*|N=f!t44%?4qvbh&mb?Dg`_yHZme>H7Dgplm
zh^v!77|e0BtjjkNvM5NW-vF;_S^0z#aRuqOD>+N-Nu`OxUa6ST9q42O#6SmOIY7)!
zN*yz=KQ<lC3rprNCm-#*CEkmdM!qy1S7h8J4JY+!?6~4b%54r5IPmX{er?|XtpTCS
zbjR=PiwwL4#Cb`0H#a#Im6n!PC~zQCp)2F#;}@DcUISH^A7xD{AUMl3%~DlVSfQg0
zHA6WeD9{E_7ueqP)zb_2Sp7AbyFz46|7S&#FnIAXvJ?KF>@ocRh}edErdQ@uQc+pi
zQQ&0O5GAqsVC}+%3oezE!h7HZN)6zW`&C%6b}2A{nm>SO0A-1r1@JHcC$u4b0c5OO
zb{~6ZCIG_2%=e5diHV6DmCQRsah3!)AK2W|N3v~tg<NL)bfHwzKzzpS3gT4#f=mTw
za6ZA>mk$DJ8c}=sARr~UiT8sC6*mh+8(28M2r@$S;49n8RES3F8Ip+yKtboR`s3j4
z`+%<__VaiKPE>X@6jTHvD0ef@wgZrSj^~=K`@LX^d+)$p1FW@dD`@Nq01|{XdeeW7
zjb%o1eipjyecg}L?<_kz`wcQ`CTB%66B7=0_5_9K^`&tuDAj^c{f2-Iq{TMCaH?@e
zckX-ywt-c4czBpz{GR;`?I8K5*w~CeKdnln8EYm6Aqs$tx2XM)9|$cW;o(WZorg>w
zp{Gv*i-KHG>CT1~kkrXQSI)B2VN*0<trfTw6%`dOY!~%}ni&rkCab=Dhv)wO?f%Cv
zWLtJ{%C^e$m3)E#B@bCM9N5tT_i@H<YawB6Dr~kIWcB8&;IabyA6}tXU0GsuLlV50
zkh~M8GcnwNG4=nhcXz@Od~$|xjZ=LES*A@<-k&!?bV9MTJ<CiPVG!_$*d~Dg8Jm^%
zlT?OOiefONDkX3N0JWp);o<R4vPpc&AGyKc-XmLx@9#?HJ`Emt(-b|yhB$x;btpZ+
zKq4b(H=qwpUo*HP>Xv~~5cS^3W|;(Td8+xtLoqoLa1ET8XP;ZT`@gs@;Pvkh;*w9X
zvR(teyn~2Z90y~w{LA_wE?zs^tgNJ@<>;8nGFi>+s{vu?M3T)#Fc@`T54m-M02}fl
zo>(F#A(<Td{1870LKxVcLMAUjd6Wth6NF4fNM``Vhb(Nh%kl~e8n852IK;*E0Se;#
zf1JpN#g?ne`u#eF<ui{<N3%f&g_^ewv?C22odnQLQ2DA5-Q28V)0eLrv<v_Qtu<H{
zE;0o$NeRX<hI>$n4HUc6OM2xOy3OIi=7coK3#RqWNt($lvV0}WK|px(pb7vg5|8mn
zx4T*0eflD3k_qGOgU|5tDgpay(i7z|#nH6}6haS}!VJRDlzR9{{J&`NR`I5KdN)s=
zi77UX>S+-dA%^R8FDwF=eQ@O|*0XZ%?y+r3W|P@0tStFyyclMA+!IHjp$v8U@OngX
zajPqa(eo1u_)XQ+XLnLz(q}IDk*Y|bqLTnKX92N4>DRYIPjhW^xOf__+}sUKK2x^J
z+55R#CS0&i;wu5BB7<J@6JArgepZ!)L`n8KNmTVp!$Tp)<GuxVvt)1IR&zX0caRXH
zmzb>}z{!>FO3Jdbn()zcFR-_-`7Ub~wHw<e?Z^BItC`F(O*Ivjov-(4a)_{IokK9v
zXL)YxbYxoXFf`+9Ff>+1vBsdMPnF*ZQByO!{Q7q7A?6qN7^B3=ftyO@-o4D?l1i5D
zpEHbc{Y>i4igH2XQOmEkX^I$%?ga~DxP_6^mod#wyupK>9IleOCjil@Q<XFfRVN=m
zZd6xd#q)XxyIrQ9{Q_5CjJ)M~K|*H>Dwagtuer<he2Xb2zyg5jm(IBx#ko6s;?Hpc
zAE+Up;JfZNn2^AT`BgoD;pmxCzGG!h%XAx^Us!BOIn@^eSCo&gNd9slF>9cL)?44V
z)|njjQVKO^-W2#9#F{4@=%}!c(B?Du%MW1Yj_~cUy8g5IUn9EmRQF8KQ<=%+m_-?P
zZDU$kgEYLSiB%-1o(8{VAKe~oBaduswUCu#->s;ytS7{zODbV*JKh8@I~|r}<4emf
z88apD+qhq8_M2oCZ1GX``<c28R6SQr39z-1`QwdyFF&I9`AiX)W!eBY>{ABGN@N2~
zGVA27iT@Q-JUXJ@W9XEH`E5G<J}?d`C?V<*F^7omeu_8NxvgD4f4*R9OsV?EJAwL?
zUSVa?Q7Y?rOM>-*c`$QaTAo4D>c`$y+Vi82-}_PJI92p8-QU-)62nl$kXd#otojWT
zdN8J=K7ln0FpB(W?o4%!jdeGl)+&0*a9iA(8vu=TfWt3Gb||zKpUd%OJM0x*la-{0
zJ?tlrNVhrs{Nu`?*BdbYwJxON$`kEMIT)#XbBx8UC*TWlRV+3Nj*p>yc)l2%{rV{-
zX08W|6%{(bE`#cDfk}uI613piC5?4z8g#4alOH@-tS))p(~=Hrp#81N?jkd%imDhU
zpfxZs8t6%r6LT9)4S&{?`k+AVZZnt7d+X+%2bqL0+%tPrGH&Btk_L(2<G;@@-DF~t
zocjKHA)Ox@Jrel$NgNIqKd*ax=XPdI+_~Lzuw=#vFJp=iD^)kXT#i>4I+Jtp&QZyD
zfvcXLPaz(@qaZ)A5ACq)08USyBQg&U3aXWyy{VQ(81u)q+pQOl&nE86d9W9X1W#o>
zpdjY5lz9kS3L@aX4>bBGbyUUZxl#LxwBlPq`9%)TQ#lB0w<N%&LEN}}`9W|5_v6R1
zLN_O^933sh=+T;kH*@UUM?7}+e-h(x2cBNZ>}916Y}I~R*-w$=Nty4ANiz=dGGO|M
zEnOIAYY~l%Z43JPqSng#l1myv7!{Uq2FmE6AXv5b;%L?OOZ@YVFst?`Vi|7!%TxO(
zbN>CQcW^BW9KB<(UT4)pxPJZkA;)a&*+LQ`>iFeD>dhHB`6aZHU_lLVttMr|o>QbJ
zp!bvi>vBWJ2Gi49ExNusjpNguL|Kh}N-Ji=PW24G+SyraU2Dqd{=}iqo4<eVUm_rh
zlJ}D0$D3E2V|$^~=O=9Y;McECn%up?${L$y^1xUVpU#yQ$Pd7eE@gUi?hS~KxpuO$
zQKh>BCEC9XoKIlpYX83O*K&wqTr`>m6^#{|_ee-a*HRN01rdG%B3l~~zEEaT43r__
z{K%0QpsQVGdNjd%4FJNP8JLyB+HAG%wLk@qbHu~Xhd)B7zjX<Ql4Va51!2}37DNMp
zn@$kNV!_<R0h$%4(SSku+RSGA3J|s9I@_0sSPJm1I;iLPdBT=aGgR%4x%+2)^yYal
z3dtlXxt7M27GE*_-60fE!0O8UylVAJ{S0luBE4~qkz&aD1NpY#tioFy=%QNRr2s-4
zVN*(}+)uDpC$72Ek`9~783k5@KL==7<hpM^wQ^{l+gld!=~o_Qih@R#i<9}e{}IIg
z`t{3k6Z(D-*8y^yu|Wt~JH>~cQD(~I1V%-&z(vwqikB~sYIkVHHtkw;@!l`jZ;FkJ
zvjj=Q68O%7IkvwPwB?aNDNY`qRH|ev<cIE73Q3^uofw@D8x<ebc*fz7<NjUz=oqt>
zUhlIbYOc8&I_!dWyd)&D_wUbg9dmgCXju`QK(C;eXd=83)Ft?(z*}8ieFZoi?fewr
z@>tKFl?#HRK=9uD5PfP6Ke(}s#WX*bTAzVi1xbfvNh9v^_7y>^Z`=Ws{A1(29y^Qp
zE0U6ZYnjg=Ku8}!)@<NiCV{zyqF)V2VKDr7T`Gf0b_#(lpsAp!81g}Zq&X%++AE{i
z<1p@o1XPiJXTO1@4`o6dIH)>GJl&9|7sTRle&Z)vsd}SRVSuKA2>s@|^M>_)Tw?`~
zSj+EpGI}gL;m<~c5DQOGe`*z)B1E#NxCrTAH8;1$XM}dWydDSv#?t?v8EWKC6aB@m
zZqqm767oh|k^8WzeH3}Wu8yDtBlu%$+9CoHRboUNLDL*`i9LRY;h02(ud}j-MCBf$
z*Q1X*UKCD@GSnfuqNq5z$h}W^2Z`p)73i+;W)&Acjv{hJ>xw{#!QIMh7XfYgL-xI{
z;oPrKF!RsnQd&qSI+C@M!S%HG60>|1J8e-rtv{)dKt95VwWnS`feQmw14Ql#5Q#g1
z4CptYYMx+a36}KZ;!at9LVx=<o%_iyJ~8>$0-38u&Sy0qEl^=oS}~N(FKBsM!I=ag
znoTiMR1<lw0q+7k)n`E8Y_3}vHKuQ7{3**!-l%{)Uhmi(hO-sy!FXhmIs&*B6>5x<
zUOK?RgMd}C++&qiP$YmQsQSBMDsJ>Peug=gAxNbMDLWacB-lZ1g)Y;}K#@$wh8a4#
zxnnBEe|;)hBDX=I26BUgmcDU*2K+@xqIjP~=oemnj|_83hG~EbZ>p9n%~^35_$^V#
zpML;1^W?zsA62y&&Xj+EK)T;DJKp{a7!E2&#sj;c^$5%dMH#2D1V%KZJQ$KBs7Fvl
z&4$~;7KYlZU*{0kPE{p_CEN@UlFqZ~)dJ1TQ87)wKn<7~2jgp)irBF^zbBtF>P~HP
zxw?x_q6ZOI5n!;B5?i8(wt-(}KXvL7Vq*X7i~zxMdKQYRDj3m9^nXdDlC725nZsyi
z+Y{fE{{BHRIyNR31ZKT{{Xa*8+mY{i`*VNO_)nTq4h!SM1cd~a*#vq)$Y(dxoh1XE
zTln1L7fkd2x;E+UP(&1pP32wpq*?~&P&QDoB_LLuubKdb?oHdIvzsoaG6Gk!!PhR;
zzW2%#IH}z&r;^RtB1>rJin{k>AGDijf=4m6sE*K&9%jAWcQ5&e0b*t5V{V{7U#Cbj
ztR~Qd^@GeM8>{>SNZ?oCjzG>7VQmsjFmQn?VfU~H3%mQ>%!;kbyKfW$1IRJGt6y)T
z)QFKF^Z5DgH6n4DJSI*d?<X`+f)OD8AKxc%05&{%eZbY21I<7n(}Sjlc*ows9WA7F
zfXod{TtBSMA-Dm-+5EEa;?L7eF(&~dxIp$f04=3i!CLi;;^N}Yz(N$nm6Vk5Z1d~>
zJB5K9e?SHzj~IA8fyI*!Xs2v<8=p0v5+4nI?a*Hx{FETSDfXoshE3C(_=SEtb>)dU
zwC0?z(g{xk9uOQDVL&>@FE2YIiv{S>3emu`|3fB2f=j^h3vp-V$)-|Y^j9C9RqtP*
zlIC&vC)1w(=0k=>Ho*os?@^Z_NTk5Na-l{;FbIS+!aDq~7?Fw{<gmKva7Qtes{LkP
zb+%8^>^(9Z?L`K)J0E{K%T)!Jo;`c`<B4|=6r3C#Av1rhsJM*OP7x6iD43mNm;Sqq
z6VO4~Zua-5varg!n*_FV0D55zxE@pdhe{Ci?X@}fg3Dlo{-z8>?W&iI$)*kAA)7uB
zH=sd|RTi4#VxZm9(z-uD)z+4`H{tN?z;(W-W!#Xs<rJO~E`7o=B)Vs4{1{HmHmUnm
zUQTZ`{x4~$qt=o>yAW=rhN>xv%uh}LdhO;!%b=!CY)C@wizopQomRlcvh;BYYVLoj
zzk7v&M1b*mJg$@b*Lpj5eqVUhpY`@srI6z<$BDY}-2{u>i}<?dJ_}C*0|6njw)ZiL
zG&1hn9&L=!(@}h=y?58hDS0GpVv1b_=pjNtoGYwHy|gZ%NotB3SzD*9&h*AZ3l^e?
zfZNh5zPCWKjbvVMxvD})P0I+s|K#KcnFog@<%6ET4D%l8J5@`skV_|aSk7-p*QHtv
z%emE=vbZ}g^@5E`=xo3$!^9E^8QC=I(aDYmJ0|$P<3X)V25FsM*fs);a)QQH2OtR7
zcJRUF0xR;W!8Exc^ZrKSODa`ImTZqC(E)-4l70JPpVm}VumBAlQfV5X6w^=S4c4;!
z3L7xiAT$ILkfz1Je@=_nr@ri(%?*rvOj?q9Sw^<-<Q2IdC%HMx^ZoZO$uEV5n8+1K
zEw%paNu%gBow4h+&E3j%;b|4}#pmSSRFAW1Na{5;oC-5JYl1f^D3GcX%pfLi2r52x
zntXh%`pZxDyocw^c+<v5hKa`NhwIH=uD^3LDjoM44l7>flJhyzb8ftC4ig|MiY#Xs
zA{3IQrWg3O?QL$J0NsUgw=19M>Kr;K{g0O}i(&J4KgwosTB?DL+2{5xAtgn_!Xgpu
zcHH;_<XFcUE?r&xsUNgF;&yHASI_Xd7G=9<P3J|2gy6&8@jtvDnbk6fSzUCy7?Lwq
zrlF#mk{_eCx@z8S=bKb@cGS60?=EyQ=b3kRM-_FfiF?VHH{NFgb36D8CVjjXf65xT
zkA~1{5fdLP>hBn?tk9p-vRot~ico?S=OgZA5<AU5edylDL4twtzG02A={(L|=ZpjI
zgPA!@sCYt`sCV6!-Xz)C&~zFuND-XC)z#J>1&_ZE12`sHv-rlz?0v%BKVzL|%#Up_
z4teXI$keb5_O#}*VKt#lyBIIq(M5<Wcl?JJpl-rthwT=TGiW}@=he1-54H=ErykP~
z5G1^_d-K`gBK;E|vU^6KTf$GMUh<4=`!qFevMjoLB;5%|$E%-!WH*S3A-9hqGeI_(
zpp)we3FR*go_$zkFK}R+YQ8jo)uC3(!}j5Gu_UK(4et_v!*O4@O|VeGd}gq*SjWd&
zphaj6{?D%f;p6_}x30dmHYxQ<t`>Ni?#SV(UibI=UH4ZbV+i2r1W>@Idh^o2(g|&)
z+ci^}e%GBuV;RLhy*kKo>az?lFMrW3zqU!v())G~R>=1*C%Pos;N>$uPDuehp)1d&
zPK)du;@K>bZ?sTbSvl_!$$Xv=I<=-mdLc4#Cwz$oT|KlUx9oeqLre$j?-XrAe*g4y
zm2ahdV}=W__pba7(N4kFKbjI&-6q{yR6JG%Fg2BTn4z?NZo-A~N?9`V_&2@M@6%qg
z<u9eRwX|x_9SE$3{!^DN;-MnKANyzg?j0$8=0k=J(7>#wfjt@eDy~Uwb~mxeB5NnI
zguF;5q-Q&cZ>IV~c=-K<8#ZJ>AV#<i9b6OO3uw-2l1xTcU?`B5bj-1Nwj6_JGuqnb
zWMAITD=v$bpFW#5Ej>j39b*JHCgdgrKoF}9j{SH*fHw_jo7p=kb|P!${`n1lBs!Ak
z5=i=vz62M(Rd=QdayFuVHE{CW2m&qD+`&Y8MAPWm_VyZ{D|_UJenDFrz8TLe#8oF~
zQ^_~@$nL`+0d3NBA=QRnIYmC!pRCPC$t<aFSG@3KkMBm@EjNDuMPJvV;5TP9HFXVY
z+|zs4Z&_T15ntBtI@C^RyT2Ov+m2XyI%K*;{%q7c@8BYe5%1q%ymbmcO>Rj*hqKEc
zCgm|I)rV!=I?)o716B(prEY*Uj0e0>J++A8YMSRT3I_uj6&Q*^DlA!~YopF#FzF?M
z&h7Ez8Cwc31s2Y*O_h$-k&)&=lrTeN@%alEa=@3GEKlHPdaaD8_?vCm;-WrX{Lj%t
zQ4epZ>>Xx-OWb1)up@uA4DH){vYZ%E4A?Dbeu0N`_xDb?c>CuG7Ycu#aB=pZZ}W?`
z|9jD{9K*1C%mnoFL`w0&?>Iuj<e!`x9WP=!@{s`N5TVp{>&yB9ITV7@&s{oIMxk0!
zK`i3d9Q=?3y+v17%_4C;w^gmn74sA6czAeyJ%Bmbe%7AT$tG?waIEgthpm|PTK)I3
ze^4C;?Tq^9(YL+8Kp3bWkh^R1P0ix*_$4_mxsaUuuLEwuXt$*GBMN3PzaGJ_xT_Ec
z-0(JHgcRtTy1UhM-j3&{pX+k<{j7fpEB(kp%KGuL)qn_I;>?N9aHPj?f+7RQi=VVo
zS>7u?4|J2r?M=m}+<wbNULF=PhkMk<!RRS*lXFe{2G`BEnbp5rzLC-gz`OPp8BBF4
zCnW_fE^e`j%h%Ho58X^6WBoC5LYVz^2&<-bmb?{X>+5?p>9R6EXfj$(c}huHHkxOv
zBvW7qf1{7@`aToD{D*y0$iQv4?e))Vxl3sjJNFNnV-5IRk^ATa1M|~Qz-?;a^;F2M
z(DP`dIbc--*RaEJkoQ;BI*=~r9%eiXCc9EK{GUJkUR?lc3jXh?a0`DD+jTB=N*LaK
z3rw1YCf|Z)iQH#ud9Q9ps&3zRW)>D$vqqyCmEachnPx1UcKRg|M#um}YTKQD2Ta-z
z4(Bk}(~qJmziejk;FOthmz8BE(W!%AHaI!d(L}(jOm^$_l>E8HCDuxNqX)N3#|69Z
zA50)`miAlP7{|imiPCSpF?jGfhamx+Fy!{3UR2*_@GDi@4)Ezd$u+#6hxDl0A7Yfm
z7bVFs;umL&h#y*ss8ACM931-1fF0xx|8@8PepFfled34{CY^4cDSs|MlxXD^>*bT>
znroXW++EoL`4_eC*-kaxe@GU&u`$gJ<sPmxb)-E;CYQJH58x#SI_`22XSSc5f)(;W
z+R2xSL42Y;%*Zq;sn#XuN22;V1a<}4(^uizsnjoG5qW84jgC*pEj2Y`H#V$n*NHK5
zahU$`nfAnRyGhkv*WXVN4+AHTJbU&GDlWr-%`IpKR|lYo+E|hMyN-rX@x@n?7LB(R
z1VhC+wv6%W@rzb{RD&A2x`{~5gYXCbCrvOvfK@v<BI4Hoo3OApc*r{d5_N!e8$|_L
ztpgF!Q`cgX=3W?;Bm{raFne1~F5tJ+372;Hvbf#Kp~Hvcm?ZK$o~HE_x+FuW>WGn{
zMjc{E_W}E$#Hu^Ef&`)I4ISYS<UIVCPEVDynMFtcpdx68L&XNpuMK2nsbGM7HE}R$
zA;CX~2rn~hX=r=`e_8=^byZao9KvB7jUp?Y$bmY6c}tenKu>DyH6c`daxJzw7_Jc~
ztMB|8K+P@a=Sc1A(+5vgLVCIiGG;>+4n~g2jOfbYSMV<jDyspr(`rl`K_0xQykb7(
z1TEBR?&}4;H*<o1C;REbWRqV82NdeQ*>r&3Z7@5CuOhcH7|7c}!9mVr=)ctfKOBg#
z603D!uOFj+Bl{*O2ztp9Ow$GjVFdP3Q!VY|FnlMf?4NoXwZpH_hbs(YekW1m0}OMf
zzepRhn;x9^0=GeYwE#4}zB7K%58nzhB$`Z$B$7tv7Ze{k_z5YoOY8yy8i-i{hCmZ;
zBs#4CH65J^6yI}+fSnH*v<_r^1JM<^;*j|s+T2ok8O;S~>8`4yy{e~Vwv=|_<yn4E
zje+55H135yk-ouRYat$;bZPLf&%0V+Lv0YA!>NSzNHAi8F?g3~pE(V7koZ453yfae
z5$(Fx{u~=JtehIyOo7N>)?ERbO-E_RdO(@+gTVcd_kk14FYarA76C@^5<&}{Y4o2W
zKCwu=y4;OdmoMuZMJgqOWL2K$aOl**FFNigFl_1AebNWVxa^3qTTHN(@1zt{Vt($#
z<|qHiSQNsrF%%N2YJpaLD+8PCuNS%vV-Fbi-a`f91vIz4WHY25p=41_HK;%I>}&i<
z7&hidooG0eSge2EpPBC7;$xhi!A+mixen5~({KS={+-yl$}srDXwQ&={#9~Zce=(!
zEX&LNKARDqa^NfM2-8Y)%D<lh_Pp8pSSrD-P(HqS4Q(3*Z3<qv>&@0be(1LeS;Eu^
zCPodc02m;Nf~N>N=;FLJyIzHi(vv6P74H=4rK8&<1hANyGic7Ag<#&`@4R1NsprLc
z>ci@!CGkZ0m!niz!;{zw0%D@LWD@4<a0|FX_mdr*c#Eg?f*Hdw$**DBr&DCr4Sr9|
zP#h|l&sacdx7_UaqTHh#i|3P;1BRooG-;aPumueJeZaTGx455obhLoitG@1nluJYJ
zT(416R@z<)jy{w^_tSmTn{3+MmJ{{}Hgw^EOOBLX+1xFOCPh%&FwE--U)t605yji3
z4_Cy6k+zP_AMc?1nPW+#zZST*>b+E;yt_UuiN+9*#?VmHyVfu+xg7w3<zE1R!NdQB
z7zdxiS$|(r-ozvpHHJXv4HxLVAcBKKPX)rEXEbaw)G`MaJuTQ(%ix*Q)8U6NDt>lo
zA`8yi(S#6+oS&#|20Wt3=*qMy4vnqha9kB|FehaT=PM{u*=gWCv?rybk#7qe7*0|z
zA$o#es9F;B9KmKqtu+Q7ld*bvwvl;tpn&*Ydt1>$4h@W25G8>4L$WjS$%C5#yamhO
zn}+S-+=4&d-4PKC6iF-0K+HtMc-X%L{|ytt2KWDN1Tr;^V?kK~nkXz+H@0<^l&HoN
zm;xkZ02zg_(j3hXwP<J@8;Hg*0b2EojY0oS42t=;7&JBREcfo6nosH7WB9ap@9;A)
z^Wucdy3o@KVdiuIqnH*D%$;EU=!jQlmRS6D26Yp%SO9|#E;1zbBeIe$+Di>A6tvn@
z?c7}QoIERiT6$Bk9k6=i{W6rus{jhSyY>4EW-m8A!J$GzL2C}jehlzZo=e4*_kk#_
zB77ZpAETo)vlYW|79#%!>&`uG9E3oCg+=l_e(Y{bLE!|ENq}#hD*k+zDd(<v>4D?5
zQsHU!Iqx2sOF^6<jOgbTVNmlbzMEB0D$39Y)~DiAR#!shxWxIca-_0rc9Qa5T`;SM
z(Bs#ybc+<@I?9l`{@_hdZl%;Yh{ss7!jGOVug|hJy~E6VaJGTvPl#mr$KlTN3f;!}
z2fs$62!n+`Lv_?Xs;_y0VTwZN|Hcoltoqa^FQJczMyKnmKCXmLez%GL2Eg{De+R_A
z%;RRa_mjc2Q%5LqTu|rS9`0-as<QfU9?{ppy{-QHl2&KnaDkphqLr8mKttGi4<G1t
z3Mbl#Jvf54_Yj#wPKupcuPZSd+L-0knOU*tnp~9k!nFoqEezW#w+#c~P0}#4Ti=M0
ziudFrHV_Xn%$^h|1OGylgdoKJM_k_iICh@{Eo2A`Ykv~EPn8IQ5a@!Ht&bA+j;q57
zaV!2o|EE~^&nI=N#?zXNEiSkFy&#A1%Q{sk$#=G@uuxpXi2ZHIXn5`+y7B!@`qIbR
z_pU5-{{8gQFINsez`2e<?~_TO@uw4ajej=Ui{GsIlGHoT(Is<#UU8q-M#1S8AY;JO
zzwgzUI(fTXe0%4>V6_3Xl=+$bawWX5Mg1Q#f?wA9)_%V9!vMZ8tWl$DKVQA@{k?CV
z|0&E>T)r3V{4QivP)vb128RtE;VQ;4p)dL8lYS?2pvfOLFb6-sDo7||<Lhv~2kI9E
z1xFnmF6d|%iX>w0^8I`XWRRHrPW}&}SjHeUlV!uUdI7Kh0<R~6u^xPsYuld$3K2S@
z1gQidEg&B%a0pl#+z|M@e+5kw5!;5IGlD>(n!^z)2+9b{6Jya|A!SGhAcYOOv=br5
zdH|&=0Hha?*`DLS#_ccu1PNO9DK)bebV7!4t$+r{8W9gsX{ZjC_`ydA3tbPUpr8Ob
zaU??-te~5ly>?{=)RMn%1yrL(!!hdQY<QDJdxEN>k-Z-{^M}Pd9y$>lRl#?YUq;~>
z`p2M{)-*K4qw@mKoT)U92A(g!%m|_ew4BV$oF^z}`(K1S%2M0es80f{JjfdbT(VXe
zB_@e#$V-;s<U+(=buRep^UI>e-0)+Kk<&AoSot{rp40JvQftD3_!Bk!7l7s8w*vNe
ziTc;u%4I|%)h<6cdicl2l(U|}A&i)V{(R$QKT@w(Z5$@1a_j3sxARL)P2~oC?OaT-
zgLf^hzCD`2fiKQWMJrT8Z>k_QTZ5A`GmZ%(*UJ(7nZj)~bzd2hv19@cH#RdmI*>|)
zJutMPEXbW9-wy>U+1IZf#U+|tbDad(rL^29`mota2nB{7!d6c|*JJ~>8wG#C6->H0
zDfN_@*-dq7FDSX^FF`|={f&i%poTH%%DSPds-;beU5d+nVx4!>;5heDOBVvn(ljr+
zH`>XOn}Y;<a8r@;_@#jV0g(Hz=S8|4s&>%mSiY=uOlB9JCO$ojtSp2N_GThB7GM2m
zICHq(b_WGP;zubd%#}_#s5r-QplvQK&CY%<kI{V3jzCWC;4qd;6tHJiAUGv<b3esW
zA1wl&0<0)Hkdyz*HQ5uK3TJ|ChhYSjwruBO*8hu8_Mh(`y!=m`D(wwX_td5R`y9Mo
zL{stPv_BAp!CV|fW9t<Xv)flLN)4{){D^D2^c+o!U*TPo?PF2NA)$r2XV~`=z?7QG
zEn6`{0Iy)N<V9mYGxX;lg8t4{GAe*LA&`Fkw<vtz4$4g!PSREE{&8D>^qT)>u!u$}
zB!y#()ZusuD9XU!SL_sJ1iik{vxC|<;3TQZ$8?72I63$no%_AtBV#i&)!^K$c60_4
z2*@D5fP<S|M5Noe3`i`nA77)@+y^gCK1b*s3dI*EL^%nMWz-yuj(&uWhn}fS`1{Zq
zI1mjQ+SK4Mu)aWJV`DIEJ2(2Im8^!tNn--g=`{oA8op;Ggo(YZFl_z^<YaT`1KE?w
zJZD83pcxU40T>Y36k(u(e`QevCiHRqoKf4Z1sxF1@e<WFc%nSQpbIwUrm4eTdaA02
zO7H}~^Yo<}U*}A?xMUrFa$CpX5g+hyFTgyR3cXG@gHV@}Y%6q+>yPY&goL0*K4=k`
z7N5mkdWg;l6SB!5c;ciX)p*E@aW_Edz<8oh#F^jc<Lo0?!R`RAder(>TUVC@?YA&9
z_kKA+&+6J5IF?jFF)1-P3)jwFar9-@qbE<`ypn6EcZ-vY%LPsoNm7oM^!WK&1|2#F
zZ=HZ|>x2%lqhLsd1E+Wl3freLA!81)5>L?kgWwJp`late&~>u52IaVm@{D;xeSQ7e
zM4nd;FQ=+S9;A(wwLF1<8i)ewpWQsfd+8Ugi1#&rwL!~;={r8d7fx|PxtrhV9mpsa
z2i<w-EHlHDX3-U%5TB`b4#b@T{sl4chRT3ke5sQL7<>)iG?<Rax#&DG)E4tF*q(5K
zbUI;r-fln~bhqm;6>$EHLS(=&Iz#cg|FPk(Qz_~t8XHiH#zH?RoK0YY93j<`FX;FX
z<VFW3_5*q9G4bIkcTy)krHD)ra<EIRw5uZ(l0=h`ceyaUWqL{`?V;2eY2i1Z$Xmhb
zgh8J`g*6B5PH+}!cBs<<wG0Z^WhnqclUnLi3>T1v##NPbiXC<nj*DgM4R*`TSE$hm
z3A7sf@|&kEucsw5mr(8aWBm%~_*M||Zv>%!aPWvmWhrZFhCm0bWJ9#nVDG4I+$s6k
z=mKTXLQrG-j~{o^IxS+KM8tp=7LAf8Y6Y_kBfE#|>hL$zmhxcSI-ZvyMvcyPD23BH
zoN~>~%sj<&yRG41i}7D;K(EW9-efp6B&*|(a&_QxP#<5PFN5Kgw1+e>t0v$toreK*
zZuDKFGx!AQF9F&*kZ+l{c)r4E6|G+(o1RZn+mp)RYEXrSWAKGcM_iehgA|-?Hz>I_
z0um&Mt?l{Bpp_x!BSt2WQ&hC4H&!pM)CtWpYRaM)v<zmMgl<PwXs<<y5*kX}i}Z;J
zJR;qf3pZM(hq?|3ynq8B2H!$@ynXBmKWvtf`(#WBPP?q~p~oODEMK<>h1GfvOt7|I
z&^}cKrILSe@X-<IGF|&_1`EEb$!GOK<+%gKFzl)T;Q{k&iJ=zXL`2)*7?}WIA-=qz
zA(Z;~_uM(bQibO9Y@clp1IHRx4``u<Q&zcUTcJ@Y3E)KkkEs-=F}2CRs)~O3*fia&
z>p{-FHhGDl62q#6){{%pfodI}PK}e#)BL@Jj9D+kW-KS(!=e8(mRUdr=#=2l(6Jpj
zK{X!AupGNVLpZQv3Oez8)`|vYgQ{0iM<F!6<mG!{=Epo3Riv~|6Vsz}fKhKxbuM&E
z{)4Mmin#az!YgX;59;exhBjGnup?(MqOE9RiKHL4O#5v><`H-eC!3%X+EBY!^(xq(
zz*VFO=!I636jOX^NOll+pHvLg{7vm}&~3!W*I_|=!@;prZSnAWIBt>)*g9ah=Cu;x
z<hW%~oDUET^a*iBR&skMg@QvGPW3Q{yN%uf)h8Gi&Vr?h6W&&~HTwUw_vPVKuWj2)
ziY6o_v!XO8Lx{|kA!$HFWeP<knPqM(sSKsiAY+3NDN5#crY1v%Wfm)uF;isx&Rcsw
zd+%rZ-uL~!<NfP-j(r@DmSqjU-+f>Ab)DCFo|h=m94tv>pg49pD^D-1C^(xKHLKtF
zosfh;b|)16Fc{tLxIj5G)8$+c0)$QS8H!xQyTd=x{{@uWOu5Xj&QqA&Ufp$Vd4%jw
z8JhS*5%T~@7+*9!Cb9JOldOg0PMFM<N`I{y{x7H3A2rZ_{_L-?$UoFy|GAR?xg`IH
zrDEUxrAq(bUnM|k&}S#E8QohclFx{qmB=M(j!W)=9~@3H0N5v;H!AzpEqyJPEP{m4
zzz&K*5)h5i_CMiu7sAe5U~gIQ<+(HJ?EB@{g_6I#g&;L|NH#EzVrAXZZ9?f*_w}bj
z@@ob0E)l*!;vseX?FyuL8lglJ2|6W&mwO;SZ9quOEqMmN40z8GWF0=m+AQ+ld=nEB
z*J|M8t$|X{3ArZ@&?8Wr*W!pH-lf8jaI8X`;__Y=;D$eE1Q3BFgsV@>sHJb-Xyfc9
zR(a(07A;x?Bdz{UyO$m8)361lLY-foVkAc-CUcTS7|p4wsgabB;A0-M@PO)uG5J9d
zl}OqFd$=Z}*TlL4&H$SrH391d-qQB7O>kQw9~uS-3E-;qbKz3OUlUX(hVk8XY4Pxl
zXt(#qgR~Slf!xZ19%u#ZUeUkqf^^woU=pHvxS?nu`v!Zf`9A)%8x=sp*Z@U&c=ztz
z6>JCOV{xDpr#ybxYCoPnhU%Sf))+=;s<6~L3g4FsbRFQVG~nG2#!!Wxz7ep{4Mt9t
zs_t+EBIt+0%s7qcjd63KyE<ZN3v)Vs<Q$On-+;wOXS7etbQA2>h~WcCJt2pI?Avz+
zT3?M&;@tboFZ0OR3FQ-v7<^K3vO;AwHoLakwHFH99^_Py1Ox=CfiMyW-cIKzx~ioG
z?_jAGgn|Y3t)&}a!47wi<5Os7Z$tmvR7G8#8$1mO_^t$kJlM#g=;@YQa@``3OX@9o
zGQ_e$nj$X?-H<0Sy%q~3`3^iG#E8FQd|ceG@UoHe{_)O)3rrv}DDZ;NU*UvAUK-Ra
zy-u4`O_`NMn`w`A!*%@laWz0Dtnh5_ZOC{hw)FK)5NGXQEm*~YOB$<<_0A*b@;(#h
z<Uzf?r)U6hiv5|m1EC2Xn<2C}aX8n;!>vvR`l~cpH^(O@zlUS9B(7f|ypyUxzmX);
zGr>)p-j2hvF9-r)L3j{(1%eC39qlvok|b{^$7pFKd=e->g0Q@}K;0RVQ9|C{FN<#!
z2q9iw0#Z3*$z=Za#zy~R9K|TCELBugI7pcW&#Ysc4KphiGBG(vYf@TP@$=i=d+>k@
zbTH@djJQNtI$5*1DWFKreG}yTTriU?hDW)zp-cU7@_Miq3<RB<d$_Y)8f2ciuRY!s
zAA+482!Grc{>`WDv}}>ZNP8o@<|h7yij%z(@wm9;iZqohVq%garo!<#IZbX*A_pNB
zke{dXVBt8RNT|3>8pI~XFBo9F4GmmBN#wt_?;$_p*lJPwBg(p9#XQLz5k<uoHyB$r
zhw!_7qN7<VP2RFcNybI?y~iZn4TAt0nWKVp<~tBNh=~s``IUy>OB6$Gh)XG~?AlZN
zVcHOd!rV=*e^!#)2UEC<*REdG>a@{i?7|Hgdk$Vt<xN1=$2d|<2@`P4k>B&%Ir~c0
zRh)=J{kFgWrMxE_Te<KrHQbFrY{KYaS%8ERr!gEN#FmsbX-;2t;xVnl3Z@H#aBptL
z${Ve-L_fEf*R|;wrU9fD4?CXl%_BBg<+-Rh)KJ{xLk=Qe<BTwNe>C|rF>#aNwr$}Q
zCW8(5-Uc~tX~#I6#~5FKTUtRibPO7Vg;?fT5*AUzb6Xo2H*ykZjiB`ObS{(BSh}p3
zn1Y3i7KxK74=vk}i-XM1*-wk)v=+unBDa%#tx2yM_84x5UKs9VMo+e15e-C*Qhiin
zt_!Q_)ZbJ;?FEZIbXJKoH{l4sfH}4>Vk+*|C|@e5H2niK8cpGWcjS*IQqh0>sKdk~
zqXnKhkz5kL1GrD7v`}HG;RgjQHHKQB?~@O5^8@(jhfH6AD<draaV@{%-4#2fu!Tpv
zlTx#6U^rO~@1F!xNrALNA`v`74?BPTsEU|(7a8|D4H8#GUyDh)eKQ^XTBQz)NVCTS
z@edq*PJF0=vkkFkRuHA!)*sD>GtldL^4cb|qbB{+*n<`v{>17I94fBK5Ka=?AVx<#
zYHCdMXsUPC)PcD<UVvM2OX?c5g+lNV#npL?Rek--yWlHP!jnrVSSVi5*BPDZg-1|R
z>#7wi5<u<hFM)^Xv=>@KHL$Pug4HXbU<dPTk3&V5wadbYSODR#CA*x7wH1Q?%T6=>
z0e%QNaJv#w@$xx_kg);#AnJ9cx<Ggb7eC@Z9=ewFX5fWJh4Qf)ydvVI7HY^Z1P`Me
zGNu4SDE5=!<I&={i(D$0ee{?3A9IsuFb(Q%(J}FnIS4Fg1j9Mb_IiMmM~p79=aNQi
z!!5t3Wx?n~uhnNZ7wYu>mP5_7#JlwoNr{PQc6IgNa?ahqE11auy@+UW+XhKu4GSB6
zDo)lBm?a(A<-b%L&U(2@lb^32=!WKi)Y(KS8FHnoYz637a{ZaRfJ&zl*FX&&blY0I
zC;J#`#tCZ~ZFAs8*NL5Fvl}f)s$bfobQwPUS8SeKocTAdUysM?vF@&ZM6ABy29!t&
zaHIkGh>RJ>-?0<F1Y$`C*<TU?iEzL08+a!KgS_ti?b8lo#!0Nufus?0V7@kQ_}nxf
zdjSu06#0DdtxWdK&eW5*Yv>+^k@}msvbTB;2R$;USSdP>UVPSkw)5_7RMmuuu0bFt
zvQ+nk7KiYsVwqaO1&`S15Nln;WrI*0rmHpr*P{8%*nEVU3AxzBzzN6Kh|lbp4;Zj%
zBxZ~$;!vvs-^+Po<m48^Of5iiwc!1Cd*5x?>G}ILw})R*ccnUijn7nx4*?h8b(IPO
zxgmz>UdUZju_8#>NjAsn?7rtm#=Jo6A}2NE&bq4%&zvI$9mE%KG*694g~19T+96nO
z5i?z!vwE~V(uu$>9VOFp5Wcf|QnE)P@~4V&XQzG)*QA;#5px=3U@QtFSCMH||K@Q@
zEY}Gb36jUxH|wKy5n}P%bkkS!$UEQ+a!2Yl+Nm^qICKQRw}KlOkuxEgAnOAykoX;)
zDMxVS$@T1h#mKLVJ_O<20^1->jR(Q>F=wl5_{)gRyaHgM+p^AyBAK7&5$g%N+|ls0
zWL^&07WnX_7$zV@14xT&UotKLVM5BZz*n}17(u|;v>JXA`yyqu^Cz1J<okOF0Af>g
zc_ksd8GHXY4ne&REU#{?Re&e1dv96^@eP57{f^=`Qz>?xbFD}^(eTg%YV$0YWC*x}
zUt(^5)B+|1Ob*ivz}Rc(cA!+7W2y)l)-zfaJyYGSlk<G_64RByF*0-UAc+V>tq&28
zq5rW=pN_=^t|-#G!DT)KX3S%e1?T{DaV0|q4yzVWw24h#*;zvDMkLpUNM9|8Yo%^{
z=%G678{}v<Q&l&RxDIle&>%0d{YV`=P8?_8a+HjqVhxQsS=QvNa6W7$G?G>D4c&9%
zkxcB5Vbe1tq~Kzk>k9wJeG{_2`NtNQ9LD~EQ_}&W(uPMgBw|!d@0ep3bwbNg4Ffzv
zuO^))VxB}U-jbv_5Q0+S7N{J396OD0%R~4S$jBoioPmGa_c}|gKBDwqtATK+*m;)n
z&3QzYVnrIzh{*D%XZX0eJ<4|S5svsvhx=>4Vs%q-x?^xcZ;dS7C(jC)Dh}>^EK<M^
zw9ka-Nr=rEy75G(fP7EJQt9IBHTX|s0*4CB9y@_I@5iG2I=PBbxGYSHm@nYeNqYBf
zvAfZ5d+V9kXQGl>_vWUE1MdbHHE5vl=*g3Bku@QR&)s}sXv0^c<avOpQ;xx5@N<Qa
zxCRh#9lw4lo?hTF#Bv(1wj@&6W+_P6-OF#!&n<JG;y%q$c3o=TeX7)|nTY<)|Dsif
z0QAz%Pzf*{@e)6DV=*O2EtTH4SWr27-5(60nrv4)Vp#n<LkRfVZingCGsqj@`tb-L
zC~nug_C&e@9(P`fB;Sj);pN_s>&2!LQ`%(oYdIvXu99Iw_>_l7vixa3zw?gNByPtT
zIyOXDLTGsUbZe3(sk%mC!q(Wmb>1N)I7EU5v%n)pi$6J!WPRF^YQr;6(sU<yR5&ef
zT)n#b&W95xx0HnN>#L}ozb`d^(`zwZjJN=p8Lp<L6-Qmb=x#(4XXFx#O-|kmzCgH7
zOw3tS0+88H7Itx)E^s$G6GN3kD@E)Q%Z*P3trEY$Eaew^pYkA>rA!6F{0%sFr?)%|
z17NJtIAV{vO9YrB6EqNuz$Md1M;6iTtx)oJb&$b`f!*AIm@G(k$|r9U6D8OiHy?>M
z!4^PVssA)hH_q(d9Ma{ScM<lG(XipjAVa;NdhE>kvfw)$)^M~%tR+@N<kf`s-LBR>
z?p_3HRkx=%P;BxbWE|Y-N|e?esgta^Wn;j_BnAIy@s|xNodEm{4(;fBiSo!{Lu2Ep
zl+5<Uq=O6FI(XAw{X9`IUL@)ri|R0ujLn*Coi*+mR1f0n$a5y<!Q|q;kqRBU9zCT(
zQa+KFpO-6)fAZ^(#yRBb1S6l57~MriZs_Ug2xt(l3c~n(n7?Q0rzgF`xQ=T050My&
z0UJ(sK0qAdTXzp-sUfy`6+(LRK+;4m(rit1by09vV%Axz&lHWHl(T%uu}NzQ73Wah
zEQT9Z!p9etU?MafJ=puBwTv+*Mr7TMo`B_GCG?Z88=l423A(OkXtTF>NfbyoXhA(B
z2D{(uyus)ubBc)m&=_^-bo;9OpT-M}%sdq&)kB&G;cC3Lpcu(3K<fsF@F*F@M!Xae
zxspowJob`w5>fsk48(U5cUI!yNHPo5Y*C~;cKq!P*^pec7)X9X{L`1dt^m(Z64e;d
z_mHB9l&Fv5;&#GjuL6v91@Grb<A7~ovp1{*QH6MoqFM*cAQikVEF!`Ng%!y^K@KG2
zkTPr=Ye#-ilvCrW3)X|caw0Trl*YJ+v^KvFJEK6fLY6{s35Dk{yH<r+=k>KGk@}Mf
zPtE1cXI?WQh0z9y23?H01S&5y5(F(k=8YEJVMv?cOB|B!*ts9(k9&b$gN77Ae35Z^
z$OyZv&!L*|rEr4jGxcmQ@(C@ZlDc4kLk^J=J@t7#Nu<aKI2<L$kmhM)TY?~PrmXxB
zl>qK1JKlyNB}y93Aa;=?Nkm)<HSCm-yfHXGM!XA~+*HMv;r1JreC<=;T%Hj#5@Pa<
z+{vn92C0dQj|p|&J&cjT86SGj@wBbe7jfdFiewdKGBG0}P&#ofBORo{_M#IlZ_h4a
zVyeO2Bw#@<7A$e@%bA!^7K^8zWyvjb1wwveIe0+XhmkE1<1lr4*q<eYi<*dN0D~cx
z9baieK!%Vhh3a-ry|N}n68V-OZte~xP(9)Ru?Iz4uL>pDIt|eF58o}}#9u<~dLMNO
z$)U+zCblq8#}L~Uk|9FkgS0cq(fj$mdn#lI4^W<wi%L^F;3Am6Rs*nxRM{9jsy`yZ
zBH7Dshb`4-a68>wg?$|>d%B!o!3I(Cdhe;Py{#!1F1&m$9fq)T0L?N7r5PIkYywb-
z4NyK4?;VovRF&jFy*-x=q0k?baj_+-6X+$@xN&3?k{J=?UZW0xu`>aGh+DOhTiXRb
zr57kH;fa$qYlwdmlQdHMFmY~GLCTvA*ZtJrdcAl*V#!5zA5>Q!JY>U?n1wUT2`L~M
zXc&4AnR_=$*NBLi|8yuhq#qlk?IgQR=iotNHk)>pQn8!tajmVNR^m#hXU)%lf~gou
znh8~?A@}voz7_Uo<+tP5cBiLDI91YU3z9gSEWaO_KvlQ_9$?XkF);I3!7}-SaO%;j
z)-i0j+@Xe<%f{3*c}<Xs)W%SMu$%~`|4@2bX*Q{14LsF}IuLiJ3Wh!8n<r_`JHy~^
zR7IW$E<p4cHGih2dpRYhzOrYE5yuzc(SIdB_$CN1pWiq;hz^A0;*Ch$axWG8`XP$8
zJ)z|IhKuU3;uj(&A?=h5#b}fpBn~pQ@GG0fQD0Hv+EEY!_{j!`W9U5uD+8QlS-6i|
zQWDz$AePBS!yQ-~X2lp%(GW9DH($t|*OOPULfO&b51z!Sp@>Y}Gx_jXCKi^6&(;;-
z_K`_Vq>`)ebK&$~qk%AYhd8BSu@S*KX&`!dKtPBaLe7!u$p#pIAH<MB652^B+}}r9
zvoNxR#@uwyS{tixod&@?5#ZNp5Izk~Dq=l{`@Kn2^Z-fsXTDujBF@e@--qga@Ye*F
zCBYqshg<UV1qEj0u3sgQUF5Hk-T0{eCtRp1GG-*jD4Q5b5ertRi{prA0Jh*EwDpM|
z3JBU)q@UG3Go3zjU38eS;ZQk;@C-14e0K6>i9I5Ym?Ol27Y0G2h<)ABvr`X|TsRf2
z)Q8v6Xk>IW+?<F~5D}yYEl2__kTw;;+ICkFjA^p5&xLsSkx+Q}@GV59+utqdtdI)j
ziYx?)@XH@D+C|hWfs9!~N;~I%iPDqU01?2Abl5d?KtCnM!#D-H@LpW_)N|pR%o4Fy
z_zu=%96YFSE%%2S?wiH&b9HSH&vw+a+M)M|C=3iz#m-KUH89GP_<wc53_M}ji$}@Z
z5TFS$oSL)!Ke3<Q_lrvEb(Pd5+u;5Dl4wy{b|b3xvt9=D6SCwK&hx&RJWM+P43)n0
zwH6uphe9`mhLMc-G<p)t(FC6B)B$qPb8&f&w5Y9y%2=TS+62-rA<zgutP?3~`mVMI
zkTANEYc&YN0UzS-<a)SH0zq~3NdV8R3ka|lY`3A89ECJ~PiW-m`(r$q8Pno3>jkKI
zH)?4jf+Fh@@rAfH69-OGA|uH3MMghPO#FzwB<Vh4rGV*!vhE`ZfQ$%ajW7`X%m*hB
zpp=X`^DSIBUH9(+t0*=Vh@Cq~?2PC=wQ=)%$D;-}Z9LerXu)zLXwV8nu!#aj8wXo=
zQ$EDLBt1$YBLz?wkf}DU`%k_JT}Oz$5UVw<YJ(TqF+aRbvk!_qNkVXdathDof)6gS
zohRPHh#VOVEPTlb{~MtM19Uega|rjFRN<f-A3#8+ih2(&0jh5eXHeoT+PtbZ@+*e9
z*21*%F)(#f!~+!A#}=tT=?pSfPhFzMkP`=?+@Xk1tB5+xd66@rOuXcG>sa|ZAVhj8
z=E;nP>EBKcM@21FpLV+ewG2V0FJE44J~cjAjKY6QN`k%h#3Ua+>VmLw7I&l9KOFMo
zsZvXGMcj}&TkUtrri#Q3t|Pk|w=D^YI&p6uHI|_-C?{vo4!BXoywPhdUqPA2`SXtu
z6$u;erbVRlt3~tJe99NPOmQz{o>GuzMyef)t-nb`<U!kfnYn!Zy#T|)BS`O+Q*Gq2
zaD)mW)qv!J_zBnC2UH#SfEtMNiDq2Cbm6u?l>vxG#9Jci-PN^3r2#$Jp$NLw6m-V3
z{udZ|_>TXZ-Li!`C~5yf!xG)Fp#tzl4cbvi?7Il6vjtXeo`it^i`H*McH%WB=h_dc
z$vcwPg?JIyOj655dwsu!6bEFKtaowNbOkR!tX;j3m<#+f9y^@miAQ><6NZQYSE;$I
zLs!sw{s>I|3MNrF6EJ3s1_*+Tbl{tO52t1VMbBW6Za3!95KabRHjS^v85*O8`&LC}
zJ(5&~-~^ahGKv9C!fOkMBO)-!5dVb|w*uT$1FgrisDRrHr6`??@~~l{TtT6Ylh9E~
zWb<a-y?X=a_(8Qmb=PV{7`fa6z(r8qxs+uR2RfPA7z$6;dkCCTc7t>R>(^g&tf54%
z9~Zz{Q(i_K5J9Ua5q@aMfe0Frd-F{e78RX@?`G&dV(d>+H5@F-aQX9GJa=_bU6HA%
zXip@<U6~|}L@z}KT_DU|TRUAoI~78@hQ#{?OZNyEa{z^l!svZR!GQP&HYrT(YN3|O
zHDafX%m0_?CRly{Pi<MTcVBNTUk<e{RsUgC_vKHjv4=1koAOEjCntQ1vqD+9t3vqN
z?B=YjPf~L{!JF5@M$O$)2+=+`W{$Wm7EieJ^HGz7sL1(~i8#m4eqUM^{_C&uFGxx%
zxZ&qRPa4-<+;ur5;?y>({zbnf3;h-n?((eNz>Ht-=6BiWOt^PmU2tJp0e^P=`OIb4
zH;CL`!TCUvL&`<YqJUwyOx2D@r)zGUy+0FoUHtxD>iJj!D_t3@jy|V5Ma<6^vFogH
zQn#+G<{wcNd@Z|g+Z*;+v&)6P=RV~5@VLAdy7Rdy+j4As=l1CGz-Qab6_-%Ys`XEQ
zQ5^GXraW4_;M~jLTcFPS&IyUul8?K3PB%Iy^4U!kE6Vfe+*d&Q5ex$fH5(fnU}ODH
z1_qj5N{9fEl^MtxZU?R%gcU97w~3BY`U}M?00zSb8P<ADKC{zl;P@a}*@aZpwot)m
zRNn`EA$!#+?2q97{{HbUb*0<d85GLiZFCdM(AWs7t*u2PXg|hLZGn$EE4TD%9{U5k
zcI}!%T@V})fckV>%Qt^%3++u53g?;SZEGebC(Y5WeVuye&K(*~tC;|g)0u7lZsi^=
zY$j9ft7stE?UP3Vf1NN4Iay;$OtHKNVScNmI5QYh0^!$kpAmJR6-qze958!(;~LvU
zk(v1AA~O{g^jX)#26Ah{^XI+I*;L0(*DV)N>X*f^R9087z!Ayi4YE(am9;gOeK|Gn
zdq!Q(Sw|HW!!xfIkUmkNsLzjgGH8XVH&H7qS$TPRSy@2~YeG|Lu#i2P9#b2l4&IU%
zrj+_iRa`Vkx6&T@{=FE5gE>lr%I0QXL<?^F1Dcwe5UZ~f5b*x~EV>A@!IxP^+N-g5
z%%h0CncuZ*?_F6bH1zlZO$Wir8p)d>D=PE=AXSBMBoGj2cZ{AiYyx&MSo<PR3U~`A
zS{3?|l9Cq0WAXZ{<-ywNFJ7p7cuWS^z}SiQExNNjnxBlzTc5p|LP_-H1R3}Dr2taV
zaw5{WN_N4&Up4>yspp<LJFC8r;uUY?H3Zr9Cl}!Op076g`n4<R3O?Vt?M$a^+tq!!
zt1F=HMe0aaYU+rrK%7=-iU{TAsS`2ho+cZ6iOI9nzesiBrF268dE<CzS$YrUs{TAl
zt?=;hzx`NN7EoM#jmpKXNu`U4g8KCE;tK(0;%2qU0sj6Usb-Z7J1N~EB0S3V*an=G
z_dnPApFeeR$&VIFHRqj$=h7VWq$$p&OC`lz%ElOCm={$P)NOC&_LW_$OmE)lIdJ<q
zE<Kokr_-N5FGfISo}7Wt>?&JZ+sfKn_RdJh{gLk7k2rM~=@mw?7z^Q#q#J&;u(EoL
zO%&;LFfcHf0o}3(Twj8O(G1IzY#Y`wPjq%ez+{l*18*%8emOL+0S^<knw={-8ip&W
ztHfp8&XiwFOA#TThw1Ksf_ZUiPq{p8m1bBYTDs)J06Q~77)~=Qx?G16n{}WSn=#q&
zrNeUJ^P9}e8Hp;5c5s7Aq~}l@b=vHa_4M>K1}wgBqkj1Cp|Z7AsXG_`@M-9lxpm65
zZa>_yV>GSPPrss=nL{bxz~OOrwj9n<K8yosN@x{jzoNaFT~NRbrA=j3mBgi*niT91
z_d~}MZg&C&VVxY~_x5T?53o)M(O8E!rhe2p!Xne)ObV@oX!zTVgC~(4|LD=lYLD#f
zY?`LY#7EZI!`M2NgATI;2tmb|jVpY_#&~pO#BJ_nt6P<x4tS}YWZ9j@q0KZZ{%CU>
zrPwb0!}*|~`#rk%k;toJp46$<JOy;>_?w%X)3BRf_j};DHpj;#)B+2wb-LGL75JZ>
zP?eOtk;i_8IVkbqJV`Nc@3iVjWX{DX=*=OC5C;c56A-!?>i9H}7tjN!`sPJm5?9yC
z%Bu9qk%Z+Jb`}QnvNAH?$m2U1Sk*5whR?GpUTfJ+o4p<$9@uFinJ%;->0mc6R4IGS
zRIk$S>g)FpcVC*3eML;|O8|D8@1Dy36cMt`!s3LrjYw49)UmtyqSLy!D>1i%huhNE
z{o&j%&b{Smkg58)5NLtRn5x-~E_K5rQ`bL8g0imTd3OAXMHR}&>tyJa?*i<3`^28X
zR~Ev^uw%5P6lZ8CHoY$9^z~!z+cQ!b(yentFnEYzQ;B^GC7!=?IxLlEDvy?H6E*jd
zxE>2v#yV|GJc%Pud0kke*Y1e8+qj1mU7eTsXd4p`&;25(Hmpl4YekJD9oJ2si~YM_
z(DRliRc0;s`Yf2Wq9Z_rr=XDARy~WiOF8Q6o*OpU(3hkef}P|l=n}MX$p?eetD2ad
zDArdgOnY+#`y}!AzFcIN9V~lL5bH59v4`8z_zii+jeX~pvmCZ9;BznkWV%s>g*n_M
z#?__4lwNt?R(hs;!j=a=S&c2d(8()z-jZUCjy`!PCdiNS`NrRVR&ELGP)-n^3Uqv!
z7>!4{`))Sul3oz*$fEq1qk&pU)jtxvEq#n;_0?sml0>)<)Aa+s3?GAB_V&e=eEj*W
zcjvUs$2mLq*Dx5#_@-^_ym>^ue%@K}a4%jw+^BBthpQL!z~ZUexWS~Xny~d!j~%D@
zA2H!#X-RB7D(IQ+#jCz4F<0TNXh5G}29ZWttDUepAM!kvwp+E?)nv$c8F>|4j?L3I
zeHXuV3OW<3CA+Y@;M_IFoxG0|1gGC~%t+}X^}Q)GAx^i_<z7a?!@JvFItQAzIW(gP
zzZB2UrgiM@UDm1xyx!B){IYKYU)V*cyubg!p$#fI6Ezo2d<Kh_&YnK~+aJ-)_nFR{
z>ei1piB8DB<El`*@|oRT-h(p^d*0_cyErM-I**UdS~~CjQBZc%OH<RSxpAX>XCyu)
z55LVkHq))Mr&~r1^qra-S-r}?7nv?qYklTbX>V6@CP1-v?9F=e<}0TR4oa|?mRig-
zdya<Jqzx2Z^Vt49H&gVG-C-m0B=xiOY)SJ<&ay}ib2Nm=RKGXx;%BLjh@I`%PZinH
zXmxVqh0aSHEG~+26IoYddrBr0e^@YZCAi-{OSyPE+4#+thk<z_7F`o*c?T4+4Qi_1
z)0gh)3@x*g_x|jDt(e9g@2FvUxOIIrkbv%!U+66Ee+%EaLUdzs^1&0TyCe(A{|$zk
zt!UBN(1Z56R7JtPuiadYw~NZkuGz#0N)A(m4zz8#5YNzV?iD>HPS3RW@?d0lU*J4S
zBF-iJ<MjN<WTst|GZ5Q|Z9Fx9{%X<rR;D_23l}OIgD-sVOrkX9?6pRU7f|$KKu=Fa
z=!b*{PxZwH`BKZeF87JcF_`U<cW&?9pfY;*jgd`>KlcDdNdGkb2nL4mA`^1)>^>NM
z8O==j9*5N?#dSLI%Zvsubjq2zj1=AA2)rJxVO1oqQCziUt#GlN@0+kj%CxJ`nCl?$
zQJI$UV(DU%E@giDWl5aD=3<O?k8nPzHCf!C6VFigIpy4P7W16=sG5fLYM*zVrgu^)
zv@ssOZII`AHrXIeAj)&-&3Fl#;%<$NE$k%EdM3hT-Dq&sx>{K{m}`T*Zn0dq_7-n5
z6~)inl<gYQH)49wQ=~rjIpA)kp}m|5K!c_<JFG+_2}*6p4qI*;%^)tTB5nG0e8q0+
zxrs#w0!uG8F?0dpT2HDqH}r+u(U!4@gkd`?Yh_&>hsk6Eb!!T<wc++HQdwC7`;~H?
zoJ@D^`dZ%Cr`LKpo;~{l$3TL9#>6Go4AB>K699z*T7TSa<@BD$xIdt#asmQoz%w}Y
z7HqdqewLA)C6Mw`sQs2D^9*1=70=NmR(%tr8cPRids-X5klycK$OQ9P#E-vy^#Ud_
zB}D)#qK>Ivl~9*E>M`0KFg?jKdux?RVCFzxqSE#3+F&N>7VT8)K{mmmHTV$V7+i*K
zkSRZEo9&Ta{=QYRT4&WKr>ZTJR$Au78`re|yx=38mXWML3Xx<&(JKRMGl}Gd-KCX^
zN~87GMVtc(mKonKvHrM<^#hxU67aI>txlf21Y?TgebUV<@WIq;kMHp6>J4-&3ha79
z3mL2Lo-fw2gDW?lGxgEoRbtw6OJFQ-EVm9_P8E^@yy4WJ?`&bUxNCMQ+K+pY<@<ol
zzHL>z*b3YjJGi@f6{=@WC%8JDZ@V<qu}xWBWolSy)^Q3pNI|QN-HwenK$5c+pK*8W
zrx%rLx3=s!8(b^ba8>oCL-u(ZJoV2sp&dAg-kUfBBO{G}rj0)?HB}S_@0-f-`c-nb
z_mtZD&Wxd)A8*W`ZoH}J<vvwpGNmdHDKr4C0~SW?%J||jp6==g7rkVvNOiAC5{PqJ
zP9*?92}V%!LnL8QtVgFCHOTz-p{r!r)kdibh<G%n24-jMJkin%e39E{TUk-DjDMoy
z;!}|&G3|k_T6wX~Wl8!G;kN1NH-3FKT7IvzW9bO?BUu^tFykyHA63o7#6%wCOnV&O
z;iJR<P!W+)tt8n)&Szh?y|eJq!>A))Ur;i3DJAR;a?SB`zvn7R3wuXVK&uSJQO&m=
zw*4||_HXaF@zmSJ)O0o4B`cvj@<E}r1+2?usF%3y%fa6G;f4+(4HT3hG{d*G;C|_)
zdna)2nvR*65MW?r`R=zoZv6+h;%xJM>?$grH15^_5GP8|hip$wN=e-Pfs!E-K_@%6
z&h60+EW|vgkL7?nq>datYK|dWpcSPXq?xILh&dQAW>@!Q4T^;AEe^5UcpvkxxPD?Y
z&1n1e1@J0NL#J-M-!-3`$Z2R`t2}o4N|D-#B^@l+AAkm@{C=6X^KKt(jM*?dHi;&c
zdiI03PfC8i5}8FjmG$!Fb`)#sxi4_n(Wv4Z@a&WmJ|tbeVmc=ASA*hg2cabd)F-7Q
zhQ%oe2$abWzhC6RLdjVB;?BZogn>n~u1>D5B|{zWXqmKN?$OD=tJH*Y9vpdrVOHgj
z_Ua8LpU&Q5Haqh=^}rqWYtFXL7Mp3!R@WCG<c+B}NmMT$K#dlNQWeD_cYE%D*TDGY
z3q0NK$-+2)c;%v-Ik!AaSHW?yA*blqZuxgunq{te65XR$M6$XPl&F7S0*<zjdiG;P
zo>QUEA$4_iryNhkF3j&XM=!1<;dWFM-_K7qF{vJPiCepyQE#)Fh0#%7&K@_{&7gJb
zp!!q8kh9*FTsgc$9Eo<!jWV`RO;07Zi@uVIX^5^$jhfYypYE-)@3Tn_a<l!~P^IK0
zoA+|^j(x+WhsTfdunFO2z!;5hW|$O;q@I3}KixM7Nf+;ECvck;<TfIiN6mVfeB8<W
z*~z~q!@4;x37iu<^e&_Yx_0bkmz0(+0Yb%(L&h`Gx%Mzw+bT$?H>3Ah=q0qI_`#-`
z<%aZ$)7w8FFBnnq7=IJ}Z2zs5a*EH*hnAS>Ao#aJxh7~l09Sa$?`a{t!<`TI-{k^r
zC<NWNVoU}$bo?NdX8P(EgJ&o9fi)`z+5<c5{rXM}-BH=K>-G6+7AcZHOKb}tOo@E+
z3y4*cWP|XJ7cjPiCUsmMu|XUY*)CIAGEQ|qiy!zQ^U--O(rTfG8GIH!ot*4}sYLO)
zxn6}CyF3zG&<qaS#1s?vmUH0kt<45mPA5Gl8(V|PwgIzm6$ovPAPbm>wk)Xj4?lQB
ztxS0~KaBOLBDY`p^TOWLNiC672ksa&wHw;%+rU991aX(k4)3uwWpT_<v1|*k&C{6J
z*fpnLc7&7G)25?hpLb&=b!EN~M1qsLB5EL4qP^(4K)b2U)WS8gc6c?o+sRCIl4=9U
zG!0Ah{SyJ-OUH8jdY3vk8sr{#9<lw=d&Z4#Yc-dNZ&?J6MX0_@l8i?a+t?lH>FLwJ
zskAzE>V4+c)dG7Su59o$U@ev_J%s#ZL(ctW2WK{$rPBh_i_4>%`0hQmcoRz4+80e=
zu{u1M;+iY``m*n_U`px+8n5m}Qbbp$;2jXv*X&pP@wttP{)<De+(_xM)!e^7tE@~%
zcuL>;)MlggCiwgZWWU}UuMF{)oSHxz_gJJ``z6jp^Ori#cT+{kmR+&<zT>?UA5bQ3
z@Y=%uc>lZjrj&DepuEvw$QL4=)5at&Bg2soW=1>od}UdiSI3Q<J<;06$TA-3cnU#U
zA-y=2<>lRT#=uEP93Biy^;~y4fBO4)eYrQgozG<BGm#}id-e4Td+S*bKDd#0Xw;!y
z(5tMzMe&QY_Q8StU*WNPT^RcVMWxQv7wP;<0ERvc;)m^kK)Fq3h9bx-sC?*q!`<gt
zmXsPio<Hll*Dmf$i_Znl-x4qMzc?`!x%*oWL=+|X*R5l!V)oV!2aIEvVG2m$Ir{sl
zW6jcd$Lj<6zx?)&?g;NV@zir#09gC4d{y|SC@RN3=aOO0tUV6o>4eS>>zcLrWC%ib
zh7fL~VPs+*ERs;CZ}rC6rquJ)bm!-|u`Zbw6_J)%8IFM%#p+3i4C_!F{#E;=i{x5=
zlx2e7MgyZg<V8-$7C~k%U->tx1C@WgjaS0h&$n?0vBDmAzXb4>IiPhC9g4tn!*M92
zBH|trOxyRleLBQ%khyzTuBd*J(90frV*OW$mLI_;>_M_9!KhD8R@T{x)Y;qi$Hmed
zDm|Qd>;zT~C|$#!U16W75p<QPuye1v(;j<~@3Ng&&(T4<+3$At5XTl{k3~2irXbt#
zo6s3}SCgpGKW~KG3^#D8`O)IE_w3rme;Ithz`DE4S63CwC5VpZ89u)KSwz_cybOYk
zf&A!Z=s8$KsOZIL67ll?C{=F<snw;ngH9Loq}+7}Y@3*EV@2gH1zqXfOEuQtU7>H}
ztk;BSlm^sp4!OiY`L|h{Z;D9`Y98lNtu35<7l6Nv4dSj5!0iZwo`Mh91Gyc^n&%d2
zv*qg+Yg_Eg$=`G=N3cBIKK@8Gn8`UwmrOR%Harz-;TLNo4=${@N{?48(rzx0(QAAZ
zA0A#KO2_38)8PFm@a^1#2f3v+-*w-iFm2b`m#-UK`Y#?-UZATv_spaH!9)J;dTi>4
z<r29|7nN93r6m9lT-v-P$&QrW0zY4kBF0ejl+PnMy^UMW2B<;zdoAv~*!=>#Y{vzx
z2Sn!`7K2-WmL+}-vp-5qnbm7)MN)`Hbf0B<!e(3oI~Pj#{#^c|$4sr`z#o*GAB{;B
zG<*vRp)gq_MJerCu+^ODOy7&PccFOTVGDUc;+T!^yIU*7?6rpBBqcXG!@23|two3F
z4a@HNUh*>p>d-sBbY*Gj`D8i~OPzwk9U?N9mX+?;)I9n9IIudlWUYj6S*;F23-<DS
z{)H2-FtLH|tWMdl)=W1Ycli4AB;K0d`17H;*L!lF-r<Cm&)bUS*|F0bS@mvf3gaD>
z)%=T})+A|aZas2j&#Un;bHnXpTBGx=N^iyY9S~Qf-}_|l4XWieWma4E9-#BRv`%?f
zSzYuz3Y(^{e`K1{0*Vk2g6;8X3n+Ux-&wfk;@p$W8U73hQmprUh<v{1^kH1Y>cHD_
zwqb?<8t`NM-RFz16AuMz|Efce80JcQJW&e3jqmNYbiG7C>bf_Y2Mbgj8aCA^uVRid
z>kVcU@~b-Z<~FC*+&^OY08kdy@v<l@;EG*NljHMs2Qs(P+*SUnri;4cLad*-B;odv
zD$VSV++2Hn_H)vN2Pa?kyRCNhcTpc3Rb9%=*TlO;g(_tvvzig55b78tKYrAxvqJ6k
z{2Qi!7|=ZrZI-|rJ9JoO$1}M&zgsH#xuCm7@b&Q3RVJHL>HYgTDe9)KlJU!8)-VQd
z3yzQ|o}AW-2Hl4TgjE8}8-T@rtTi9}<n};`vViB$pIh13Tp_t0r>;hBLc{T57j=on
zxW<gQ4evyPM{?6<a#~q_NQsFp1JmF#rdzGPedmrjw9W*DgQMw8z$Bp%8svL>sp;v6
z20sr>ezXx)SX}y&iLfHoI^G~ot5QEexhJcb^S!4AJAwwYEUTlzha$ZQ3~OEwjGJBf
zP;vlV*iIO+J0~VaT&w5(umSh}+|D)ab~8d|#esg*6Ym4YC#dhZY17^bN`D7&`qX)W
zd~+ZY`DGf2-F0BRTZHfCeKc|0;tw#7i>D`Q9nV!ZTAz2U2g5}|2Qveu%l_9z>kEN>
zqD=I1mIt}wVzd5c_IQKMYAS`8!WZ}Op=GG><9&!XqshiDjkfW2=cBfLaEy~(R6%i0
z0M1z$I7AT7uUCs6NEi=ZD`r$0e>GnHP*JK|DJec)P}bP^3DowaXJFs4q%4+Xk=@6q
zr{hH>MA!xQE7m*4O&I)2r*-j9xm_Jj8ImQV@u}7{3|Rp4>{&1m33~ZP<htxJPR@+t
zrIO0LpmSSeNR}dKyj(~i>@mQ&5LJU27#(S6+I}?vlgqPhiAex@)<w`+(bB8-UqKZv
z-1GE}_`A{sID{zQwYIYh`TqSoD-|QBi`$gImfHp?fL#4?I$Z@#I05~WZqm;!Rd5d&
z|4jusG+aJXQ&N(J=lP~p7Bk;GV`L<FfD^Q)+~A6_mA(zwf;Z13>BF2)I&K6!%TwKS
zzJu5zi(WWdHaJrfP5@eIJ96Y>3caq>9og<E;4YbeM*gkeRy+vK%L{n*%I}Ws`K~Mh
zWqP(0t$+>ef`3a!Y*){>zPgY1Jg~{N8&}Lfv6}Hz_iEA(xKaLq85k*LLo|Mi2CSCi
z9odB>u70em)*d6z&|S3Ufek_Ud^d?zo4F@wwrE}Xx{$4<1{av3yt2Y9%Z0$WddgS6
zHUM^|=+fc&7s*R4NWYoFOG5YGo@B1qmgo4#f?Hf(qrojl@lC66O*u_^bR0{(I=d01
zC}QjrWi?y=p(}T~+9Pi;$gU9lBXa=M_8^k&bGSfJ{xWhm>1eGozu9j^CUjfMB&&~8
zy<!HV0db$1sWuOH4-a!N^9C_0UYeYG;4eA(ucW8k>gwfr$soJ4^`-1FhJW3lD<^po
z+2P2F)Kv;^Sw`JbeMC_XCvQD(<S#9{`22jILWKkJgDNC%w}+c$v0TewFj{BB77>dm
zy7c&#q3d|ezd@5lko%F5kwWYcTHF<p%V45z-Vf6Y*0GgWiZ-xcVP#DN1*Gfi*EcZ@
z4Nrg8eNihZ7Q1{ERsO!AP^f@nU@q+;K!N35=R=W>W!ozo8@b_Lz-4bTSFts*IA9t}
zn%iLeqSX9a(qq_vd?=Z+nmhaqj~SvCc{*6KAz-YiySln=Mi0MG0i4LM-`=;9ZXc4~
z^=7DbIdzv%3M|-SSfC`VCczEUBP7x0UPXGJn+=|<h6ex}u${{R0i*~5<<1P^TYG4h
zN+84I!Em(FOt@wX`K{xLxBBZAla`b$1chMvizbJ677+{-<>G(Rx%wZq=Kg8#_qW}8
z3@6Ho^5C0W?~i083HMy4#JKPMkVT2fa&lI6a#HDS)OT;D>bvu@YHzU6Uq#VfB2iJW
zFiD?(Y5A`2eZodChPC>%OwRRz5u>lWpfbH+diuTJe{^zDliU`ZQ(I?{1!t>@f;Iox
zrsIfP!g%K9&K;%<zJh>RDU`por~h(B|2KM)ZWmHg0_vX$KG?g_tFQZQmO$c%>f|LD
zyToo}gl;YC<o7V+vC%MNb(Iy1>uXe<$vb~TmcioRxdy*WdA$rt)TiU*TRjB6OreO{
zoYctP@3kBe!nn4cv$9TR=!m)bSJ|b+qsSOI`-&^9f9o-ATedY6Bbq3wS{oL==edg*
z^8c^N`fu*!pIW!NX@B?qDW=x73n=kD|Frp<;Cdn^cF*HKHD61EDNK4`jUtOH_`D3i
z%pf<FB82#+m-tK~F3tXPMse{+XUb~1d6K%1cCLvt2)rpJ92Xlql3|d3ZW-n2gC&f*
zX>+&L`tMS$|Hf^cd#dt<-`dK*x01Cr4nD(7zSRe?aAjo$un2Se$)OUkVa3y=D{slY
z)Y|dK0$tA!CpzE74KF47WHU$_oDeft1rw{Om|tY&1i84?jcrJ+`}2vLRtQ`!3OM{s
zfQxI7$rV-Zg~KYCFz35ic*XoL@)G}@C8}`w!yV}o;6(8m8B$<cy4H;iKmjI03zUd^
z{CG=4LqmI#$+R_t-aN|H1?yS){!JKX9V2G{Y2MQlfr;}etA$U*un?f#%GNdr&TWP0
zR-#Y2G7V^QG7Rrx_AIQp*flT^0>qd(|Mcn8!80>6uluthW}IVm5O@JpQ=w-n6v|eM
zMI{`Dkawjybi4(|$JR~|KM+V|F)AhYb-(a$jbDQ2vIEGdzh4_U`y+bxl0=z|V9!?T
zUo68Qp>(k@+NswwFd$tU_QvOR$6(~sW65$^{;yv3e0UCO#v(cml^wdTUL{{i5Q2DQ
zcN1WmLDAVEQG|IQj{yPWfSZT42lzQS0;=}QZGkYgK7Pg#zzGMECYDV{Pcfg}-!Yt=
zpf=-kSg|}&K8mfTbb0VDKvf&a`3CA#8kllj{r$lx=z=mk!NTU58AbwWhWV5~*&%7?
zY0E%(3IZ*$1a<-DI5SgTez|P`pBy;t2;sYNfR%AY6p2w9F#MZHp(0%xPTE8Nciadb
zy4O{F`BeYh=SA*?bM>{R+}oPUO6_Y6LuHPeqAgXWhGyRlojrtWkf2tJ#6Wo-$t#d8
zDeOQtZBmpC{)NnB?=u<>WLMPbodIJrgaGs9%a`#grRge-WU-?qvLe$!_iPDbP}$JH
z1#bXvqZd`%YoBpB0OtjE=GWZ4cawQ@8elH-<oNJ~6>Jueou&c(>FVnX0z;n1S#12h
zr6L>587kKvig+AL`-tH)Gj$Nqo&pXGGQzg@s8p{$8*vDr#$gD+hc9a6IZ+i|me6o2
zY&cm!APQ-7CIr>(XsDFO?CHLh*53u>P`p;#W`5VAr&w&6kW2~+v_oVV>@)Guhif#C
zA4KTE>VpT0JX(gM-fR{<*ufJq^S1K^$-V)5RMQ-$9YVrq$tR1R8F@(4OLf;{(9<ao
zbXK>_5e`WUJ|d`m;3{yKO~#N5AmQjS9?NnAQqP>-38wqtV2F?W44?`3r|;hBZZEl?
zKNLlPwT%xWmt&yo`+K#Ancki~q%=_(Vuq0rgN#1k7_S@gar6Yga?|HC=gm;#&bD`>
zX1A;)cocJXOMU(Jc4u&qK$_ukX$9ukK@Mfyi$zJ-Jmt=OQl6@hp1x)LdG7{u-op~L
z+P3q%?A<WuJY+nV(1He$gBL@rJ3(alG&Gjvd!s;@Tg&G?-ZnlIk<WwjAP6Z%2^OSh
zwbc2u75{)ON8!!fph#)QkaJjwo!IO0=$2gVGvYhhzFz5v#%ALj88Czfub31f?KeZZ
zPg3|dA`y{WYlH3%rpOn|T9>V6YiEldFZ=|a1ta+%zPsG`n*-iM31dXX&TojegvE{U
zmjr?Nmi(-YYbq50&nx0K;D@Nt{|QgM&N_yg!TZ~Rna>A+X)!WN!?Tyz16XQXf5Amf
z9r4H?^YEJ;<ewG4v{Gl_k6UKWG&29?%tQ}y#R>qIU5*c<;4|?6)&{Dc9iHF(_}^z{
zoyr}`9QlpNB<^WA@sq<8GCLkZLt1$WRVYDJyUiY0OLoW$mxlha+#*$Am9*&JW<#-i
zK~ynJ&&Xh!wL|w092K^Qkq&$C00F?-Nw5(VP&4$Yy3?`(S8Dy%6uuZ-I+<<jS$E#r
zkA#)<V$=;A0rgGw56qnQ>^a(cxfNg{>HLsJlS<*YQS3uSScT+d_C9^rPJrAvC*P->
zJ)i2{EAP4x>GV-!@cWB=1>)KelIe|vQSIo_qa$Numb#qE9UMK%zWz@YfXcq+KWPjr
zh>Kpc<ZGil;XyERR_a)9JsUuLVvVp5##QDglKsq|ou_{**-&zul@-FIGm)4<f!^c|
za&1W^K)xk;JZJQ=T~JF)%XnRhB73SwbI;ZUuGK+l%)$G7{q4Dq%(~v|A-Sz<4vp6y
zG%^)@EQXcy0;AKInV+7g7pgmKd3Pn%#K)`2scPThh<Rei0-dYji7WI9cyHS{dLwLd
zF1@8$D|{Cr*ceMLWoBkRVFVNB>cs@B`X4jK{%gq$h4Qc1i~rN~@V}da{O1Gw=L-Bk
eU4d`2bXWKSuLW?Y<$;e((NNv5lCsnEm;VEvwHcWJ

literal 0
HcmV?d00001

diff --git a/examples_skyline/fair_prep_results.py b/examples_skyline/fair_prep_results.py
new file mode 100644
index 0000000..e2f9565
--- /dev/null
+++ b/examples_skyline/fair_prep_results.py
@@ -0,0 +1,92 @@
+import os
+import sys
+import shutil
+import warnings
+warnings.simplefilter('ignore')
+
+from helper import extract_info
+
+sys.path.append(os.getcwd())
+
+from fp.traindata_samplers import CompleteData
+from fp.missingvalue_handlers import CompleteCaseAnalysis
+from fp.dataset_experiments import GermanCreditDatasetSexExperiment
+from fp.scalers import NamedStandardScaler, NamedMinMaxScaler
+from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          
+from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing
+from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing
+
+import numpy as np
+import matplotlib.pyplot as plt
+
+#creating list of parameters that we will alter to observe variations
+seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]
+learners = [NonTunedLogisticRegression(), LogisticRegression()]
+processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing())]
+
+def calculate_metrics(seed, learner, pre_processor, post_processor):
+    '''
+    Experiment function to run the experiments
+    '''
+    exp = GermanCreditDatasetSexExperiment(
+        fixed_random_seed=seed,
+        train_data_sampler=CompleteData(),
+        missing_value_handler=CompleteCaseAnalysis(),
+        numeric_attribute_scaler=NamedStandardScaler(),
+        learners=[learner],
+        pre_processors=[pre_processor],
+        post_processors=[post_processor])
+    exp.run()
+
+def run_exp(seeds, learners, processors):
+    '''
+    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.
+    '''
+    accuracy, disp_imp, fnr, fpr = [], [], [], []
+    for processor in processors:
+        for learner in learners:
+            learner_acc, learner_di, learner_fnr, learner_fpr = [], [], [], []
+            for seed in seeds:    
+                calculate_metrics(seed, learner, pre_processor=processor[0], post_processor=processor[1])
+                extract_info(learner_acc, learner_di, learner_fnr, learner_fpr)
+            accuracy.append(learner_acc)
+            disp_imp.append(learner_di)
+            fnr.append(learner_fnr)
+            fpr.append(learner_fpr)
+    
+    return accuracy, disp_imp, fnr, fpr
+
+accuracy, disp_imp, fnr, fpr  = run_exp(seeds, learners, processors)
+print(accuracy)
+print(disp_imp)
+print(fnr)
+print(fpr)
+
+def plotter(x, y, x_ticks, x_label, main_title):
+    '''
+    Function to plot various comparison plots.
+    '''
+    title_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']
+    label_list = [('NonTunedLogistic', 'TunedLogistic'), ('NonTunedDecisionTree', 'TunedDecisionTree')] 
+    fig, axs = plt.subplots(6, 2, figsize=((10,20)))
+    axs = axs.flatten()
+    for i in range(0, len(y), 2):
+        loc = i//2
+        axs[loc].scatter(x[i], y[i], c='b', marker='o')
+        axs[loc].scatter(x[i+1], y[i+1], c='r', marker='o')
+        axs[loc].set_xticks(x_ticks)
+        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))
+        axs[loc].set_title(title_list[i//4])
+        axs[loc].grid(True)
+        axs[loc].set_xlabel(x_label)
+        axs[loc].set_ylabel('Accuracy')
+        axs[loc].legend(label_list[int(i%4/2)])
+    fig.suptitle(main_title)
+    plt.subplots_adjust(wspace=0.3, hspace=0.43)
+    fig.savefig('examples/' + main_title + '.png')
+    plt.show()
+    
+plotter(x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di')
+plotter(x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr')
+plotter(x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr')
+	
diff --git a/examples_skyline/helper.py b/examples_skyline/helper.py
new file mode 100644
index 0000000..05af226
--- /dev/null
+++ b/examples_skyline/helper.py
@@ -0,0 +1,53 @@
+import os 
+import sys
+import shutil
+import pandas as pd
+import numpy as np
+
+path = 'logs'
+def extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal):
+    try:
+        for folder_name, sub_folders, file_names in os.walk(path):
+            for sub_folder in sub_folders:
+                file_list = os.listdir(os.path.join(path, sub_folder))
+                file_list.sort()
+                for file in file_list:
+                    file_path = os.path.join(path,sub_folder, file)
+                    df = pd.read_csv(str(file_path), header=None, names = ['split_type', 'label', 'parameter', 'value'])
+                    test_data = df['split_type'] == 'test'
+                    
+                    flags = file.split("__")
+                    pre_processor = flags[1]
+                    post_processor = flags[2]
+ 
+                    if(len(flags) == 8 and flags[-1] == "OPTIMAL.csv"):
+                        learner_optimal.append(True)
+                    else:
+                        learner_optimal.append(False)
+                        
+                    if(post_processor == "no_post_processing"):
+                        learner_label.append(pre_processor)
+                    else:
+                        learner_label.append(post_processor[:17])
+
+                    di = (df['parameter'] == 'disparate_impact') & test_data
+                    di_value = df.loc[di, :]
+
+                    learner_di.append(di_value.iloc[0, -1])
+                    
+                    acc = (df['parameter'] == 'accuracy') & test_data
+                    acc_value = df.loc[acc, :]
+                    learner_acc.append(acc_value.iloc[0, -1])
+
+                    fnr = (df['parameter'] == 'generalized_false_negative_rate') & test_data
+                    fnr_value = df.loc[fnr, :]
+                    learner_fnr.append(fnr_value.iloc[0, -1])
+
+                    fpr = (df['parameter'] == 'generalized_false_positive_rate') & test_data
+                    fpr_value = df.loc[fpr, :]
+                    learner_fpr.append(fpr_value.iloc[0, -1])
+
+                shutil.rmtree(path + '/' + sub_folder)
+    except:
+        pass
+
diff --git a/examples_skyline/missing_data_results.py b/examples_skyline/missing_data_results.py
new file mode 100644
index 0000000..29186ce
--- /dev/null
+++ b/examples_skyline/missing_data_results.py
@@ -0,0 +1,99 @@
+import os
+import sys
+import shutil
+import warnings
+warnings.simplefilter('ignore')
+
+from helper import extract_info
+
+sys.path.append(os.getcwd())
+
+from fp.traindata_samplers import CompleteData
+from fp.missingvalue_handlers import CompleteCaseAnalysis, ModeImputer, DataWigSimpleImputer
+from fp.dataset_experiments import AdultDatasetWhiteExperiment
+from fp.scalers import NamedStandardScaler, NamedMinMaxScaler
+from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          
+from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing
+from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing
+
+import numpy as np
+import itertools
+import matplotlib.pyplot as plt
+
+#creating list of parameters that we will alter to observe variations
+seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]
+learners = [NonTunedLogisticRegression(), LogisticRegression(), NonTunedDecisionTree(), DecisionTree()]
+processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()),
+              (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]
+impute_column_list = ['workclass', 'occupation', 'native-country']
+label_column = 'income-per-year'
+datawig_imputer = DataWigSimpleImputer(impute_column_list, label_column,out='out')
+missing_value_imputers = [CompleteCaseAnalysis(),  ModeImputer(impute_column_list), datawig_imputer]
+
+
+def calculate_metrics(seed, learner, missing_value_imputer,pre_processor,post_processor):
+    '''
+    Experiment function to run the experiments
+    '''
+    exp = AdultDatasetWhiteExperiment( 
+        fixed_random_seed=seed,
+        train_data_sampler=CompleteData(),
+        missing_value_handler=missing_value_imputer,
+        numeric_attribute_scaler=NamedStandardScaler(),
+        learners=[learner],
+        pre_processors=[pre_processor],
+        post_processors=[post_processor])
+    exp.run()
+
+def run_exp(seeds, learners, processors):
+    '''
+    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.
+    '''
+    accuracy, disp_imp, fnr, fpr = [], [], [], []
+    for learner in learners:
+        for processor in processors:
+            for imputer in missing_value_imputers:
+                imputer_acc, imputer_di, imputer_fnr, imputer_fpr = [], [], [], []
+                for seed in seeds:    
+                    calculate_metrics(seed, learner, imputer, pre_processor=processor[0], post_processor=processor[1])
+                    extract_info(imputer_acc, imputer_di, imputer_fnr, imputer_fpr)
+                accuracy.append(imputer_acc)
+                disp_imp.append(imputer_di)
+                fnr.append(imputer_fnr)
+                fpr.append(imputer_fpr)
+    return accuracy, disp_imp, fnr, fpr
+
+accuracy, disp_imp, fnr, fpr  = run_exp(seeds, learners, processors)
+
+def plotter(x, y, x_ticks, x_label, main_title):
+    '''
+    Function to plot various comparison plots.
+    '''
+    learner_list = ['NonTunedLogistic','TunedLogistic', 'NonTunedDecisionTree', 'TunedDecisionTree']
+    processor_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']
+    title_list = list(itertools.product(learner_list,processor_list))
+    label_list = [('CompleteCase',  'ModeImputer', 'datawig_simple')]
+    fig, axs = plt.subplots(len(learner_list), len(processor_list), figsize=((10,20)))
+    axs = axs.flatten()
+    for i in range(0, len(y), 3):
+        loc = i//3
+        axs[loc].scatter(x[i], y[i], c='b', marker='o')
+        axs[loc].scatter(x[i+1], y[i+1], c='r', marker='o')
+        axs[loc].scatter(x[i+2], y[i+2], c='g', marker='o')
+        axs[loc].set_xticks(x_ticks)
+        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))
+        axs[loc].set_title(title_list[i//3],fontsize=8)
+
+        axs[loc].grid(True)
+        axs[loc].set_xlabel(x_label)
+        axs[loc].set_ylabel('Accuracy')
+        axs[loc].legend(label_list[int(i%3/2)])
+    fig.suptitle(main_title)
+    plt.subplots_adjust(wspace=0.3, hspace=0.43)
+    fig.savefig('examples/' + main_title + '.png')
+    plt.show()
+
+plotter(x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='missing_data_accuracy_vs_di')
+plotter(x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='missing_data_accuracy_vs_fnr')
+plotter(x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='missing_data_accuracy_vs_fpr')
+
diff --git a/examples_skyline/results_play_skyline_formula.ipynb b/examples_skyline/results_play_skyline_formula.ipynb
new file mode 100644
index 0000000..e6ed2c6
--- /dev/null
+++ b/examples_skyline/results_play_skyline_formula.ipynb
@@ -0,0 +1,1778 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Import error: No module named 'tensorflow'\n"
+     ]
+    }
+   ],
+   "source": [
+    "#load relevant libraries\n",
+    "import os\n",
+    "import sys\n",
+    "import pathlib\n",
+    "import shutil\n",
+    "import warnings\n",
+    "warnings.simplefilter('ignore')\n",
+    "from helper import extract_info\n",
+    "\n",
+    "os.chdir('..')\n",
+    "#parent_dir = os.path.dirname(os.getcwd())\n",
+    "#sys.path.append(parent_dir)\n",
+    "\n",
+    "from fp.traindata_samplers import CompleteData\n",
+    "from fp.missingvalue_handlers import CompleteCaseAnalysis\n",
+    "from fp.dataset_experiments import GermanCreditDatasetSexExperiment\n",
+    "from fp.scalers import NamedStandardScaler, NamedMinMaxScaler\n",
+    "from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          \n",
+    "from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing\n",
+    "from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing\n",
+    "\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#parameters in this cell can be adjusted for experimentation\n",
+    "\n",
+    "seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]\n",
+    "#seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe]\n",
+    "learners = [NonTunedLogisticRegression(), LogisticRegression()] \n",
+    "#tuned and non tuned DecisionTree() can also be used.\n",
+    "\n",
+    "#pairs of pre and post processors\n",
+    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()), (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]\n",
+    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing())]\n",
+    "pre_processors = [NoPreProcessing(), DIRemover(1.0)]\n",
+    "post_processors = [NoPostProcessing(), RejectOptionPostProcessing()]\n",
+    "\n",
+    "skyline_formula = {'accuracy': 0.6, 'selection_rate': 0.2, 'false_discovery_rate': 0.2}\n",
+    "\n",
+    "# TODO: add more test cases for above parameters\n",
+    "#update these for the purpose of plotting as per your experiment\n",
+    "#title_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']\n",
+    "#add pair wise tuples for each pair of learners.\n",
+    "label_list = [('NonTunedLogistic', 'TunedLogistic')]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def calculate_metrics(seed, learner, pre_processor, post_processor, val_strategy):\n",
+    "    '''\n",
+    "    Experiment function to run the experiments\n",
+    "    '''\n",
+    "    exp = GermanCreditDatasetSexExperiment(\n",
+    "        fixed_random_seed=seed,\n",
+    "        train_data_sampler=CompleteData(),\n",
+    "        missing_value_handler=CompleteCaseAnalysis(),\n",
+    "        numeric_attribute_scaler=NamedStandardScaler(),\n",
+    "        learners=learner,\n",
+    "        pre_processors=pre_processor,\n",
+    "        post_processors=post_processor,\n",
+    "        optimal_validation_strategy=val_strategy)\n",
+    "    exp.run()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  50 tasks      | elapsed:    4.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    4.2s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.0s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.0s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n"
+     ]
+    }
+   ],
+   "source": [
+    "def run_exp(seeds, learners, pre_processors, post_processors, optimal_validation_strategies):\n",
+    "    '''\n",
+    "    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.\n",
+    "    '''\n",
+    "    total_experiments = len(pre_processors) * len(post_processors) * len(learners)\n",
+    "    #pd.set_option(\"display.max_rows\", None, \"display.max_columns\", None)\n",
+    "    accuracy, disp_imp, fnr, fpr, label, optimal = [], [], [], [], [], []\n",
+    "    for j in range(total_experiments):\n",
+    "        accuracy.append(list())\n",
+    "        disp_imp.append(list())\n",
+    "        fnr.append(list())\n",
+    "        fpr.append(list())\n",
+    "        label.append(list())\n",
+    "        optimal.append(list())\n",
+    "    for seed in seeds:  \n",
+    "        learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal = [], [], [], [], [], []\n",
+    "        calculate_metrics(seed, learners, pre_processors, post_processors, optimal_validation_strategies)\n",
+    "        extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal)\n",
+    "        for i in range(len(learner_acc)):\n",
+    "            accuracy[i].append(learner_acc[i])\n",
+    "            disp_imp[i].append(learner_di[i])\n",
+    "            fnr[i].append(learner_fnr[i])\n",
+    "            fpr[i].append(learner_fpr[i])\n",
+    "            #label[i].append(learner_label[i])\n",
+    "            optimal[i].append(learner_optimal[i])\n",
+    "    label = learner_label\n",
+    "    \n",
+    "    return accuracy, disp_imp, fnr, fpr, label, optimal\n",
+    "    #return learner_acc, learner_di, learner_fnr, learner_fpr, label, optimal\n",
+    "\n",
+    "s_accuracy, s_disp_imp, s_fnr, s_fpr, s_label, s_optimal  = run_exp(seeds, learners, pre_processors, post_processors, skyline_formula)\n",
+    "\n",
+    "#Converting string types to floating values for plotting\n",
+    "\n",
+    "accuracy, disp_imp, fnr, fpr = [],[],[],[]\n",
+    "for item in s_accuracy:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    accuracy.append(temp_list)\n",
+    "\n",
+    "for item in s_disp_imp:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    disp_imp.append(temp_list)\n",
+    "\n",
+    "for item in s_fnr:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    fnr.append(temp_list)\n",
+    "\n",
+    "for item in s_fpr:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    fpr.append(temp_list)\n",
+    "\n",
+    "title_list = s_label"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plotter(s_optimal, title_list, label_list, x, y, x_ticks, x_label, main_title):\n",
+    "    '''\n",
+    "    Function to plot various comparison plots.\n",
+    "    '''\n",
+    "    #update label list and title list as per the experiment being performed.\n",
+    "    fig, axs = plt.subplots(2, 2, figsize=((10,20)))\n",
+    "    axs = axs.flatten()\n",
+    "    for i in range(0, len(y), 2):\n",
+    "        loc = i//2\n",
+    "        for k in range(len(x[i])):\n",
+    "            if s_optimal[i][k]:\n",
+    "                axs[loc].scatter(x[i][k], y[i][k], c='y', marker='o')\n",
+    "            else:\n",
+    "                axs[loc].scatter(x[i][k], y[i][k], c='b', marker='o')\n",
+    "            if s_optimal[i+1][k]:\n",
+    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='g', marker='o')\n",
+    "            else:\n",
+    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='r', marker='o')\n",
+    "        axs[loc].set_xticks(x_ticks)\n",
+    "        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))\n",
+    "        axs[loc].set_title(title_list[i])\n",
+    "        axs[loc].grid(True)\n",
+    "        axs[loc].set_xlabel(x_label)\n",
+    "        axs[loc].set_ylabel('Accuracy')\n",
+    "        axs[loc].legend(label_list[int(i%(len(label_list)*2)/2)])\n",
+    "    fig.suptitle(main_title)\n",
+    "    plt.subplots_adjust(wspace=0.3, hspace=0.43)\n",
+    "    fig.savefig('examples_skyline/' + main_title + '.png')\n",
+    "    plt.show()\n",
+    "\n",
+    "plotter(s_optimal, title_list, label_list, x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di-skyline_formula')\n",
+    "plotter(s_optimal, title_list, label_list, x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr-skyline_formula')\n",
+    "plotter(s_optimal, title_list, label_list, x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr-skyline_formula')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/examples_skyline/results_play_skyline_order.ipynb b/examples_skyline/results_play_skyline_order.ipynb
new file mode 100644
index 0000000..1ab6950
--- /dev/null
+++ b/examples_skyline/results_play_skyline_order.ipynb
@@ -0,0 +1,1767 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Import error: No module named 'tensorflow'\n"
+     ]
+    }
+   ],
+   "source": [
+    "#load relevant libraries\n",
+    "import os\n",
+    "import sys\n",
+    "import pathlib\n",
+    "import shutil\n",
+    "import warnings\n",
+    "warnings.simplefilter('ignore')\n",
+    "from helper import extract_info\n",
+    "\n",
+    "os.chdir('..')\n",
+    "#parent_dir = os.path.dirname(os.getcwd())\n",
+    "#sys.path.append(parent_dir)\n",
+    "\n",
+    "from fp.traindata_samplers import CompleteData\n",
+    "from fp.missingvalue_handlers import CompleteCaseAnalysis\n",
+    "from fp.dataset_experiments import GermanCreditDatasetSexExperiment\n",
+    "from fp.scalers import NamedStandardScaler, NamedMinMaxScaler\n",
+    "from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          \n",
+    "from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing\n",
+    "from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing\n",
+    "\n",
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#parameters in this cell can be adjusted for experimentation\n",
+    "\n",
+    "seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]\n",
+    "#seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe]\n",
+    "learners = [NonTunedLogisticRegression(), LogisticRegression()] \n",
+    "#tuned and non tuned DecisionTree() can also be used.\n",
+    "\n",
+    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()), (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]\n",
+    "skyline_order = ['accuracy', 'selection_rate', 'false_discovery_rate']\n",
+    "pre_processors = [NoPreProcessing(), DIRemover(1.0)]\n",
+    "post_processors = [NoPostProcessing(), RejectOptionPostProcessing()]\n",
+    "#add pair wise tuples for each pair of learners.\n",
+    "label_list = [('NonTunedLogistic', 'TunedLogistic')]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def calculate_metrics(seed, learner, pre_processor, post_processor, val_strategy):\n",
+    "    '''\n",
+    "    Experiment function to run the experiments\n",
+    "    '''\n",
+    "    exp = GermanCreditDatasetSexExperiment(\n",
+    "        fixed_random_seed=seed,\n",
+    "        train_data_sampler=CompleteData(),\n",
+    "        missing_value_handler=CompleteCaseAnalysis(),\n",
+    "        numeric_attribute_scaler=NamedStandardScaler(),\n",
+    "        learners=learner,\n",
+    "        pre_processors=pre_processor,\n",
+    "        post_processors=post_processor,\n",
+    "        optimal_validation_strategy=val_strategy)\n",
+    "    exp.run()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    2.0s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.6s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.6s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from training data\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
+      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n",
+      "complete_case removed 0 instances from validation data\n",
+      "Injecting zero columns for features not present set()\n"
+     ]
+    }
+   ],
+   "source": [
+    "def run_exp(seeds, learners, pre_processors, post_processors, optimal_validation_strategies):\n",
+    "    '''\n",
+    "    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.\n",
+    "    '''\n",
+    "    total_experiments = len(pre_processors) * len(post_processors) * len(learners)\n",
+    "    #pd.set_option(\"display.max_rows\", None, \"display.max_columns\", None)\n",
+    "    accuracy, disp_imp, fnr, fpr, label, optimal = [], [], [], [], [], []\n",
+    "    for j in range(total_experiments):\n",
+    "        accuracy.append(list())\n",
+    "        disp_imp.append(list())\n",
+    "        fnr.append(list())\n",
+    "        fpr.append(list())\n",
+    "        label.append(list())\n",
+    "        optimal.append(list())\n",
+    "    for seed in seeds:  \n",
+    "        learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal = [], [], [], [], [], []\n",
+    "        calculate_metrics(seed, learners, pre_processors, post_processors, optimal_validation_strategies)\n",
+    "        extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal)\n",
+    "        for i in range(len(learner_acc)):\n",
+    "            accuracy[i].append(learner_acc[i])\n",
+    "            disp_imp[i].append(learner_di[i])\n",
+    "            fnr[i].append(learner_fnr[i])\n",
+    "            fpr[i].append(learner_fpr[i])\n",
+    "            #label[i].append(learner_label[i])\n",
+    "            optimal[i].append(learner_optimal[i])\n",
+    "    label = learner_label\n",
+    "    \n",
+    "    return accuracy, disp_imp, fnr, fpr, label, optimal\n",
+    "    #return learner_acc, learner_di, learner_fnr, learner_fpr, label, optimal\n",
+    "\n",
+    "s_accuracy, s_disp_imp, s_fnr, s_fpr, s_label, s_optimal  = run_exp(seeds, learners, pre_processors, post_processors, skyline_order)\n",
+    "\n",
+    "#Converting string types to floating values for plotting\n",
+    "\n",
+    "accuracy, disp_imp, fnr, fpr = [],[],[],[]\n",
+    "for item in s_accuracy:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    accuracy.append(temp_list)\n",
+    "\n",
+    "for item in s_disp_imp:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    disp_imp.append(temp_list)\n",
+    "\n",
+    "for item in s_fnr:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    fnr.append(temp_list)\n",
+    "\n",
+    "for item in s_fpr:\n",
+    "    temp_list = list()\n",
+    "    for i in item:\n",
+    "        temp_list.append(float(i))\n",
+    "    fpr.append(temp_list)\n",
+    "\n",
+    "title_list = s_label"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plotter(s_optimal, title_list, label_list, x, y, x_ticks, x_label, main_title):\n",
+    "    '''\n",
+    "    Function to plot various comparison plots.\n",
+    "    '''\n",
+    "    #update label list and title list as per the experiment being performed.\n",
+    "    fig, axs = plt.subplots(len(pre_processors), len(post_processors), figsize=((10,20)))\n",
+    "    axs = axs.flatten()\n",
+    "    for i in range(0, len(y), 2):\n",
+    "        loc = i//2\n",
+    "        for k in range(len(x[i])):\n",
+    "            if s_optimal[i][k]:\n",
+    "                axs[loc].scatter(x[i][k], y[i][k], c='y', marker='o')\n",
+    "            else:\n",
+    "                axs[loc].scatter(x[i][k], y[i][k], c='b', marker='o')\n",
+    "            if s_optimal[i+1][k]:\n",
+    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='g', marker='o')\n",
+    "            else:\n",
+    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='r', marker='o')\n",
+    "        axs[loc].set_xticks(x_ticks)\n",
+    "        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))\n",
+    "        axs[loc].set_title(title_list[i])\n",
+    "        axs[loc].grid(True)\n",
+    "        axs[loc].set_xlabel(x_label)\n",
+    "        axs[loc].set_ylabel('Accuracy')\n",
+    "        axs[loc].legend(label_list[int(i%(len(label_list)*2)/2)])\n",
+    "    fig.suptitle(main_title)\n",
+    "    plt.subplots_adjust(wspace=0.3, hspace=0.43)\n",
+    "    fig.savefig('examples_skyline/' + main_title + '.png')\n",
+    "    plt.show()\n",
+    "\n",
+    "plotter(s_optimal, title_list, label_list, x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di-skyline-order')\n",
+    "plotter(s_optimal, title_list, label_list, x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr-skyline-order')\n",
+    "plotter(s_optimal, title_list, label_list, x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr-skyline-order')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/fp/dataset_experiments.py b/fp/dataset_experiments.py
index 08b07aa..91e0229 100644
--- a/fp/dataset_experiments.py
+++ b/fp/dataset_experiments.py
@@ -6,7 +6,7 @@
 class AdultDatasetWhiteMaleExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -27,12 +27,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}, {1.0: 'Male', 0.0: 'Female'}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultwhitemale')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'adultwhitemale', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -41,7 +45,7 @@ def load_raw_data(self):
 class AdultDatasetMaleExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -62,12 +66,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'Male', 0.0: 'Female'}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultmale')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'adultmale', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -76,7 +84,7 @@ def load_raw_data(self):
 class AdultDatasetWhiteExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -97,12 +105,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultwhite')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'adultwhite', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -112,7 +124,7 @@ def load_raw_data(self):
 class PropublicaDatasetWhiteExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
         label_name = 'two_year_recid'
@@ -132,12 +144,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'Caucasian', 0.0: 'Non-white'}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'propublicawhite')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'propublicawhite', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         """The custom pre-processing function is adapted from
@@ -163,7 +179,7 @@ class GermanCreditDatasetSexExperiment(BinaryClassificationExperiment):
 
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -186,12 +202,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'male', 0.0: 'female'}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'germancreditsex')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'germancreditsex', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/german.csv', na_values='?', sep=',')
@@ -253,7 +273,7 @@ class RicciRaceExperiment(BinaryClassificationExperiment):
     '''
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -274,12 +294,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'W', 0.0: 'NW'}]
         }
-        
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names,attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'riccirace')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'riccirace', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/ricci.txt', na_values='?', sep=',')
@@ -305,7 +329,7 @@ class GiveMeSomeCreditExperiment(BinaryClassificationExperiment):
     Fairness intervention for the Age attribute (priviledge for age>=25) while predicting if a person will experience 90 days past due delinquency or worse. 
     '''
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
+                 learners, pre_processors, post_processors, optimal_validation_strategy):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -327,12 +351,16 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
         dataset_metadata = {
             'label_maps': [{1.0: 1, 0.0: 0}]
         }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'givecredit')
+        if optimal_validation_strategy:
+            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
+                             pre_processors,
+                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                             unprivileged_groups, dataset_metadata, 'givecredit', optimal_validation_strategy)
+        else:
+            print("Need to specify the strategy that is used to select the optimal results!")
+            exit()
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/givemesomecredit.csv', na_values='?', sep=',',index_col=False)
diff --git a/fp/dataset_experiments_old.py b/fp/dataset_experiments_old.py
new file mode 100644
index 0000000..08b07aa
--- /dev/null
+++ b/fp/dataset_experiments_old.py
@@ -0,0 +1,341 @@
+from fp.experiments import BinaryClassificationExperiment
+import pandas as pd
+import numpy as np
+
+
+class AdultDatasetWhiteMaleExperiment(BinaryClassificationExperiment):
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'income-per-year'
+        positive_label = '>50K'
+        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
+        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
+                                       'native-country']
+        attributes_to_drop_names = ['fnlwgt']
+
+        protected_attribute_names = ['race', 'sex']
+        privileged_classes = [['White'], ['Male']]
+
+        privileged_groups = [{'race': 1, 'sex': 1}]
+        unprivileged_groups = [{'race': 1, 'sex': 0}, {'sex': 0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
+            'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}, {1.0: 'Male', 0.0: 'Female'}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultwhitemale')
+
+    def load_raw_data(self):
+        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
+
+
+class AdultDatasetMaleExperiment(BinaryClassificationExperiment):
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'income-per-year'
+        positive_label = '>50K'
+        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
+        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
+                                       'native-country']
+        attributes_to_drop_names = ['fnlwgt', 'race']
+
+        protected_attribute_names = ['sex']
+        privileged_classes = [['Male']]
+
+        privileged_groups = [{'sex': 1}]
+        unprivileged_groups = [{'sex': 0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
+            'protected_attribute_maps': [{1.0: 'Male', 0.0: 'Female'}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultmale')
+
+    def load_raw_data(self):
+        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
+
+
+class AdultDatasetWhiteExperiment(BinaryClassificationExperiment):
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'income-per-year'
+        positive_label = '>50K'
+        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
+        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
+                                       'native-country']
+        attributes_to_drop_names = ['fnlwgt', 'sex']
+
+        protected_attribute_names = ['race']
+        privileged_classes = [['White']]
+
+        privileged_groups = [{'race': 1}]
+        unprivileged_groups = [{'race': 0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
+            'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultwhite')
+
+    def load_raw_data(self):
+        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
+
+
+
+class PropublicaDatasetWhiteExperiment(BinaryClassificationExperiment):
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'two_year_recid'
+        positive_label = 1
+        numeric_attribute_names = ['age', 'decile_score', 'priors_count', 'days_b_screening_arrest', 'decile_score',
+                                   'is_recid']
+        categorical_attribute_names = ['c_charge_degree', 'age_cat', 'score_text']
+        attributes_to_drop_names = ['sex', 'c_jail_in', 'c_jail_out']
+
+        protected_attribute_names = ['race']
+        privileged_classes = [['Caucasian']]
+
+        privileged_groups = [{'race': 1}]
+        unprivileged_groups = [{'race': 0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: 1, 0.0: 0}],
+            'protected_attribute_maps': [{1.0: 'Caucasian', 0.0: 'Non-white'}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'propublicawhite')
+
+    def load_raw_data(self):
+        """The custom pre-processing function is adapted from
+        https://github.com/IBM/AIF360/blob/master/aif360/algorithms/preprocessing/optim_preproc_helpers/data_preproc_functions.py
+        https://github.com/fair-preprocessing/nips2017/blob/master/compas/code/Generate_Compas_Data.ipynb
+        """
+        df = pd.read_csv('datasets/raw/propublica-recidivism.csv', na_values='?', sep=',')
+        df = df[['age', 'c_charge_degree', 'race', 'age_cat', 'score_text',
+                 'sex', 'priors_count', 'days_b_screening_arrest', 'decile_score',
+                 'is_recid', 'two_year_recid', 'c_jail_in', 'c_jail_out']]
+        ix = df['days_b_screening_arrest'] <= 100
+        ix = (df['days_b_screening_arrest'] >= -100) & ix
+        ix = (df['is_recid'] != -1) & ix
+        ix = (df['c_charge_degree'] != "O") & ix
+        ix = (df['score_text'] != 'N/A') & ix
+        df = df.loc[ix, :]
+        df['length_of_stay'] = (pd.to_datetime(df['c_jail_out']) - pd.to_datetime(df['c_jail_in'])).apply(
+            lambda x: x.days)
+        return df
+
+
+class GermanCreditDatasetSexExperiment(BinaryClassificationExperiment):
+
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'credit'
+        positive_label = 1
+        numeric_attribute_names = ['month', 'credit_amount', 'residence_since', 'age', 'number_of_credits',
+                                   'people_liable_for']
+        categorical_attribute_names = ['credit_history', 'savings', 'employment']
+        attributes_to_drop_names = ['personal_status', 'status', 'purpose', 'investment_as_income_percentage',
+                                    'other_debtors', 'property', 'installment_plans', 'housing', 'skill_level',
+                                    'telephone', 'foreign_worker']
+
+        protected_attribute_names = ['sex']
+        privileged_classes = [[1.0]]
+
+        privileged_groups = [{'sex': 1.0}]
+        unprivileged_groups = [{'sex': 0.0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: 1, 0.0: 0}],
+            'protected_attribute_maps': [{1.0: 'male', 0.0: 'female'}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'germancreditsex')
+
+    def load_raw_data(self):
+        df = pd.read_csv('datasets/raw/german.csv', na_values='?', sep=',')
+
+        def group_credit_hist(x):
+            if x in ['A30', 'A31', 'A32']:
+                return 'None/Paid'
+            elif x == 'A33':
+                return 'Delay'
+            elif x == 'A34':
+                return 'Other'
+            else:
+                return 'NA'
+
+        def group_employ(x):
+            if x == 'A71':
+                return 'Unemployed'
+            elif x in ['A72', 'A73']:
+                return '1-4 years'
+            elif x in ['A74', 'A75']:
+                return '4+ years'
+            else:
+                return 'NA'
+
+        def group_savings(x):
+            if x in ['A61', 'A62']:
+                return '<500'
+            elif x in ['A63', 'A64']:
+                return '500+'
+            elif x == 'A65':
+                return 'Unknown/None'
+            else:
+                return 'NA'
+
+        def group_status(x):
+            if x in ['A11', 'A12']:
+                return '<200'
+            elif x in ['A13']:
+                return '200+'
+            elif x == 'A14':
+                return 'None'
+            else:
+                return 'NA'
+
+        status_map = {'A91': 1.0, 'A93': 1.0, 'A94': 1.0, 'A92': 0.0, 'A95': 0.0}
+        df['sex'] = df['personal_status'].replace(status_map)
+        #         group credit history, savings, and employment
+        df['credit_history'] = df['credit_history'].apply(lambda x: group_credit_hist(x))
+        df['savings'] = df['savings'].apply(lambda x: group_savings(x))
+        df['employment'] = df['employment'].apply(lambda x: group_employ(x))
+        df['age'] = df['age'].apply(lambda x: np.float(x >= 25))
+        df['status'] = df['status'].apply(lambda x: group_status(x))
+        return df
+   
+class RicciRaceExperiment(BinaryClassificationExperiment):
+    '''
+    Check for fairness based on race (white vs minority i.e Black and Hispanic) while predicting if a candidate will pass i.e obtain total 
+    marks greater than or equal to 70.0  
+    '''
+
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        
+        label_name = 'combine'
+        positive_label = 1
+        numeric_attribute_names = ['oral', 'written']
+        categorical_attribute_names = ['position']
+        attributes_to_drop_names = []
+                
+        protected_attribute_names = ['race']
+        privileged_classes = [[1.0]]
+        
+        privileged_groups = [{'race': 1.0}]
+        unprivileged_groups = [{'race': 0.0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: 1, 0.0: 0}],
+            'protected_attribute_maps': [{1.0: 'W', 0.0: 'NW'}]
+        }
+        
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names,attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'riccirace')
+
+    def load_raw_data(self):
+        df = pd.read_csv('datasets/raw/ricci.txt', na_values='?', sep=',')
+        df.columns = map(str.lower, df.columns)
+
+        def group_race_minority(x):
+            if x in ['B', 'H', 'B']:
+                return 'NW'
+            else:
+                return 'W'
+
+        post_map = {'Captain': 0.0, 'Lieutenant': 1.0}
+        df['position'] = df['position'].replace(post_map)
+        
+        #group minorities i.e Black and Hispanic are combined to 'NW'(non white)
+        df['race'] = df['race'].apply(lambda x: group_race_minority(x))
+        df['combine'] = df['combine'].apply(lambda x: int(x >= 70))
+
+        return df
+
+class GiveMeSomeCreditExperiment(BinaryClassificationExperiment):
+    '''
+    Fairness intervention for the Age attribute (priviledge for age>=25) while predicting if a person will experience 90 days past due delinquency or worse. 
+    '''
+    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
+                 learners, pre_processors, post_processors):
+
+        test_set_ratio = 0.2
+        validation_set_ratio = 0.1
+        label_name = 'SeriousDlqin2yrs'
+        positive_label = 1
+
+        numeric_attribute_names = ['RevolvingUtilizationOfUnsecuredLines','age','NumberOfTime30-59DaysPastDueNotWorse',
+                                  'DebtRatio','MonthlyIncome','NumberOfOpenCreditLinesAndLoans','NumberOfTimes90DaysLate',
+                                  'NumberRealEstateLoansOrLines','NumberOfTime60-89DaysPastDueNotWorse','NumberOfDependents']
+        categorical_attribute_names = []
+        attributes_to_drop_names = []
+
+        protected_attribute_names = ['age']
+        privileged_classes = [[1.0]]
+
+        privileged_groups = [{'age': 1}]
+        unprivileged_groups = [{'age': 0}]
+
+        dataset_metadata = {
+            'label_maps': [{1.0: 1, 0.0: 0}]
+        }
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'givecredit')
+
+    def load_raw_data(self):
+        df = pd.read_csv('datasets/raw/givemesomecredit.csv', na_values='?', sep=',',index_col=False)
+        df['age'] = df['age'].apply(lambda x: int(x >= 25))
+        return df
+        
diff --git a/fp/experiments.py b/fp/experiments.py
index cb5f33c..bbc5d5e 100644
--- a/fp/experiments.py
+++ b/fp/experiments.py
@@ -10,6 +10,7 @@
 from sklearn.base import clone
 from sklearn.metrics import roc_auc_score
 from sklearn.model_selection import train_test_split
+from fp.utils import filter_optimal_results_skyline_order, filter_optimal_results_skyline_formula
 
 
 class BinaryClassificationExperiment:
@@ -35,7 +36,8 @@ def __init__(self,
                  privileged_groups,
                  unprivileged_groups,
                  dataset_metadata,
-                 dataset_name):
+                 dataset_name,
+                 optimal_validation_strategy):
 
         self.fixed_random_seed = fixed_random_seed
         self.test_set_ratio = test_set_ratio
@@ -59,20 +61,21 @@ def __init__(self,
         self.dataset_name = dataset_name
         self.log_path = 'logs/'
         self.exec_timestamp = self.generate_timestamp()
+        self.optimal_validation_strategy = optimal_validation_strategy
 
 
 
     # --- Helper Methods Begin ------------------------------------------------
 
 
-    def unique_file_name(self, learner, pre_processor, post_processor):
-        return '{}__{}__{}__{}__{}__{}__{}'.format(self.dataset_name,
+    def unique_file_name(self, pre_processor, post_processor, learner):
+        return '{0}__{1}__{2}__{3}__{4}__{5}__{6}'.format(self.dataset_name,
+                                                   pre_processor.name(),
+                                                   post_processor.name(),
                                                    learner.name(),
                                                    self.missing_value_handler.name(),
                                                    self.train_data_sampler.name(),
-                                                   self.numeric_attribute_scaler.name(),
-                                                   pre_processor.name(),
-                                                   post_processor.name())
+                                                   self.numeric_attribute_scaler.name())
 
 
     def generate_file_path(self, file_name=''):
@@ -131,7 +134,6 @@ def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processo
 
         feature_names_in_train_but_not_in_current = set(train_feature_names).difference(
             set(current_feature_names))
-
         print("Injecting zero columns for features not present", feature_names_in_train_but_not_in_current)
 
         validation_data_df, _ = adjusted_annotated_data.convert_to_dataframe()
@@ -147,14 +149,12 @@ def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processo
         if learner.needs_annotated_data_for_prediction():
             adjusted_annotated__data_with_predictions = model.predict(adjusted_annotated_data)
         else:
-            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features)
-
+            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features).reshape(-1,1)
             try:
                 class_probs = model.predict_proba(adjusted_annotated_data.features)
                 adjusted_annotated__data_with_predictions.scores = class_probs[:, 0]
             except AttributeError:
                 print("WARNING: MODEL CANNOT ASSIGN CLASS PROBABILITIES")
-
         return adjusted_annotated_data, adjusted_annotated__data_with_predictions
 
 
@@ -235,10 +235,11 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
         post_processor : fairprep pre-processor abstraction from 
             aif360.algorithms.post_processors
         """
-        
+
         adjusted_annotated_train_data = self.preprocess_data(pre_processor, annotated_train_data)
-        model = self.learn_classifier(learner, adjusted_annotated_train_data, self.fixed_random_seed)
 
+        model = self.learn_classifier(learner, adjusted_annotated_train_data, self.fixed_random_seed)
+        
         adjusted_annotated_train_data_with_predictions = adjusted_annotated_train_data.copy()
 
         if learner.needs_annotated_data_for_prediction():
@@ -260,13 +261,12 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
             adjusted_annotated_test_data_with_predictions)
 
         results_file_name = '../{}{}-{}.csv'.format(
-            self.generate_file_path(), self.unique_file_name(pre_processor, learner, post_processor), self.fixed_random_seed)
+            self.generate_file_path(), self.unique_file_name(pre_processor, post_processor, learner), self.fixed_random_seed)
         results_file_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), results_file_name)
         
         results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
         if not os.path.exists(results_dir_name):
             os.makedirs(results_dir_name)
-        
         results_file = []
         
         results_file = self.log_metrics(results_file, model, adjusted_annotated_validation_data,
@@ -280,45 +280,70 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
         results_file.to_csv(results_file_path, index=False)
 
 
+
     def filter_optimal_results(self):
         """Identifies the experiment(s) with the highest accuracy as optimal 
             result. Keeps the test metrics just for the experiment(s) with the
             optimal result.
         """
+
         results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
         results_dir = os.listdir(Path(results_dir_name))
-        accuracies = dict()
-        max_accuracy = 0
 
-        # Fetching the accuracy from the row('val', 'None', 'accuracy') of all the experiment results
+        ##### SKYLINE FORMULA IMPLEMENTATION
+
+        privileged_metric_names = ['num_true_positives', 'num_false_positives', 'num_false_negatives',
+                                   'num_true_negatives', 'num_generalized_true_positives',
+                                   'num_generalized_false_positives', 'num_generalized_false_negatives',
+                                   'num_generalized_true_negatives', 'true_positive_rate', 'false_positive_rate',
+                                   'false_negative_rate', 'true_negative_rate', 'generalized_true_positive_rate',
+                                   'generalized_false_positive_rate', 'generalized_false_negative_rate',
+                                   'generalized_true_negative_rate', 'positive_predictive_value',
+                                   'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value',
+                                   'accuracy', 'error_rate', 'num_pred_positives', 'num_pred_negatives',
+                                   'selection_rate']
+         
+        dictionary = {}
+        filenames = list()
         for result_filename in results_dir:
             file_path = os.path.join(results_dir_name, result_filename)
             result_df = pd.read_csv(file_path)
             result_df.fillna(value='', inplace=True)
-            accuracy = (result_df.loc[(result_df['Split'] == 'val') & 
+            for privileged_metric in privileged_metric_names:
+                if privileged_metric not in dictionary:
+                    dictionary[privileged_metric] = list()
+                p_metric = (result_df.loc[(result_df['Split'] == 'val') & 
                                       (result_df['PrivilegedStatus'] == '') & 
-                                      (result_df['MetricName'] == 'accuracy'), 'MetricValue'].values[0])
-            accuracies[result_filename] = accuracy
-            if accuracy > max_accuracy:
-                max_accuracy = accuracy
+                                      (result_df['MetricName'] == privileged_metric), 'MetricValue'].values[0])
+                dictionary[privileged_metric].append(p_metric)
+            filenames.append(result_filename)
+
+        privileged_metric_values = pd.DataFrame(dictionary)
+        privileged_metric_values['filenames'] = filenames
+
+        if isinstance(self.optimal_validation_strategy, dict):
+            skyline_result = filter_optimal_results_skyline_formula(privileged_metric_values, self.optimal_validation_strategy)
+        else:
+            skyline_result = filter_optimal_results_skyline_order(privileged_metric_values, self.optimal_validation_strategy)
 
-        # List of non optimal and optimal filenames and accuracy
         non_optimal_filenames = list()
         optimal_filenames = list()
-        for filename, accuracy in accuracies.items():
-            if accuracy != max_accuracy:
-                non_optimal_filenames.append(filename)
+        filenames_list = privileged_metric_values['filenames'].tolist()
+        for file_name in filenames_list:
+            if file_name == skyline_result[-1]:
+                optimal_filenames.append(file_name)
             else:
-                optimal_filenames.append(filename)
+                non_optimal_filenames.append(file_name)
 
         # Removing the test results from the non optimal experiment results      
+        '''
         for file_name in non_optimal_filenames:
-            file_path = os.path.join(results_dir_name, result_filename)
+            file_path = os.path.join(results_dir_name, file_name)
             result_df = pd.read_csv(file_path)
             result_df = result_df[(result_df['Split'] != 'test')]
             os.remove(file_path)
             result_df.to_csv(file_path, index=False, header=False)
-
+        '''
         # Renaming the optimal experiment results file (or files if tie) 
         for file_name in optimal_filenames:
             file_path = os.path.join(results_dir_name, file_name)
@@ -341,6 +366,7 @@ def run(self):
                                                                     self.validation_set_ratio,
                                                                     random_state=self.fixed_random_seed)
 
+        #Just returns complete data
         train_data = self.train_data_sampler.sample(all_train_data)
 
         second_split_ratio = self.test_set_ratio / (self.test_set_ratio + self.validation_set_ratio)
@@ -348,6 +374,8 @@ def run(self):
         validation_data, test_data = train_test_split(test_and_validation_data, test_size=second_split_ratio,
                                                       random_state=self.fixed_random_seed)
 
+
+        #fit just passes and hadle_missing does drop_na operation
         self.missing_value_handler.fit(train_data)
         filtered_train_data = self.missing_value_handler.handle_missing(train_data)
 
@@ -374,10 +402,8 @@ def run(self):
             features_to_drop=self.attributes_to_drop_names,
             metadata=self.dataset_metadata
         )
-
         for pre_processor in self.pre_processors:
             for learner in self.learners:
                 for post_processor in self.post_processors:
-                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, 
-                                        pre_processor, learner, post_processor)
+                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, pre_processor, learner, post_processor)
         self.filter_optimal_results()
diff --git a/fp/experiments_old.py b/fp/experiments_old.py
new file mode 100644
index 0000000..cb5f33c
--- /dev/null
+++ b/fp/experiments_old.py
@@ -0,0 +1,383 @@
+import os
+import numpy as np
+import pandas as pd
+
+from time import time
+from pathlib import Path
+from datetime import datetime
+from aif360.datasets import StandardDataset
+from aif360.metrics import ClassificationMetric
+from sklearn.base import clone
+from sklearn.metrics import roc_auc_score
+from sklearn.model_selection import train_test_split
+
+
+class BinaryClassificationExperiment:
+
+
+    def __init__(self,
+                 fixed_random_seed,
+                 test_set_ratio,
+                 validation_set_ratio,
+                 label_name,
+                 positive_label,
+                 numeric_attribute_names,
+                 categorical_attribute_names,
+                 attributes_to_drop_names,
+                 train_data_sampler,
+                 missing_value_handler,
+                 numeric_attribute_scaler,
+                 learners,
+                 pre_processors,
+                 post_processors,
+                 protected_attribute_names,
+                 privileged_classes,
+                 privileged_groups,
+                 unprivileged_groups,
+                 dataset_metadata,
+                 dataset_name):
+
+        self.fixed_random_seed = fixed_random_seed
+        self.test_set_ratio = test_set_ratio
+        self.validation_set_ratio = validation_set_ratio
+        self.label_name = label_name
+        self.positive_label = positive_label
+        self.numeric_attribute_names = numeric_attribute_names
+        self.categorical_attribute_names = categorical_attribute_names
+        self.attributes_to_drop_names = attributes_to_drop_names
+        self.train_data_sampler = train_data_sampler
+        self.missing_value_handler = missing_value_handler
+        self.numeric_attribute_scaler = numeric_attribute_scaler
+        self.learners = learners
+        self.pre_processors = pre_processors
+        self.post_processors = post_processors
+        self.protected_attribute_names = protected_attribute_names
+        self.privileged_classes = privileged_classes
+        self.privileged_groups = privileged_groups
+        self.unprivileged_groups = unprivileged_groups
+        self.dataset_metadata = dataset_metadata
+        self.dataset_name = dataset_name
+        self.log_path = 'logs/'
+        self.exec_timestamp = self.generate_timestamp()
+
+
+
+    # --- Helper Methods Begin ------------------------------------------------
+
+
+    def unique_file_name(self, learner, pre_processor, post_processor):
+        return '{}__{}__{}__{}__{}__{}__{}'.format(self.dataset_name,
+                                                   learner.name(),
+                                                   self.missing_value_handler.name(),
+                                                   self.train_data_sampler.name(),
+                                                   self.numeric_attribute_scaler.name(),
+                                                   pre_processor.name(),
+                                                   post_processor.name())
+
+
+    def generate_file_path(self, file_name=''):
+        dir_name = '{}_{}/'.format(self.exec_timestamp, self.dataset_name)
+        return self.log_path + dir_name + file_name
+
+
+    def generate_timestamp(self):
+        return datetime.fromtimestamp(time()).strftime('%Y-%m-%d_%H-%M-%S-%f')[:-3]
+
+
+    def load_raw_data(self):
+        raise NotImplementedError
+
+
+    def learn_classifier(self, learner, annotated_train_data, fixed_random_seed):
+        return learner.fit_model(annotated_train_data, fixed_random_seed)
+
+
+    def preprocess_data(self, pre_processor, annotated_dataset):
+        return pre_processor.pre_process(annotated_dataset, self.privileged_groups, self.unprivileged_groups)
+
+
+    def post_process_predictions(self, post_processor, validation_dataset, validation_dataset_with_predictions,
+                                 testset_with_predictions):
+        return post_processor.post_process(validation_dataset, validation_dataset_with_predictions,
+                                                testset_with_predictions, self.fixed_random_seed,
+                                                self.privileged_groups, self.unprivileged_groups)
+
+
+    def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processor, learner, model):
+        filtered_data = self.missing_value_handler.handle_missing(data)
+        print(self.missing_value_handler.name(), 'removed', len(data) - len(filtered_data),
+              'instances from validation data')
+
+        for numerical_attribute, scaler in scalers.items():
+            numerical_attribute_data = np.array(filtered_data[numerical_attribute]).reshape(-1, 1)
+            scaled_numerical_attribute_data = scaler.transform(numerical_attribute_data)
+            filtered_data.loc[:, numerical_attribute] = scaled_numerical_attribute_data
+
+        annotated_data = StandardDataset(
+            df=filtered_data,
+            label_name=self.label_name,
+            favorable_classes=[self.positive_label],
+            protected_attribute_names=self.protected_attribute_names,
+            privileged_classes=self.privileged_classes,
+            categorical_features=self.categorical_attribute_names,
+            features_to_drop=self.attributes_to_drop_names,
+            metadata=self.dataset_metadata
+        )
+
+        adjusted_annotated_data = self.preprocess_data(pre_processor, annotated_data)
+
+        train_feature_names = adjusted_annotated_train_data.feature_names
+        current_feature_names = adjusted_annotated_data.feature_names
+
+        feature_names_in_train_but_not_in_current = set(train_feature_names).difference(
+            set(current_feature_names))
+
+        print("Injecting zero columns for features not present", feature_names_in_train_but_not_in_current)
+
+        validation_data_df, _ = adjusted_annotated_data.convert_to_dataframe()
+
+        for feature_name in feature_names_in_train_but_not_in_current:
+            validation_data_df.loc[:, feature_name] = 0.0
+
+        adjusted_annotated_data.feature_names = train_feature_names
+        adjusted_annotated_data.features = validation_data_df[train_feature_names].values.copy()
+
+        adjusted_annotated__data_with_predictions = adjusted_annotated_data.copy()
+
+        if learner.needs_annotated_data_for_prediction():
+            adjusted_annotated__data_with_predictions = model.predict(adjusted_annotated_data)
+        else:
+            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features)
+
+            try:
+                class_probs = model.predict_proba(adjusted_annotated_data.features)
+                adjusted_annotated__data_with_predictions.scores = class_probs[:, 0]
+            except AttributeError:
+                print("WARNING: MODEL CANNOT ASSIGN CLASS PROBABILITIES")
+
+        return adjusted_annotated_data, adjusted_annotated__data_with_predictions
+
+
+    def log_metrics(self, results_file, model, annotated_data, annotated_data_with_predictions, prefix):
+        metric = ClassificationMetric(annotated_data, annotated_data_with_predictions,
+                                      unprivileged_groups=self.unprivileged_groups,
+                                      privileged_groups=self.privileged_groups)
+
+        privileged_metric_names = ['num_true_positives', 'num_false_positives', 'num_false_negatives',
+                                   'num_true_negatives', 'num_generalized_true_positives',
+                                   'num_generalized_false_positives', 'num_generalized_false_negatives',
+                                   'num_generalized_true_negatives', 'true_positive_rate', 'false_positive_rate',
+                                   'false_negative_rate', 'true_negative_rate', 'generalized_true_positive_rate',
+                                   'generalized_false_positive_rate', 'generalized_false_negative_rate',
+                                   'generalized_true_negative_rate', 'positive_predictive_value',
+                                   'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value',
+                                   'accuracy', 'error_rate', 'num_pred_positives', 'num_pred_negatives',
+                                   'selection_rate']
+
+        for maybe_privileged in [None, True, False]:
+            for metric_name in privileged_metric_names:
+                metric_function = getattr(metric, metric_name)
+                metric_value = metric_function(privileged=maybe_privileged)
+                results_file.append([prefix, maybe_privileged, metric_name, metric_value])
+
+        if hasattr(model, 'predict_proba'):
+            auc = roc_auc_score(annotated_data.labels, model.predict_proba(annotated_data.features)[:, 1])
+        else:
+            auc = None
+
+        results_file.append([prefix, '', 'roc_auc', auc])
+
+        global_metric_names = ['true_positive_rate_difference', 'false_positive_rate_difference',
+                               'false_negative_rate_difference', 'false_omission_rate_difference',
+                               'false_discovery_rate_difference', 'false_positive_rate_ratio',
+                               'false_negative_rate_ratio', 'false_omission_rate_ratio',
+                               'false_discovery_rate_ratio', 'average_odds_difference', 'average_abs_odds_difference',
+                               'error_rate_difference', 'error_rate_ratio', 'disparate_impact',
+                               'statistical_parity_difference', 'generalized_entropy_index',
+                               'between_all_groups_generalized_entropy_index',
+                               'between_group_generalized_entropy_index', 'theil_index', 'coefficient_of_variation',
+                               'between_group_theil_index', 'between_group_coefficient_of_variation',
+                               'between_all_groups_theil_index', 'between_all_groups_coefficient_of_variation']
+
+        for metric_name in global_metric_names:
+            metric_function = getattr(metric, metric_name)
+            metric_value = metric_function()
+            results_file.append([prefix, '', metric_name, metric_value])
+
+        return results_file
+
+
+    # --- Helper Methods End --------------------------------------------------
+
+    
+    def run_single_exp(self, annotated_train_data, validation_data, test_data, scalers, pre_processor,
+                       learner, post_processor):
+        """Executes a single instance of experiment out of all the possible 
+        experiments from the given parameters.
+        
+        Parameters:
+        -----------
+        annotated_train_data : annotated aif360.datasets.StandardDataset of 
+            train data
+
+        validation_data : pandas dataframe of validation data
+
+        test_data : pandas dataframe of test data
+
+        scalers : dictionary with (key='feature name', value='type of scaler')
+
+        pre_processor : fairprep pre-processor abstraction from 
+            aif360.algorithms.pre_processors
+    
+        learner : fairprep learner abstraction from sci-kit learn or 
+            aif360.algorithms.inprocessing
+    
+        post_processor : fairprep pre-processor abstraction from 
+            aif360.algorithms.post_processors
+        """
+        
+        adjusted_annotated_train_data = self.preprocess_data(pre_processor, annotated_train_data)
+        model = self.learn_classifier(learner, adjusted_annotated_train_data, self.fixed_random_seed)
+
+        adjusted_annotated_train_data_with_predictions = adjusted_annotated_train_data.copy()
+
+        if learner.needs_annotated_data_for_prediction():
+            adjusted_annotated_train_data_with_predictions = model.predict(
+                adjusted_annotated_train_data_with_predictions)
+        else:
+            adjusted_annotated_train_data_with_predictions.labels = model.predict(
+                adjusted_annotated_train_data_with_predictions.features)
+
+        adjusted_annotated_validation_data, adjusted_annotated_validation_data_with_predictions = \
+            self.apply_model(validation_data, scalers, adjusted_annotated_train_data, pre_processor, learner, model)
+
+        adjusted_annotated_test_data, adjusted_annotated_test_data_with_predictions = \
+            self.apply_model(test_data, scalers, adjusted_annotated_train_data, pre_processor, learner, model)
+
+        adjusted_annotated_test_data_with_predictions = self.post_process_predictions(post_processor,
+            adjusted_annotated_validation_data,
+            adjusted_annotated_validation_data_with_predictions,
+            adjusted_annotated_test_data_with_predictions)
+
+        results_file_name = '../{}{}-{}.csv'.format(
+            self.generate_file_path(), self.unique_file_name(pre_processor, learner, post_processor), self.fixed_random_seed)
+        results_file_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), results_file_name)
+        
+        results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
+        if not os.path.exists(results_dir_name):
+            os.makedirs(results_dir_name)
+        
+        results_file = []
+        
+        results_file = self.log_metrics(results_file, model, adjusted_annotated_validation_data,
+                                        adjusted_annotated_validation_data_with_predictions, 'val')
+        results_file = self.log_metrics(results_file, model, adjusted_annotated_train_data,
+                                        adjusted_annotated_train_data_with_predictions, 'train')
+        results_file = self.log_metrics(results_file, model, adjusted_annotated_test_data, 
+                                        adjusted_annotated_test_data_with_predictions, 'test')
+        
+        results_file = pd.DataFrame(results_file, columns=['Split', 'PrivilegedStatus', 'MetricName', 'MetricValue'])
+        results_file.to_csv(results_file_path, index=False)
+
+
+    def filter_optimal_results(self):
+        """Identifies the experiment(s) with the highest accuracy as optimal 
+            result. Keeps the test metrics just for the experiment(s) with the
+            optimal result.
+        """
+        results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
+        results_dir = os.listdir(Path(results_dir_name))
+        accuracies = dict()
+        max_accuracy = 0
+
+        # Fetching the accuracy from the row('val', 'None', 'accuracy') of all the experiment results
+        for result_filename in results_dir:
+            file_path = os.path.join(results_dir_name, result_filename)
+            result_df = pd.read_csv(file_path)
+            result_df.fillna(value='', inplace=True)
+            accuracy = (result_df.loc[(result_df['Split'] == 'val') & 
+                                      (result_df['PrivilegedStatus'] == '') & 
+                                      (result_df['MetricName'] == 'accuracy'), 'MetricValue'].values[0])
+            accuracies[result_filename] = accuracy
+            if accuracy > max_accuracy:
+                max_accuracy = accuracy
+
+        # List of non optimal and optimal filenames and accuracy
+        non_optimal_filenames = list()
+        optimal_filenames = list()
+        for filename, accuracy in accuracies.items():
+            if accuracy != max_accuracy:
+                non_optimal_filenames.append(filename)
+            else:
+                optimal_filenames.append(filename)
+
+        # Removing the test results from the non optimal experiment results      
+        for file_name in non_optimal_filenames:
+            file_path = os.path.join(results_dir_name, result_filename)
+            result_df = pd.read_csv(file_path)
+            result_df = result_df[(result_df['Split'] != 'test')]
+            os.remove(file_path)
+            result_df.to_csv(file_path, index=False, header=False)
+
+        # Renaming the optimal experiment results file (or files if tie) 
+        for file_name in optimal_filenames:
+            file_path = os.path.join(results_dir_name, file_name)
+            optimal_file_name = '{}{}'.format(file_name[:-4], '__OPTIMAL.csv')
+            optimal_file_path = os.path.join(results_dir_name, optimal_file_name)
+            os.rename(file_path, optimal_file_path)
+
+
+    def run(self):
+        """Executes all the possible experiments from the combination of  given
+            learners, pre-processors and post-processors.
+            
+            No. of experiments = (#learners * #preprocessors * #postprocessors)
+        """
+        np.random.seed(self.fixed_random_seed)
+
+        data = self.load_raw_data()
+
+        all_train_data, test_and_validation_data = train_test_split(data, test_size=self.test_set_ratio +
+                                                                    self.validation_set_ratio,
+                                                                    random_state=self.fixed_random_seed)
+
+        train_data = self.train_data_sampler.sample(all_train_data)
+
+        second_split_ratio = self.test_set_ratio / (self.test_set_ratio + self.validation_set_ratio)
+
+        validation_data, test_data = train_test_split(test_and_validation_data, test_size=second_split_ratio,
+                                                      random_state=self.fixed_random_seed)
+
+        self.missing_value_handler.fit(train_data)
+        filtered_train_data = self.missing_value_handler.handle_missing(train_data)
+
+        print(self.missing_value_handler.name(), 'removed', len(train_data) - len(filtered_train_data),
+              'instances from training data')
+
+        scalers = {}
+
+        for numerical_attribute in self.numeric_attribute_names:
+            numerical_attribute_data = np.array(filtered_train_data[numerical_attribute]).reshape(-1, 1)
+            scaler = clone(self.numeric_attribute_scaler).fit(numerical_attribute_data)
+            scaled_numerical_attribute_data = scaler.transform(numerical_attribute_data)
+
+            filtered_train_data.loc[:, numerical_attribute] = scaled_numerical_attribute_data
+            scalers[numerical_attribute] = scaler
+
+        annotated_train_data = StandardDataset(
+            df=filtered_train_data,
+            label_name=self.label_name,
+            favorable_classes=[self.positive_label],
+            protected_attribute_names=self.protected_attribute_names,
+            privileged_classes=self.privileged_classes,
+            categorical_features=self.categorical_attribute_names,
+            features_to_drop=self.attributes_to_drop_names,
+            metadata=self.dataset_metadata
+        )
+
+        for pre_processor in self.pre_processors:
+            for learner in self.learners:
+                for post_processor in self.post_processors:
+                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, 
+                                        pre_processor, learner, post_processor)
+        self.filter_optimal_results()
diff --git a/fp/utils.py b/fp/utils.py
new file mode 100644
index 0000000..6d9113e
--- /dev/null
+++ b/fp/utils.py
@@ -0,0 +1,43 @@
+import pandas as pd
+
+def filter_optimal_results_skyline_order(_df, _order_list):
+    _df['selection_rate'] = abs(1 - _df['selection_rate'])
+    higher_is_better = ['num_true_positives', 'num_true_negatives', 'num_generalized_true_positives',
+                        'num_generalized_true_negatives', 'true_positive_rate', 'true_negative_rate',
+                        'generalized_true_positive_rate', 'generalized_true_negative_rate', 'positive_predictive_value',
+                        'accuracy', 'num_pred_positives']
+    lower_is_better = ['selection_rate', 'num_false_positives', 'num_false_negatives',
+                       'num_generalized_false_positives', 'num_generalized_false_negatives', 'false_positive_rate',
+                       'false_negative_rate', 'generalized_false_positive_rate', 'generalized_false_negative_rate',
+                       'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value', 'error_rate',
+                       'num_pred_negatives']
+    order = []
+    for item in _order_list:
+        if item in higher_is_better:
+            order.append(False)
+        else:
+            order.append(True)
+    _df = _df.sort_values(_order_list, ascending=order)
+
+    return _df.values[0]
+
+
+def filter_optimal_results_skyline_formula(_df, _formula):
+    df = pd.DataFrame()
+    for key in _formula:
+        df["norm_" + key] = (_df[key] - _df[key].min()) / (_df[key].max() - _df[key].min())
+
+    df_temp = list(_formula.values())
+    keys = list(_formula.keys())
+    for col in range(len(keys)):
+        keys[col] = "norm_" + keys[col]
+
+    # Multiplying with the multiplier to perform sorting operation
+    df['norm_avg'] = df[keys].multiply(df_temp).sum(axis=1)
+    frames = [_df, df]
+    df_fin = pd.concat(frames, axis=1)
+
+    df_fin = df_fin.sort_values(by='norm_avg', ascending=False)
+    cols = [c for c in df_fin.columns if c[:4] != 'norm']
+    df_fin = df_fin[cols]
+    return df_fin.values[0]
\ No newline at end of file

From 19ccc5953d51ebae07cf80b4127d1bc0f7b0c8a8 Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Mon, 22 Jun 2020 09:38:59 -0400
Subject: [PATCH 2/7] Add pipeline constructors

---
 pipeline/fairness_label.py                  |  74 +++
 pipeline/label_pipeline.py                  | 506 ++++++++++++++++++++
 pipeline/model/classifiers.py               |  87 ++++
 pipeline/model/fair_classifiers.py          |  69 +++
 pipeline/model/inprocessor.py               |  74 +++
 pipeline/postprocess/fair_postprocessors.py |  66 +++
 pipeline/postprocess/postprocessor.py       |  59 +++
 pipeline/preprocess/categorizers.py         |  51 ++
 pipeline/preprocess/encoders.py             |  98 ++++
 pipeline/preprocess/fair_preprocessors.py   |  82 ++++
 pipeline/preprocess/filters.py              |  69 +++
 pipeline/preprocess/imputers.py             |  88 ++++
 pipeline/preprocess/preprocessor.py         |  71 +++
 pipeline/preprocess/samplers.py             |  96 ++++
 pipeline/preprocess/scalers.py              |  47 ++
 pipeline/preprocess/splitters.py            | 134 ++++++
 pipeline/preprocess/transformers.py         |  24 +
 pipeline/step.py                            |  49 ++
 18 files changed, 1744 insertions(+)
 create mode 100644 pipeline/fairness_label.py
 create mode 100644 pipeline/label_pipeline.py
 create mode 100644 pipeline/model/classifiers.py
 create mode 100644 pipeline/model/fair_classifiers.py
 create mode 100644 pipeline/model/inprocessor.py
 create mode 100644 pipeline/postprocess/fair_postprocessors.py
 create mode 100644 pipeline/postprocess/postprocessor.py
 create mode 100644 pipeline/preprocess/categorizers.py
 create mode 100644 pipeline/preprocess/encoders.py
 create mode 100644 pipeline/preprocess/fair_preprocessors.py
 create mode 100644 pipeline/preprocess/filters.py
 create mode 100644 pipeline/preprocess/imputers.py
 create mode 100644 pipeline/preprocess/preprocessor.py
 create mode 100644 pipeline/preprocess/samplers.py
 create mode 100644 pipeline/preprocess/scalers.py
 create mode 100644 pipeline/preprocess/splitters.py
 create mode 100644 pipeline/preprocess/transformers.py
 create mode 100644 pipeline/step.py

diff --git a/pipeline/fairness_label.py b/pipeline/fairness_label.py
new file mode 100644
index 0000000..ce74e6d
--- /dev/null
+++ b/pipeline/fairness_label.py
@@ -0,0 +1,74 @@
+from sklearn.metrics import confusion_matrix
+import numpy as np
+def get_static_label(df, sensi_atts, target_name, round_digit=3):
+    groupby_cols = sensi_atts+[target_name]
+    placeholder_att = list(set(df.columns).difference(groupby_cols))[0]
+    
+    count_all = df[groupby_cols+[placeholder_att]].groupby(groupby_cols).count()
+    values_all = count_all.get_values()
+    index_all = list(count_all.index)
+
+    if len(sensi_atts) == 1:
+        norm_cols = [target_name]
+    elif len(sensi_atts) == 2:
+        norm_cols = [sensi_atts[0], target_name]
+    norm_values = df[norm_cols+[placeholder_att]].groupby(norm_cols).count().get_values()
+    
+    res_dict = {}
+    if 0 < len(sensi_atts) <= 2:
+        s1_n = len(df[sensi_atts[0]].unique())
+        t_n = len(df[target_name].unique())
+        for idx, tuple_i in enumerate(index_all):
+            if len(tuple_i[:-1]) == 1:
+                key_tuple = (tuple_i[0])
+            else:
+                key_tuple = tuple_i[:-1]
+            idx_denom = idx % 2 + int(idx / (s1_n*t_n))*t_n # only work for binary 2nd sensitive att
+            if key_tuple not in res_dict:
+                res_dict[key_tuple] = {tuple_i[-1]: round(values_all[idx][0]/norm_values[idx_denom][0], round_digit)}
+            else:
+                res_dict[key_tuple].update({tuple_i[-1]: round(values_all[idx][0]/norm_values[idx_denom][0], round_digit)})
+    else: # for more than 2 sensitive atts
+        pass
+    return res_dict
+
+def compute_evaluation_metric_binary(true_y, pred_y, label_order):
+    TN, FP, FN, TP = confusion_matrix(true_y, list(pred_y), labels=label_order).ravel()
+    P = TP + FN
+    N = TN + FP
+    ACC = (TP+TN) / (P+N) if (P+N) > 0.0 else np.float64(0.0)
+    return dict(
+                PR = P/ (P+N), P = TP + FN, N = TN + FP,
+                TPR=TP / P, TNR=TN / N, FPR=FP / N, FNR=FN / P,
+                PPV=TP / (TP+FP) if (TP+FP) > 0.0 else np.float64(0.0),
+                NPV=TN / (TN+FN) if (TN+FN) > 0.0 else np.float64(0.0),
+                FDR=FP / (FP+TP) if (FP+TP) > 0.0 else np.float64(0.0),
+                FOR=FN / (FN+TN) if (FN+TN) > 0.0 else np.float64(0.0),
+                ACC=ACC,
+                ERR=1-ACC,
+                F1=2*TP / (2*TP+FP+FN) if (2*TP+FP+FN) > 0.0 else np.float64(0.0)
+            )
+def get_performance_label(df, sensi_atts, target_name, posi_target, output_metrics=["TPR", "FPR", "TNR", "FNR", "PR"], round_digit=3):
+    
+    groupby_cols = sensi_atts+[target_name]
+    placeholder_att = list(set(df.columns).difference(groupby_cols))[0]
+    
+    count_all = df[groupby_cols+[placeholder_att]].groupby(groupby_cols).count()
+    index_all = list(count_all.index)
+
+    res_dict = {}
+    target_label_order = [posi_target, set(df[target_name]).difference([posi_target]).pop()]
+    
+    for tuple_i in index_all:
+        if len(tuple_i[:-1]) == 1:
+            key_tuple = (tuple_i[0])
+        else:
+            key_tuple = tuple_i[:-1]
+        cur_q = []
+        for idx, vi in enumerate(tuple_i[:-1]):
+            cur_q.append("{}=='{}'".format(sensi_atts[idx], vi))
+        tuple_df = df.query(" and ".join(cur_q))
+        metrics_all = compute_evaluation_metric_binary(list(tuple_df[target_name]), list(tuple_df["pred_"+target_name]), target_label_order)
+        res_dict[key_tuple] = {x: round(metrics_all[x], round_digit) for x in metrics_all if x in output_metrics}
+        
+    return res_dict
\ No newline at end of file
diff --git a/pipeline/label_pipeline.py b/pipeline/label_pipeline.py
new file mode 100644
index 0000000..5cc8321
--- /dev/null
+++ b/pipeline/label_pipeline.py
@@ -0,0 +1,506 @@
+"""
+    Class to run the pipeline
+
+"""
+import pandas as pd
+import numpy as np
+import os
+import warnings
+warnings.filterwarnings("ignore")
+from pipeline.preprocess.splitters import *
+from pipeline.preprocess.samplers import *
+from pipeline.preprocess.imputers import *
+from pipeline.preprocess.scalers import *
+from pipeline.preprocess.categorizers import *
+from pipeline.preprocess.encoders import *
+from pipeline.preprocess.fair_preprocessors import *
+from pipeline.model.classifiers import *
+from pipeline.model.fair_classifiers import *
+from pipeline.postprocess.fair_postprocessors import *
+
+# for integrity check of user inputs
+SUPPORT_STEPS = {"Splitter": "Splitter", "Sampler": "Sampler", "Imputer": "Imputer", "Scaler": "Scaler",
+                 "Categorizer": "Categorizer", "Encoder": "Encoder", "SensitiveEncoder": "SensitiveAttEncoder",
+                 "FairPreprocessor": "AIF_", "model": "SK_OPT_", "FairPostprocessor": "AIF_Postprocessing"}
+
+ALL_STEPS = ["RandomSplitter", "BalanceTargetSplitter",
+             "RandomSampler", "BalancePopulationSampler",
+             "DropNAImputer", "ModeImputer", "DatawigImputer",
+             "SK_StandardScaler", "SK_MinMaxScaler",
+             "SK_Discretizer", "SK_Binarizer",
+             "SK_OrdinalEncoder", "SK_OneHotEncoder",
+             "CustomCateAttsEncoder",
+             "AIF_Reweighing", "AIF_DIRemover",
+             "SK_LogisticRegression", "SK_DecisionTree", "OPT_LogisticRegression", "OPT_DecisionTree", "AIF_AdversarialDebiasing",
+             "AIF_EqOddsPostprocessing", "AIF_CalibratedEqOddsPostprocessing"]
+PRINT_SPLIT = "\n===================================================\n"
+
+def init_input_steps(step_tuple, input_df):
+    step_str = step_tuple[0] + '(input_df, '
+    for pi in step_tuple[1:]:
+        if isinstance(pi, str):
+            step_str += "'"+pi + "', "
+        else:
+            step_str += str(pi) + ", "
+    end_idx = step_str.rfind(", ")
+    step_str = step_str[0:end_idx] + step_str[end_idx:].replace(", ", ")")
+    print(step_str)
+    return eval(step_str)
+
+
+class FairnessLabelPipeline():
+    def __init__(self, data_file_name, target_col, target_positive_values, sensitive_atts=[], protected_values={}, sep_flag=None, na_mark=None):
+        """
+
+        :param data_file_name: str, file name that stores the data.
+        :param target_col: str, the name of the target variable in above data.
+        :param target_positive_values: list of str, each str is the value of the target variable in above data that represents the positive outcome.
+        :param sensitive_atts: list, stores the user specified sensitive attributes in above data. Optional.
+        :param protected_values: dict, stores the user-specified protected values for above sensitive attributes. Optional.
+                                Key is the name in sensitive_atts, value is a list of str, representing the values of the attribute.
+                                 Order should mapping to the order in the sensitive_atts.
+        """
+        # print(os.path.realpath(data_file_name))
+        if not os.path.exists(os.path.realpath(data_file_name)):
+            print("The data you specified doesn't exist!")
+            raise ValueError
+        if ".csv" not in data_file_name:
+            print("The data you specified is not valid! Only support .csv file.")
+            raise ValueError
+        data_name = data_file_name.replace(".csv", "")
+        self.data_name = data_name[data_name.rfind("/")+1:]
+        if sep_flag:
+            df = pd.read_csv(data_file_name, sep=sep_flag)
+        else: # default ',' separated data
+            df = pd.read_csv(data_file_name)
+        if not df.shape[0]:
+            print("Uploaded data is empty!")
+            raise ValueError
+
+        # integrity check for target_col
+        if target_col is None or target_positive_values is None:
+            print("Need to specify target_col and target_positive_value!")
+            raise ValueError
+
+        if target_col not in df.columns:
+            print("Need to specify a valid target attribute to be predicted!")
+            raise ValueError
+        target_values = df[target_col].unique()
+        if len(target_values) != 2:
+            print("Only support binary target feature now!")
+            raise ValueError
+        if len(set(target_positive_values).intersection(target_values)) == 0:
+            print("Need to specify a valid target positive value!")
+            raise ValueError
+        self.target_col = target_col
+        self.target_positive_values = target_positive_values
+
+        # integrity check for sensitive_atts and protected_values
+        input_indicator = sum([len(x)== 0 for x in [sensitive_atts, protected_values]])
+        if input_indicator == 0: # both are specified
+            if len(sensitive_atts) != len(protected_values):
+                print("Different size of input sensitive attributes and protected values!")
+                raise ValueError
+            if sum([len(set(protected_values[x]).difference(df[x].unique())) > 0 for x in protected_values]) > 0:
+                print("Some specified protected values do not appear in the column specified in sensitive_atts!")
+                raise ValueError
+        elif input_indicator == 1: # one of parameter is empty
+            print("Need to specify both sensitive_atts and protected_values!")
+            raise ValueError
+        else: # both are empty
+            # TODO: add auto-generation for the below two variables: sensitive_atts and protected_values
+            # for adult only
+            sensitive_atts = ["sex", "race"]
+            protected_values = {"sex": ["Female"], "race": ["Black", "Asian-Pac-Islander", "Amer-Indian-Eskimo", "Other"]}
+
+        self.sensitive_atts = sensitive_atts
+        # self.protected_values = protected_values
+        self.zero_mapping = {target_col: [x for x in target_values if x not in target_positive_values]}
+        self.zero_mapping.update(protected_values)
+        if na_mark:
+            self.na_mark = na_mark
+        else:
+            self.na_mark = None
+
+        # refer numerical and categorical attributes first
+        # DataFrame.describe() usually returns 8 rows.
+        if df.describe().shape[0] == 8:
+            num_atts = set(df.describe().columns)
+        # DataFrame.describe() returns less than 8 rows when there is no numerical attribute.
+        else:
+            num_atts = set()
+
+        cate_atts = set(df.columns).difference(num_atts)
+        if self.target_col in cate_atts:
+            cate_atts.remove(self.target_col)
+        if self.target_col in num_atts:
+            num_atts.remove(self.target_col)
+        for si in sensitive_atts:
+            cate_atts.remove(si)
+        self.num_atts = list(num_atts)
+        self.cate_atts = list(cate_atts)
+        self.df = df
+
+        # record the sensitive attributes and target variable value mapping
+        sensi_target_value_mapping = {}
+        for atti in [self.target_col] + self.sensitive_atts:
+            atti_values = list(df[atti].unique())
+            pro_values = self.zero_mapping[atti]
+            if len(pro_values) > 1:
+                pro_value_str = pro_values[0] + "&more"
+            else:
+                pro_value_str = pro_values[0]
+            other_values = list(set(atti_values).difference(pro_values))
+            if len(other_values) > 1:
+                other_value_str = other_values[0] + "&more"
+            else:
+                other_value_str = other_values[0]
+            sensi_target_value_mapping[atti] = {0: pro_value_str, 1: other_value_str}
+        # print(sensi_target_value_mapping)
+        self.sensi_target_value_mapping = sensi_target_value_mapping
+        self.pipeline_id = self.data_name[:2]
+
+    def init_necessary_steps(self, step_flag, apply_df, input_weights=[]):
+        # initialize the necessary steps
+        if step_flag == "Imputer":
+            return DropNAImputer(apply_df, na_mark=self.na_mark)
+        # elif step_flag == "Scaler":
+        #     return SK_StandardScaler(apply_df, list(self.num_atts))
+        elif step_flag == "Encoder":
+            return SK_OneHotEncoder(apply_df, list(self.cate_atts))
+        elif step_flag == "SensitiveEncoder":
+            return CustomCateAttsEncoder(apply_df, self.sensitive_atts+[self.target_col], self.zero_mapping)
+        else:
+            return OPT_LogisticRegression(apply_df, self.target_col, instance_weights=input_weights)
+
+    def print_necessary_steps(self):
+        # for printout and efficiency
+        return {"Imputer": ("DropNAImputer", "?"),
+                # "Scaler": ("SK_StandardScaler", list(self.num_atts)),
+                "Encoder": ("SK_OneHotEncoder", list(self.cate_atts)),
+                "SensitiveEncoder": ("CustomCateAttsEncoder", self.sensitive_atts+[self.target_col], self.zero_mapping),
+                "model": ("OPT_LogisticRegression", self.target_col)}
+
+
+    def run_pipeline(self, steps, return_test=True, output_interdata=False):
+        """
+
+        :param df: pandas dataframe, the data on which steps are applied.
+        :param steps: list of classes that represent the steps user want to perform on the above data.
+                      Supported steps are listed in STEPS.md.
+        :return:  two pandas dataframes: before_data and after_data. after_data is the data after applied the input steps.
+
+        """
+
+        if not steps:
+            print("Require list of steps as input!")
+            raise ValueError
+
+        if len(steps) < len(SUPPORT_STEPS):
+            print("Missing some input steps! Required steps are listed in the order below.\n"+" ".join(SUPPORT_STEPS.keys()))
+            raise ValueError
+        if sum([len(set(x[0]).intersection(list(SUPPORT_STEPS.values())[idx]))==0 for idx, x in enumerate(steps) if x is not None]) > 0:
+            print("Some input steps are not supported!")
+            raise ValueError
+        if sum([x[0] not in ALL_STEPS for x in steps if x is not None]) > 0:
+            print("Some input steps don't include enough parameters!")
+            raise ValueError
+        if sum([len(x) <=1 for x in steps if x is not None]) > 0:
+            print("Some input steps don't include enough parameters!")
+            raise ValueError
+
+        if steps[-1] is not None and len(steps[0][1])==2: # run fair-postprocessor, then require validation set
+            print("FairPostprocessor requires a validation set! Specify through split_ratio in Splitter!")
+            raise ValueError
+
+        self.pipeline_id = "_".join([self.pipeline_id]+[str(x[0]) for x in steps if x is not None])
+
+        support_step_names = list(SUPPORT_STEPS.keys())
+
+        # split the data into separated datasets for train, [validation], and test
+        if steps[0] is None: # default splitter
+            if steps[-1] is not None: # train, validation and test data
+                cur_splitter = BalanceTargetSplitter(self.df, [0.5, 0.3, 0.2], self.target_col)
+            else: # train and test data
+                cur_splitter = BalanceTargetSplitter(self.df, [0.7, 0.3], self.target_col)
+        else:
+            cur_splitter = init_input_steps(steps[0], self.df)
+        after_data = cur_splitter.apply(self.df)
+        after_data = list(after_data)
+        if output_interdata:
+            self.save_inter_data(after_data, cur_splitter.get_name())
+        print("Done "+support_step_names[0]+PRINT_SPLIT)
+        # record the before data to output
+        before_data = [x for x in after_data]
+
+
+        # run sampler on train
+        if steps[1] is not None:
+            for idx_df, cur_df in enumerate(after_data):
+                cur_sampler = init_input_steps(steps[1], cur_df)
+                after_data[idx_df] = cur_sampler.apply(cur_df)
+            if output_interdata:
+                self.save_inter_data(after_data, cur_sampler.get_name(), steps[:1])
+            print("Done "+support_step_names[1]+PRINT_SPLIT)
+
+
+        # run the preprocess steps: "Imputer", "Scaler", "Categorizer" that fit on train and apply on others
+        for idx, step_i in enumerate(steps[2:5]):
+            idx = idx + 2
+            step_i_key = support_step_names[idx]
+            # fit on train data
+            if step_i is None:
+                if step_i_key in list(self.print_necessary_steps().keys()):  # add default operation for necessary steps
+                    step_i = self.init_necessary_steps(step_i_key, after_data[0])
+                else:  # skip the step
+                    continue
+            else:  # user-specified step
+                step_i = init_input_steps(steps[idx], after_data[0])
+            # apply on train, validation and test data
+            for idx_df, cur_df in enumerate(after_data):
+                after_data[idx_df] = step_i.apply(cur_df)
+
+            if output_interdata:
+                self.save_inter_data(after_data, step_i.get_name(), steps[:idx])
+            print("Done " + support_step_names[idx] + PRINT_SPLIT)
+
+        # run the preprocess steps: "Encoder"
+        # fit and apply on the same data
+        after_data, encoder_name = self.run_encoder(steps[5], after_data)
+        if output_interdata:
+            self.save_inter_data(after_data, encoder_name, steps[:5])
+        print("Done " + support_step_names[5] + PRINT_SPLIT)
+
+        # run the preprocess steps: "SensitiveAttEncoder"
+        # fit and apply on the same data
+        if steps[6] is None:
+            for idx_df, cur_df in enumerate(after_data):
+                cur_sensi_encoder = self.init_necessary_steps("SensitiveEncoder", cur_df)
+                after_data[idx_df] = cur_sensi_encoder.apply(cur_df)
+        else: # user-specified sensitive encoder
+            for idx_df, cur_df in enumerate(after_data):
+                cur_sensi_encoder = init_input_steps(steps[6], cur_df)
+                after_data[idx_df] = cur_sensi_encoder.apply(cur_df)
+        if output_interdata:
+            self.save_inter_data(after_data, cur_sensi_encoder.get_name(), steps[:6])
+        print("Done " + support_step_names[6] + PRINT_SPLIT)
+
+        # run the preprocess steps: "FairPreprocessor"
+        # fit and apply on the same data
+        if steps[7] is not None:
+            weights = [[0 for _ in range(x.shape[0])] for x in after_data]
+            for idx_df, cur_df in enumerate(after_data):
+                cur_fair_preprossor = init_input_steps(steps[7], cur_df)
+                if "AIF_Reweighing" in cur_fair_preprossor.get_name():  # special heck for methods updating sample weight
+                    after_data[idx_df], weights[idx_df] = cur_fair_preprossor.apply(cur_df)
+                else:
+                    after_data[idx_df] = cur_fair_preprossor.apply(cur_df)
+            if output_interdata:
+                self.save_inter_data(after_data, cur_fair_preprossor.get_name(), steps[:7])
+            print("Done " + support_step_names[7] + PRINT_SPLIT)
+
+            # after fair-preprocess, rerun encoder
+            after_data, encoder_name = self.run_encoder(steps[5], after_data)
+            if output_interdata:
+                self.save_inter_data(after_data, encoder_name+"_prep", steps[:5])
+            print("Done " + support_step_names[5] + " for fair preprocessor "+PRINT_SPLIT)
+
+        # run model step
+        # fit on train data
+        if steps[8] is None:
+            if weights:
+                cur_model = self.init_necessary_steps("model", after_data[0], input_weights=weights[0])
+            else:
+                cur_model = self.init_necessary_steps("model", after_data[0])
+        else: # TODO: add the support for weight in user-specified models
+            cur_model = init_input_steps(steps[8], after_data[0])
+
+        # predict on train, validation and test data
+        for idx_df, cur_df in enumerate(after_data):
+            after_data[idx_df] = cur_model.apply(cur_df)
+        if output_interdata:
+            self.save_inter_data(after_data, cur_model.get_name(), steps[:8])
+        print("Done " + support_step_names[8] + PRINT_SPLIT)
+
+        # run fair postprocess step
+        if steps[9] is not None:
+            # encode first
+            after_data, encoder_name = self.run_encoder(steps[5], after_data)
+            if output_interdata:
+                self.save_inter_data(after_data, encoder_name+"_post", steps[:5])
+            print("Done " + support_step_names[5] + " for fair post processor " + PRINT_SPLIT)
+            # fit on validation data
+            cur_postprocessor = init_input_steps(steps[9], after_data[1])
+            # predict on validation and test data
+            for idx_df, cur_df in enumerate(after_data[1:]):
+                after_data[idx_df+1] = cur_postprocessor.apply(cur_df)
+            if output_interdata:
+                self.save_inter_data(after_data, cur_postprocessor.get_name(), steps[:9])
+            print("Done " + support_step_names[9] + PRINT_SPLIT)
+
+        # transfer back to original values for encoded sensitive and target columns
+        for idx, df_i in enumerate(after_data):
+            for atti in [self.target_col]+ self.sensitive_atts:
+                df_i[atti] = df_i[atti].apply(lambda x: self.sensi_target_value_mapping[atti][x])
+            if "pred_" +self.target_col in df_i.columns:
+                df_i["pred_" + self.target_col] = df_i["pred_" + self.target_col].apply(lambda x: int(x>=0.5))
+                df_i["pred_" +self.target_col] = df_i["pred_" +self.target_col].apply(lambda x: self.sensi_target_value_mapping[self.target_col][x])
+
+        if return_test: # only return the before and after of test data
+            return before_data[-1], after_data[-1]
+        else: # return all before and after data
+            return before_data, after_data
+
+    def run_encoder(self, encode_step_tuple, data_list):
+        # run the preprocess steps: "Encoder"
+        # fit and apply on the same data
+        if len(self.cate_atts) > 0:
+            for idx_df, cur_df in enumerate(data_list):
+                if encode_step_tuple is None:  # default encoder
+                    cur_encoder = self.init_necessary_steps(list(SUPPORT_STEPS.keys())[5], cur_df)
+                else:
+
+                    # check for user specified encoder that cover partial categorical atts
+                    non_encoded_cate = set(self.cate_atts).difference(encode_step_tuple[1])
+                    if non_encoded_cate:
+                        cur_encoder = init_input_steps((encode_step_tuple[0], self.cate_atts), cur_df)
+                    else:
+                        cur_encoder = init_input_steps(encode_step_tuple, cur_df)
+                data_list[idx_df] = cur_encoder.apply(cur_df)
+
+            # check for different dimensions after encoding for validation and test set
+            if len(data_list) > 2:
+                for idx_df, cur_df in enumerate(data_list[1:]):
+                    if cur_df.shape[1] != data_list[0].shape[1]:
+                        for feature_i in set(data_list[0].columns).difference(cur_df.columns):
+                            cur_df[feature_i] = 0.0
+                        data_list[idx_df] = cur_df.copy()
+            else: # check the dimensions for train and test set
+                if data_list[0].shape[1] != data_list[1].shape[1]:
+                    diff_features_1 = set(data_list[0].columns).difference(data_list[1].columns)
+                    diff_features_2 = set(data_list[1].columns).difference(data_list[0].columns)
+                    add_df = data_list[1].copy()
+                    for feature_i in diff_features_1.union(diff_features_2):
+                        if feature_i not in add_df.columns:
+                            add_df[feature_i] = 0.0
+                    data_list[1] = add_df.copy()
+
+                    add_df = data_list[0].copy()
+                    for feature_i in diff_features_1.union(diff_features_2):
+                        if feature_i not in add_df.columns:
+                            add_df[feature_i] = 0.0
+                    data_list[0] = add_df.copy()
+
+            return data_list, cur_encoder.get_name()
+        else:
+            return data_list, "None"
+
+    def save_inter_data(self, input_dfs, step_name, pre_steps=[], path="data/inter_data/"):
+        if len(input_dfs) == 2:
+            suffix_names = ["train", "test"]
+        else:
+            suffix_names = ["train", "validation", "test"]
+
+        for idx, df_i in enumerate(input_dfs):
+            output_df_i = df_i.copy()
+            df_path = os.path.realpath(path) + "/" + self.pipeline_id + "/" + suffix_names[idx] + "/"
+            if not os.path.exists(df_path):
+                os.makedirs(df_path)
+            if pre_steps:
+                pre_step_names = [x[0] for x in pre_steps if x is not None]+[step_name]
+            else:
+                pre_step_names = [step_name]
+
+            for atti in [self.target_col]+ self.sensitive_atts:
+                if not isinstance(output_df_i[atti].values[0], str):
+                    output_df_i[atti] = output_df_i[atti].apply(lambda x: self.sensi_target_value_mapping[atti][x])
+            if "pred_" +self.target_col in output_df_i.columns:
+                output_df_i["pred_" + self.target_col] = output_df_i["pred_" + self.target_col].apply(lambda x: int(x >= 0.5))
+                output_df_i["pred_" +self.target_col] = output_df_i["pred_" +self.target_col].apply(lambda x: self.sensi_target_value_mapping[self.target_col][x])
+
+            output_name = df_path + "__".join([self.data_name, "after"]+[x[:x.find("@")] for x in pre_step_names]) + ".csv"
+            print("!!!!!!!", suffix_names[idx], output_df_i.shape, "!!!!!!!")
+            output_df_i.to_csv(output_name, index=False)
+            print("Current "+suffix_names[idx]+" data after "+" ".join([x[:x.find("@")] for x in pre_step_names])+" \n Stored in ", output_name)
+            print()
+
+
+    
+
+if __name__ == '__main__':
+    # input_steps = [("BalanceTargetSplitter", [0.5, 0.3, 0.2], "income-per-year"),
+    #                ("RandomSampler", 10000),  # sampler
+    #                ("DropNAImputer", "?"),
+    #                ("SK_StandardScaler", ["fnlwgt", "age"]),
+    #                ("SK_Discretizer", ["fnlwgt", "age"], [2, 3]),
+    #                ("SK_OneHotEncoder", ["workclass"]),  # encoder
+    #                ("CustomCateAttsEncoder", ["sex", "race", "income-per-year"], {"sex": ["Female"], "race": ["Black"], "income-per-year": ["<=50K"]}),
+    #                ("AIF_DIRemover", "income-per-year", "sex", 0.8),  # fair-preprocessor
+    #                ("AIF_AdversarialDebiasing", "income-per-year", "sex"), # test Adversial learning
+    #                ("AIF_CalibratedEqOddsPostprocessing", "income-per-year", "sex")  # fair-post-postprocessor
+    #                ]
+    # cur_pip = FairnessLabelPipeline(data_file, y_col, y_posi, sensitive_atts=sensi_atts, protected_values=sensi_pro_valus, na_mark="?")
+    # before_test, after_test = cur_pip.run_pipeline(input_steps, return_test=True, output_interdata=True)
+
+    # data_file = "../data/adult.csv"
+    # y_col = "income-per-year"
+    # y_posi = [">50K"]
+    # na_symbol = "?"
+    # sensi_atts = ["sex", "race"]
+    # sensi_pro_valus = {"sex": ["Female"], "race": ["Black", "Asian-Pac-Islander", "Amer-Indian-Eskimo", "Other"]}
+
+    # data_file = "../data/mylsn_cleaned_2.csv"
+    # y_col = "status"
+    # y_posi = ["Ac"]
+    # na_symbol = "N/A"
+    # sensi_atts = ["sex", "race"]
+    # sensi_pro_valus = {"sex": ["female"], "race": ["black", "hispanic", "native-american", "asian"]}
+
+    #
+    #  input_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),  # splitter
+    #               None,  # sampler
+    #               ("DropNAImputer", "?"),  # imputer
+    #               None,  # scaler
+    #               ("SK_Discretizer", ["age"], [3]),  # categorizer
+    #               None,  # encoder
+    #               None,  # sensitive att and target encoder
+    #               None,  # fair-preprocessor
+    #               ("OPT_LogisticRegression", y_col),  # model
+    #               None  # fair-post-postprocessor
+    #               ]
+
+    # debias_focus_att = "race"
+    # input_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),
+    #               ("RandomSampler", 5000),  # sampler
+    #               None,  # imputer
+    #               None,  # scaler
+    #               None,
+    #               None,  # encoder
+    #               None,
+    #               None, #("AIF_DIRemover", y_col, debias_focus_att, 1.0),  # fair-preprocessor
+    #               ("SK_LogisticRegression", y_col),  # model
+    #               None  # fair-post-postprocessor
+    #               ]
+
+    data_file = "../data/german_AIF.csv"
+    y_col = "credit"
+    y_posi = ["good"]
+    sensi_atts = ["age", "sex"]
+    sensi_pro_valus = {"age": ["young"], "sex": ["female"]}
+    debias_focus_att = "age"
+    fair_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),
+                  None,  # sampler
+                  None,  # ("ModeImputer", [], ["workclass"], "?"), # imputer
+                  None,  # scaler
+                  None,  # categorizer
+                  None,  # encoder
+                  None,
+                  ("AIF_Reweighing", y_col, debias_focus_att),  # fair-preprocessor
+                  None,  # ("OPT_LogisticRegression", y_col), # model
+                  None  # fair-post-postprocessor
+                  ]
+
+    cur_pip = FairnessLabelPipeline(data_file, y_col, y_posi, sensitive_atts=sensi_atts,
+                                    protected_values=sensi_pro_valus)
+    before_test, after_test = cur_pip.run_pipeline(fair_steps, return_test=True, output_interdata=True)
+
diff --git a/pipeline/model/classifiers.py b/pipeline/model/classifiers.py
new file mode 100644
index 0000000..248311f
--- /dev/null
+++ b/pipeline/model/classifiers.py
@@ -0,0 +1,87 @@
+"""
+    Classes of supervised binary classifiers.
+"""
+
+import pandas as pd
+from sklearn.linear_model import SGDClassifier
+from sklearn.tree import DecisionTreeClassifier
+from pipeline.model.inprocessor import Model
+
+
+class SK_LogisticRegression(Model):
+    def __init__(self, df, target_col, loss_func="log", instance_weights=[], seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param loss_func: str, the name of the loss function used in linear model. Same as the loss parameter in sklearn.linear_model.SGDClassifier.
+                         The possible options are ‘hinge’, ‘log’, ‘modified_huber’, ‘squared_hinge’, ‘perceptron’, or a regression loss: ‘squared_loss’, ‘huber’, ‘epsilon_insensitive’, or ‘squared_epsilon_insensitive’.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param seed: integer, the seed for random state.
+        """
+
+        cur_step = SGDClassifier(loss=loss_func, random_state=seed)
+        super().__init__("@".join(["SK_LogisticRegression", target_col]), cur_step, df, target_col, instance_weights=instance_weights)
+
+
+class SK_DecisionTree(Model):
+    def __init__(self, df, target_col, instance_weights=[], seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param seed: integer, the seed for random state.
+        """
+        cur_step = DecisionTreeClassifier(random_state=seed)
+        super().__init__("@".join(["SK_DecisionTree", target_col]), cur_step, df, target_col, instance_weights=instance_weights)
+
+
+class OPT_LogisticRegression(Model):
+    def __init__(self, df, target_col, loss_func="log", max_iter=1000, instance_weights=[], seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param loss_func: str, the name of the loss function used in linear model. Same as the loss parameter in sklearn.linear_model.SGDClassifier.
+                         The possible options are ‘hinge’, ‘log’, ‘modified_huber’, ‘squared_hinge’, ‘perceptron’, or a regression loss: ‘squared_loss’, ‘huber’, ‘epsilon_insensitive’, or ‘squared_epsilon_insensitive’.
+        :param max_iter: integer, max number of iterations of the model.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param seed: integer, random seed.
+        """
+        # Update below parameters according to the loss function used
+        param_grid = {
+            'learner__loss': [loss_func],
+            'learner__penalty': ['l2', 'l1', 'elasticnet'],
+            'learner__alpha': [0.00005, 0.0001, 0.005, 0.001]
+        }
+        cur_step = SGDClassifier(max_iter=max_iter, random_state=seed)
+        super().__init__("@".join(["OPT_LogisticRegression", target_col]), cur_step, df, target_col, instance_weights=instance_weights, hyper_tune=True, param_grid=param_grid)
+
+class OPT_DecisionTree(Model):
+    def __init__(self, df, target_col, instance_weights=[], seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param seed: integer, random seed.
+        """
+        param_grid = {
+            'learner__min_samples_split': range(20, 500, 10),
+            'learner__max_depth': range(15, 30, 2),
+            'learner__min_samples_leaf': [3, 4, 5, 10],
+            "learner__criterion": ["gini", "entropy"]
+        }
+
+        cur_step = DecisionTreeClassifier(random_state=seed)
+        super().__init__("@".join(["OPT_DecisionTree", target_col]), cur_step, df, target_col, instance_weights=instance_weights, hyper_tune=True, param_grid=param_grid)
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
+    cur_o = SK_LogisticRegression(data, "income-per-year")
+    # cur_o = SK_DecisionTree(data, "income-per-year")
+    # cur_o = OPT_LogisticRegression(data, "income-per-year")
+    # cur_o = OPT_DecisionTree(data, "income-per-year")
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/model/fair_classifiers.py b/pipeline/model/fair_classifiers.py
new file mode 100644
index 0000000..085e968
--- /dev/null
+++ b/pipeline/model/fair_classifiers.py
@@ -0,0 +1,69 @@
+"""
+    Classes of fair supervised binary classifiers.
+"""
+
+import pandas as pd
+from aif360.algorithms.inprocessing import AdversarialDebiasing
+from aif360.algorithms.inprocessing import MetaFairClassifier
+from aif360.algorithms.inprocessing import PrejudiceRemover
+from pipeline.model.inprocessor import Model
+import warnings
+warnings.filterwarnings("ignore")
+
+class AIF_AdversarialDebiasing(Model):
+
+    def __init__(self, df, target_col, sensitive_att, seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the fair classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param seed: integer, random seed.
+
+        """
+
+        import tensorflow as tf
+        sess = tf.Session()
+        cur_step = AdversarialDebiasing(unprivileged_groups=[{sensitive_att: 0}], privileged_groups=[{sensitive_att: 1}], scope_name='debiased_classifier', debias=True, sess=sess, seed=seed)
+        super().__init__("@".join(["AIF_AdversarialDebiasing", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+
+
+class AIF_MetaFairClassifier(Model):
+
+    def __init__(self, df, target_col, sensitive_att, fairness_penalty=0.8, fair_metric="sr"):
+        """
+        :param df: pandas dataframe, stores the data to fit the fair classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param fairness_penalty: float in [0,1], fairness penalty parameter. default is 0.8. The same parameter in aif360.algorithms.inprocessing.MetaFairClassifier.
+        :param fair_metric: str, fairness metric used in this method. Value from ["fdr" (false discovery rate ratio), "sr" (statistical rate/disparate impact)].
+                            The same parameter in aif360.algorithms.inprocessing.MetaFairClassifier.
+        """
+
+        cur_step = MetaFairClassifier(tau=fairness_penalty, sensitive_attr=sensitive_att, type=fair_metric)
+        super().__init__("@".join(["AIF_MetaFairClassifier", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+
+class AIF_PrejudiceRemover(Model):
+
+    def __init__(self, df, target_col, sensitive_att, fairness_penalty=1.0):
+        """
+        :param df: pandas dataframe, stores the data to fit the fair classifier.
+        :param target_col: str, the name of the target variable in above data.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param fairness_penalty: float in [0,1], fairness penalty parameter. default is 1. The same parameter in aif360.algorithms.inprocessing.PrejudiceRemover.
+
+        """
+        # TODO: fix the bug that cannot import lib of 'getoutput'
+        cur_step = PrejudiceRemover(eta=fairness_penalty, sensitive_attr=sensitive_att, class_attr=target_col)
+        super().__init__("@".join(["AIF_PrejudiceRemover", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
+    cur_o = AIF_AdversarialDebiasing(data, "income-per-year", "sex")
+    # cur_o = AIF_MetaFairClassifier(data, "income-per-year", "sex")
+    # cur_o = AIF_PrejudiceRemover(data, "income-per-year", "sex")
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/model/inprocessor.py b/pipeline/model/inprocessor.py
new file mode 100644
index 0000000..3df3909
--- /dev/null
+++ b/pipeline/model/inprocessor.py
@@ -0,0 +1,74 @@
+"""
+    Super class for all the supported classifier classes including fair-classifiers.
+"""
+import numpy as np
+from aif360.datasets import BinaryLabelDataset
+from sklearn.model_selection import GridSearchCV
+from sklearn.pipeline import Pipeline
+from pipeline.step import Step
+
+class Model(Step):
+    def __init__(self, step_name, step, df, target_col, instance_weights=[], hyper_tune=False, param_grid={}, sensitive_att=None, fair_aware=False, target_positive=1):
+        """
+        :param step_name: str, name of the current input step.
+        :param step: object of the initialized class.
+        :param df: pandas dataframe, stores the data.
+        :param target_col: str, the name of the target attribute.
+        :param instance_weights: list of float in [0,1], each float represents the weight of the sample in above data.
+        :param hyper_tune: boolean, whether to tune the hyper-parameter. Default is False.
+        :param param_grid: dict, stores the search range of the hyper-parameter. When hyper_tune is True, this must be provided.
+        :param sensitive_att: str, the name of a sensitive attribute.
+        :param fair_aware: boolean, whether the model is fair-aware. Default is False.
+        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
+        """
+
+        super().__init__(step_name=step_name, df=df, sensitive_att=sensitive_att, target_col=target_col)
+        # assume the data set has been encoded to numerical values
+        if fair_aware: # fair classifiers
+            # intitialize a binary label dataset from AIF 360
+            aif_df = BinaryLabelDataset(df=df, label_names=[target_col], protected_attribute_names=[sensitive_att])
+            fitted_step = step.fit(aif_df)
+            input_score = False
+        else: # regular classifiers
+            if len(instance_weights) == 0:
+                instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
+            if hyper_tune: # grid search for best hyper parameters
+                if not param_grid:
+                    print("Need to specify the search range of the hyper parameters - 'param_grid' is empty!")
+                    raise ValueError
+
+                search = GridSearchCV(Pipeline([('learner', step)]), param_grid, scoring='roc_auc', cv=5, verbose=1, n_jobs=-1)
+                fitted_step = search.fit(np.array(df.drop(columns=[target_col])), np.array(df[target_col]), None, **{'learner__sample_weight': instance_weights})
+            else:
+                fitted_step = step.fit(np.array(df.drop(columns=[target_col])), np.array(df[target_col]), sample_weight=instance_weights)
+            input_score = True
+
+        self.input_score = input_score
+        self.step = fitted_step
+        self.target_positive = target_positive
+
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned discretizer.
+        :return: pandas dataframe, stores the data after discretize.
+        """
+
+        # initialize AIF360 BinaryLabelDataset
+
+        if self.input_score:  # for regular model, generate score prediction
+            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
+            after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+
+            favorable_class_idx = list(self.step.classes_).index(self.target_positive)
+            after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
+
+        else:  # for fair model, generate label prediction
+            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
+                                             protected_attribute_names=[self.sensitive_att])
+
+            after_aif_df = self.step.predict(aif_pred_df)
+            after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+            after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
diff --git a/pipeline/postprocess/fair_postprocessors.py b/pipeline/postprocess/fair_postprocessors.py
new file mode 100644
index 0000000..efa5b1c
--- /dev/null
+++ b/pipeline/postprocess/fair_postprocessors.py
@@ -0,0 +1,66 @@
+"""
+    Classes for post-process data and model outcomes
+"""
+
+import pandas as pd
+from aif360.algorithms.postprocessing import CalibratedEqOddsPostprocessing
+from aif360.algorithms.postprocessing import EqOddsPostprocessing
+from aif360.algorithms.postprocessing import RejectOptionClassification
+from pipeline.postprocess.postprocessor import Postprocessor
+
+class AIF_EqOddsPostprocessing(Postprocessor):
+
+    def __init__(self, df, target_col, sensitive_att, threshold=0.5, seed=0):
+        """
+        :param df: pandas dataframe, stores the data to fit the postprocessor.
+        :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
+        :param seed: integer, the seed for random state.
+        """
+
+        cur_step = EqOddsPostprocessing([{sensitive_att: 0}], [{sensitive_att: 1}], seed)
+        super().__init__("@".join(["AIF_EqOddsPostprocessing", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=False, clf_threshold=threshold)
+
+
+class AIF_CalibratedEqOddsPostprocessing(Postprocessor):
+
+    def __init__(self, df, target_col, sensitive_att, threshold=0.5, seed=0, cost_constraint='weighted'):
+        """
+        :param df: pandas dataframe, stores the data to fit the postprocessor.
+        :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
+        :param seed: integer, the seed for random state.
+        :param cost_constraint: str, the fairness constraints format, value from [fpr, fnr, weighted].
+                                The same parameter as in aif360.algorithms.postprocessing.CalibratedEqOddsPostprocessing.
+        """
+
+        cur_step = CalibratedEqOddsPostprocessing([{sensitive_att: 0}], [{sensitive_att: 1}], cost_constraint=cost_constraint, seed=seed)
+        super().__init__("@".join(["AIF_CalibratedEqOddsPostprocessing", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=True, clf_threshold=threshold)
+
+class AIF_RejectOptionPostprocessing(Postprocessor):
+
+    def __init__(self, df, target_col, sensitive_att, threshold=0.5):
+        """
+        :param df: pandas dataframe, stores the data to fit the postprocessor.
+        :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
+        """
+        # TODO: fix the bug that reject option doesn't return results
+        cur_step = RejectOptionClassification([{sensitive_att: 0}], [{sensitive_att: 1}])
+        super().__init__("@".join(["AIF_RejectOptionClassification", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=True, clf_threshold=threshold)
+
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_post.csv")
+    # cur_o = AIF_RejectOptionPostprocessing(data, "income-per-year", "sex")
+    cur_o = AIF_EqOddsPostprocessing(data, "income-per-year", "sex")
+    # cur_o = AIF_CalibratedEqOddsPostprocessing(data, "income-per-year", "sex")
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/postprocess/postprocessor.py b/pipeline/postprocess/postprocessor.py
new file mode 100644
index 0000000..f10494c
--- /dev/null
+++ b/pipeline/postprocess/postprocessor.py
@@ -0,0 +1,59 @@
+"""
+    Super class for all the supported postprocessor classes.
+"""
+import numpy as np
+from aif360.datasets import BinaryLabelDataset
+from pipeline.step import Step
+
+class Postprocessor(Step):
+    def __init__(self, step_name, step, df, sensitive_att, target_col, input_score=True, clf_threshold=0.5):
+        """
+        :param step_name: str, name of the current input step.
+        :param step: object of the initialized class.
+        :param df: pandas dataframe, stores the data.
+        :param sensitive_att: str, the name of a sensitive attribute.
+        :param target_col: str, the name of the target attribute.
+        :param input_score: boolean, represent whether the post-processor takes predicted score as input. Default is True.
+        :param clf_threshold: float in [0, 1], represents the threshold to categorize class labels from predicted scores.
+        """
+        if "pred_"+target_col not in df.columns:
+            print("Require the predictions for ",target_col, " existing in the data!")
+            raise ValueError
+        super().__init__(step_name=step_name, df=df, sensitive_att=sensitive_att, target_col=target_col)
+        # assume the data set has been encoded to numerical values,
+        # intitialize a BinaryLabelDataset from AIF 360
+        aif_true_df = BinaryLabelDataset(df=df.drop(columns=["pred_"+target_col]), label_names=[target_col], protected_attribute_names=[sensitive_att])
+
+        aif_pred_df = aif_true_df.copy()
+
+        if input_score:
+            aif_pred_df.scores = df["pred_"+target_col]
+        else:
+            aif_pred_df.labels = np.array([int(x >= clf_threshold) for x in df["pred_"+target_col]])
+        self.input_score = input_score
+        self.step = step.fit(aif_true_df, aif_pred_df)
+        self.clf_threshold = clf_threshold
+
+
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned discretizer.
+        :return: pandas dataframe, stores the data after discretize.
+        """
+
+        # initialize AIF360 BinaryLabelDataset
+
+        if self.input_score: # use score prediction to fit model, e.g. RejectOptionClassification, CalibratedEqOddsPostprocessing
+            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], scores_names=[self.pred_target_col],
+                                             protected_attribute_names=[self.sensitive_att])
+        else: # use label prediction to fit model, e.g. EqOddsPostprocessing
+            df["pred_label_"+self.target_col] = [int(x >= self.clf_threshold) for x in df[self.pred_target_col]]
+            aif_pred_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]), label_names=["pred_label_"+self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.step.predict(aif_pred_df)
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
\ No newline at end of file
diff --git a/pipeline/preprocess/categorizers.py b/pipeline/preprocess/categorizers.py
new file mode 100644
index 0000000..83d8d90
--- /dev/null
+++ b/pipeline/preprocess/categorizers.py
@@ -0,0 +1,51 @@
+"""
+    Classes to discretize numerical attributes into categorical attributes.
+"""
+import pandas as pd
+from sklearn.preprocessing import KBinsDiscretizer
+from sklearn.preprocessing import Binarizer
+from pipeline.preprocess.preprocessor import Preprocessor
+
+
+class SK_Discretizer(Preprocessor):
+    def __init__(self, df, num_atts, bin_size, encode='ordinal', strategy='kmeans'):
+        """
+        :param df: pandas dataframe, stores the data to fit the discretizer.
+        :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
+        :param bin_size: list of integer, each integer represents the number of bins to categorize the corresponding numerical attribute.
+        :param encode: same parameter with sklearn KBinsDiscretizer
+        :param strategy: same parameter with sklearn KBinsDiscretizer
+        """
+        cur_step = {}
+        for idx, ai in enumerate(num_atts):
+            cur_step[ai] = KBinsDiscretizer(n_bins=bin_size[idx], encode=encode, strategy=strategy)
+        self.bin_size = bin_size
+        super().__init__("@".join([strategy+"Categorizer"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+
+
+class SK_Binarizer(Preprocessor):
+    def __init__(self, df, num_atts, bin_thresholds, copy=True):
+        """
+        :param df: pandas dataframe, stores the data to fit the binarizer.
+        :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
+        :param bin_thresholds: list of float, each float represents the value to binarize the corresponding numerical attributes.
+                               Values below or equal to this threshold are replaced by 0, above it by 1.
+        :param copy: same parameter with sklearn Binarizer
+        """
+        cur_step = {}
+        for idx, ai in enumerate(num_atts):
+            cur_step[ai] = Binarizer(threshold=bin_thresholds[idx], copy=copy)
+
+        self.bin_thresholds = bin_thresholds
+        super().__init__("@".join(["BinaryCategorizer"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult.csv")
+    cur_o = SK_Discretizer(data, ["fnlwgt", "age"], [2, 3])
+    # cur_o = SK_Binarizer(data, ["fnlwgt", "age"], [100000, 30])
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/preprocess/encoders.py b/pipeline/preprocess/encoders.py
new file mode 100644
index 0000000..b2a6712
--- /dev/null
+++ b/pipeline/preprocess/encoders.py
@@ -0,0 +1,98 @@
+"""
+    Classes to encode the string values for categorical attributes.
+"""
+"""
+    Classes to discretize numerical attributes into categorical attributes.
+"""
+import numpy as np
+import pandas as pd
+from sklearn.preprocessing import OrdinalEncoder
+from sklearn.preprocessing import OneHotEncoder
+from pipeline.preprocess.preprocessor import Preprocessor
+
+class SK_OrdinalEncoder(Preprocessor):
+    def __init__(self, df, cate_atts, sort_label, sort_positive_value):
+        """
+        :param df: pandas dataframe, stores the data to fit the encoder.
+        :param cate_atts: list of str, each str represents the name of a categorical attribute in above data.
+        :param encode_order_dict: dict, key (str) represents the name of categorical attribute, value is a list of str, representing the ordered categories of each each categorical attribute.
+        :param sort_label: str, name of the target varible to determine the order of ordinal encodings.
+        """
+        cur_step = {}
+        for ci in cate_atts:
+            value_counts = {}
+            for vi in df[ci].unique():
+                value_counts[vi] = df[(df[ci] == vi) & (df[sort_label] == sort_positive_value)].shape[0]
+            value_orders = sorted(value_counts.keys(), key=lambda x: value_counts[x])
+            cur_step[ci] = OrdinalEncoder(categories=[value_orders])
+
+        self.sort_label = sort_label
+        self.sort_positive_value = sort_positive_value
+        super().__init__("@".join(["OrdinalEncoder"]+cate_atts), df, step=cur_step, focus_atts=cate_atts, fit_flag=True)
+
+
+class SK_OneHotEncoder(Preprocessor):
+    def __init__(self, df, cate_atts):
+        """
+        :param df: pandas dataframe, stores the data to fit the encoder.
+        :param cate_atts: list of str, each str represents the name of a categorical attribute in above data.
+        :param encode_order_dict: dict, key (str) represents the name of categorical attribute, value is a list of str, representing the ordered categories of each each categorical attribute.
+        :param sort_label: str, name of the target variable to determine the order of ordinal encodings.
+        """
+        # TODO: fix the bug that sklearn one-hot encoder change the dimension
+        # cur_step = {}
+        # for ci in cate_atts:
+        #     cur_step[ci] = OneHotEncoder()
+        super().__init__("@".join(["OneHotEncoder"]+cate_atts), df, step=None, focus_atts=cate_atts, fit_flag=False)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned encoder.
+        :return: pandas dataframe, stores the data after encode.
+        """
+        after_df = pd.get_dummies(df, columns=self.focus_atts, prefix_sep='=')
+        # after_df = df[list(set(df.columns).difference(self.focus_atts))]
+        # for ci in self.focus_atts:
+        #     ci_encode_array = self.step[ci].transform(np.array(df[ci]).reshape(-1, 1)).toarray()
+        #     ci_encode_df = pd.DataFrame(ci_encode_array, columns=[ci+"="+x for x in self.step[ci].categories_[0]])
+        #     after_df = pd.concat([after_df, ci_encode_df], axis=1)
+        return after_df
+
+class CustomCateAttsEncoder(Preprocessor):
+    def __init__(self, df, sensitive_atts, protected_values):
+        """ To encode sensitive attribute and target feature.
+        :param df: pandas dataframe, stores the data to fit the encoder.
+        :param sensitive_atts: list of str, each str represents the name of a sensitive attribute.
+        :param protected_values: dict, key is the str in sensitive_atts, value is a list of str, each str represent the protected values for the key sensitive attribute.
+        """
+        super().__init__("@".join(["SensitiveAttEncoder"]+sensitive_atts), df, step=None, focus_atts=sensitive_atts, fit_flag=False)
+        for x in sensitive_atts:
+            if sum([vi not in df[x].unique() for vi in protected_values[x]]) > 0:
+                print("Some input values of sensitive attribute ", x, " are not valid!")
+                raise ValueError
+        self.protected_values = protected_values
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the encoder.
+        :return: pandas dataframe, stores the data after encode.
+        """
+        after_df = df.copy()
+        for si in self.focus_atts:
+            after_df[si] = after_df[si].apply(lambda x: int(x not in self.protected_values[si]))
+        return after_df
+
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../data/train/adult__Imputer.csv")
+    # data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
+    # data = pd.read_csv("../../data/adult_pre_SensitiveAttsEncoder_sex_race_income-per-year.csv")
+    # cur_o = SK_OrdinalEncoder(data, ["sex", "race"], "income-per-year", ">50K")
+    cur_o = SK_OneHotEncoder(data, ["workclass", "education", "marital-status", "occupation", "relationship", "native-country"])
+    # cur_o = CustomCateAttsEncoder(data, ["sex", "race", "income-per-year"], {"sex": ["Female"], "race": ["Black"], "income-per-year": ["<=50K"]})
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/preprocess/fair_preprocessors.py b/pipeline/preprocess/fair_preprocessors.py
new file mode 100644
index 0000000..aeb519f
--- /dev/null
+++ b/pipeline/preprocess/fair_preprocessors.py
@@ -0,0 +1,82 @@
+"""
+    Class of fairness preprocessing interventions
+"""
+
+import numpy as np
+import pandas as pd
+from aif360.datasets import BinaryLabelDataset
+from aif360.algorithms.preprocessing import Reweighing as Reweighing
+from aif360.algorithms.preprocessing import LFR as LFR
+from aif360.algorithms.preprocessing import DisparateImpactRemover
+from pipeline.preprocess.preprocessor import Preprocessor
+
+class AIF_Reweighing(Preprocessor):
+
+    def __init__(self, df, target_col, sensitive_att):
+        """
+        :param df: pandas dataframe, stores the data to fit the scaler.
+        :param target_col: str, the name of the target variable in above data.
+        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+
+        """
+
+        cur_step = Reweighing([{sensitive_att: 0}], [{sensitive_att: 1}])
+        super().__init__("@".join(["AIF_Reweighing", sensitive_att]), df, step=cur_step, fit_flag=True, weight_flag=True, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
+
+
+class AIF_DIRemover(Preprocessor):
+
+    def __init__(self, df, target_col, sensitive_att, repair_level):
+        """
+        :param df: pandas dataframe, stores the data to fit the scaler.
+        :param target_col: str, the name of the target variable in above data.
+        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+
+        """
+        if repair_level is None or not isinstance(repair_level, float):
+            print("Input repair_level is not valid! Should be float within [0,1]!")
+            raise ValueError
+        else:
+            if repair_level < 0 or repair_level > 1:
+                print("Input repair_level is not valid! Should be float within [0,1]!")
+                raise ValueError
+        self.repair_level = repair_level
+        cur_step = DisparateImpactRemover(repair_level=repair_level, sensitive_attribute=sensitive_att)
+
+        super().__init__("@".join(["AIF_DIRemover", sensitive_att]), df, step=cur_step, fit_flag=False, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
+
+
+class AIF_LFR(Preprocessor):
+
+    def __init__(self, df, target_col, sensitive_att):
+        """ NOTE: very sensitive to input data, refer the example in AIF 360 for this preprocessor
+        :param df: pandas dataframe, stores the data to fit the scaler.
+        :param target_col: str, the name of the target variable in above data.
+        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
+        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+
+        """
+        # TODO: fix the bug of LFR for not returning categorical atts
+        # TODO: experiment with the same data used by AIF360 tutorial to compare whether the categorical atts are returned
+        cur_step = LFR([{sensitive_att: 0}], [{sensitive_att: 1}])
+        super().__init__("@".join(["AIF_LFR", sensitive_att]), df, step=cur_step, fit_flag=True, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
+
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
+    # cur_o = AIF_Reweighing(data, "income-per-year", "sex")
+    # cur_o = AIF_LFR(data, "income-per-year", "sex")
+    cur_o = AIF_DIRemover(data, "income-per-year", "sex", 0.8)
+
+    after_data = cur_o.apply(data)
+    # for Reweighing
+    # after_data, new_weights = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
+    # for Reweighing
+    # print(len(new_weights))
+    # print(new_weights)
\ No newline at end of file
diff --git a/pipeline/preprocess/filters.py b/pipeline/preprocess/filters.py
new file mode 100644
index 0000000..b2a4426
--- /dev/null
+++ b/pipeline/preprocess/filters.py
@@ -0,0 +1,69 @@
+"""
+    Classes to filter slice of data.
+"""
+import pandas as pd
+from pipeline.preprocess.preprocessor import Preprocessor
+
+# utility functions
+def wrap_filter(att, value):
+    if isinstance(value, str): # string or null
+        if value in ["?", ""]: # for null
+            return '{}!={}'.format(att, att)
+        else:
+            return '{}=="{}"'.format(att, value)
+    else: # numerical value
+        return '{}=={}'.format(att, value)
+
+# TODO: add multiple filter class
+class RowFilter(Preprocessor):
+
+    def __init__(self, df, column, value):
+        """
+        :param column: str, name of the column name to be filtered.
+        :param value: str, integer, float, the value of the column to be filtered.
+        """
+        if column is None or value is None:
+            print("Need to specify column and value to create filter!")
+            raise ValueError
+        if column not in df.columns:
+            print("Need to specify valid column!")
+            raise ValueError
+        if value not in df[column].unique():
+            print("Need to specify valid value!")
+            raise ValueError
+        self.column = column
+        self.value = value
+        super().__init__(step_name="@".join(["RowFilter", column, str(value)]), df=df, focus_atts=[column], fit_flag=False)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to be filtered.
+        :return: pandas dataframe, stores the data after filter.
+        """
+        return df.query(wrap_filter(self.column, self.value))
+
+class RemoveColumnFilter(Preprocessor):
+    def __init__(self, df, exclude_cols):
+        """
+        :param exclude_cols: list of string, each string represents the name of the column to be excluded.
+        """
+
+        super().__init__(step_name="RemoveColumnFilter", df=df, focus_atts=exclude_cols, fit_flag=False)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to be filtered.
+        :return: pandas dataframe, stores the data after filter.
+        """
+        return df.drop(columns=self.focus_atts)
+
+if __name__ == '__main__':
+
+    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
+    # cur_o = RowFilter(data, "sex", 0)
+    cur_o = RemoveColumnFilter(data, ["sex","race"])
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_" + cur_o.get_name() + ".csv", index=False)
+
+    print(cur_o.get_name())
diff --git a/pipeline/preprocess/imputers.py b/pipeline/preprocess/imputers.py
new file mode 100644
index 0000000..36aee84
--- /dev/null
+++ b/pipeline/preprocess/imputers.py
@@ -0,0 +1,88 @@
+"""
+    Classes to impute missing values in data.
+"""
+import numpy as np
+import pandas as pd
+import datawig
+from pipeline.preprocess.preprocessor import Preprocessor
+from sklearn.impute import SimpleImputer
+
+class DropNAImputer(Preprocessor):
+    def __init__(self, df, na_mark=None):
+        """
+        :param df: pandas dataframe, stores the data to fit the imputer.
+        :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
+        """
+        super().__init__("DropNAImputer", df=df, fit_flag=False, na_mark=na_mark)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to impute.
+        :return: pandas dataframe, stores the data after impute.
+        """
+        if self.na_mark:
+            df = df.replace({self.na_mark:np.nan})
+        return df.dropna()
+
+class ModeImputer(Preprocessor):
+    def __init__(self, df, num_atts, cate_atts, na_mark=None):
+        """
+        :param df: pandas dataframe, stores the data to fit the imputer.
+        :param num_atts: list of str, each str represents the name of numerical column to be imputed using the mean value.
+        :param cate_atts: list of str, each str represents the name of categorical column to be imputed using the most frequent value.
+        :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
+        """
+        if len(set(num_atts).intersection(cate_atts)) > 0:
+            print("Some attributes are both in num_atts and cate_atts!")
+            raise ValueError
+
+        cur_step = {}
+        if len(cate_atts) > 0:
+            for ci in cate_atts:
+                cur_step[ci] = SimpleImputer(strategy='most_frequent')
+        if len(num_atts) > 0:
+            for ni in num_atts:
+                cur_step[ni] = SimpleImputer(strategy='mean')
+
+        super().__init__("@".join(["ModeImputer"]+num_atts+cate_atts), df, step=cur_step, focus_atts=cate_atts+num_atts, fit_flag=True, na_mark=na_mark)
+
+
+class DatawigImputer(Preprocessor):
+    def __init__(self, df, impute_atts, na_mark=None, output_path="datawig/", num_epochs=50):
+        """
+        :param df: pandas dataframe, stores the data to fit the imputer.
+        :param impute_atts: list of str, each str represents the name of column to be imputed using datawig model. Column can be categorical or numerical.
+        :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
+        :param output_path: str, the path to store the learned datawig model.
+        :param num_epochs: integer, the maximum iteration of datawig model.
+        """
+        super().__init__("@".join(["DatawigImputer"] + impute_atts), df, focus_atts=impute_atts, fit_flag=False, na_mark=na_mark)
+
+        learned_imputers = {}
+        for ai in impute_atts:
+            learned_imputers[ai] = datawig.SimpleImputer(input_columns=list(set(df.columns).difference(ai)),
+                                                          output_column=ai, output_path=output_path).fit(train_df=df, num_epochs=num_epochs)
+        self.step = learned_imputers
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned imputer.
+        :return: pandas dataframe, stores the data after impute.
+        """
+        if self.na_mark:
+            df = df.replace({self.na_mark:np.nan})
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.step[ai].predict(df)[ai + '_imputed']
+        return after_df
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
+    cur_o = DropNAImputer(data, na_mark="?")
+    # cur_o = ModeImputer(data, ["fnlwgt"], ["workclass"], na_mark="?")
+    # cur_o = DatawigImputer(data, ["workclass"], na_mark="?")
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/preprocess/preprocessor.py b/pipeline/preprocess/preprocessor.py
new file mode 100644
index 0000000..6c5ffae
--- /dev/null
+++ b/pipeline/preprocess/preprocessor.py
@@ -0,0 +1,71 @@
+"""
+    Super class for all the supported preprocessor classes.
+"""
+import numpy as np
+from aif360.datasets import BinaryLabelDataset
+from pipeline.step import Step
+
+class Preprocessor(Step):
+    def __init__(self, step_name, df, step=None, focus_atts=[], fit_flag=True, weight_flag=False, sensitive_att=None, target_col=None, fair_aware=False, na_mark=None):
+        """
+        :param step_name: str, name of the current input step.
+        :param df: pandas dataframe, stores the data.
+        :param step: object of the initialized class. If none, initialize here.
+        :param focus_atts: lisf of str, each str represents the name of a column in above data that will be pre-processed.
+        :param fit_flag: boolean, whether to initialize step object here.
+        :param weight_flag: boolean, whether to output extra sample weight after fair-preprocessor.
+        :param sensitive_att: str, the name of a sensitive attribute.
+        :param target_col: str, the name of the target attribute.
+        :param fair_aware: boolean, whether the preprocessor is fair-aware. Default is False. If true, sensitive_att and target_col can not be null.
+        """
+        super().__init__(step_name=step_name, df=df, focus_atts=focus_atts, sensitive_att=sensitive_att, target_col=target_col)
+
+        if len(focus_atts) > 0 and fit_flag:
+            fitted_step = {}
+            for idx, ai in enumerate(focus_atts):
+                fitted_step[ai] = step[ai].fit(np.array(df[ai]).reshape(-1, 1))
+            self.step = fitted_step
+        elif fair_aware and fit_flag: # for fair-preprocessors
+            aif_df = BinaryLabelDataset(df=df, label_names=[target_col], protected_attribute_names=[sensitive_att])
+            self.step = step.fit(aif_df)
+        else:
+            if step is not None:
+                self.step = step
+
+        # address different encoding of missing values
+        if na_mark is not None:
+            self.na_mark = na_mark
+        else:
+            self.na_mark = None
+        self.fair_aware = fair_aware
+        self.fit_flag = fit_flag
+        self.weight_flag = weight_flag
+
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned discretizer.
+        :return: pandas dataframe, stores the data after discretize.
+        """
+        if self.na_mark:
+            df = df.replace({self.na_mark:np.nan})
+        if self.fair_aware: # fair-preprocessor
+            aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+            if self.fit_flag: # fit has been initialized
+                after_aif_df = self.step.transform(aif_df)
+            else: # fit and transform is combined, e.g. DisparateImpactRemover
+                after_aif_df = self.step.fit_transform(aif_df)
+
+            after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+            if self.weight_flag:
+                preprocessed_weights = after_aif_df.instance_weights
+
+        else: # regular preprocessor
+            after_df = df.copy()
+            for ai in self.focus_atts:
+                after_df[ai] = self.step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        if self.weight_flag: # for the preprocessor that updates weights, e.g. Reweighing
+            return after_df, preprocessed_weights
+        else:
+            return after_df
\ No newline at end of file
diff --git a/pipeline/preprocess/samplers.py b/pipeline/preprocess/samplers.py
new file mode 100644
index 0000000..e95ea2d
--- /dev/null
+++ b/pipeline/preprocess/samplers.py
@@ -0,0 +1,96 @@
+"""
+    Classes to sample a subset of data.
+"""
+
+import numpy as np
+import pandas as pd
+from pipeline.preprocess.preprocessor import Preprocessor
+
+class RandomSampler(Preprocessor):
+    def __init__(self, df, sample_n, random_state=0):
+        """
+        :param sample_n: integer, the size of the sampled subset of data
+        """
+        if not sample_n:
+            print("Need to specify a size greater than 0!")
+            raise ValueError
+        self.sample_n = sample_n
+        self.random_state = random_state
+        super().__init__("RandomSampler@"+str(sample_n), df=df, fit_flag=False)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to be sampled.
+        :return: pandas dataframe, stores the data after sample.
+        """
+        return df.sample(n=self.sample_n, random_state=self.random_state)
+
+class BalancePopulationSampler(Preprocessor):
+    def __init__(self, df, sample_n, balance_col, random_state=0):
+        """
+        :param sample_n: integer, the size of the sampled subset of data
+        :param balance_col: str, the name of a categorical column that the population of groups within this column are balanced in the sampled subset.
+        :param random_state: integer, the seed for random process, same as random_state in pandas.DataFrame.sample.
+
+        """
+        if not sample_n:
+            print("Need to specify a size greater than 0!")
+            raise ValueError
+        if not balance_col:
+            print("Need to specify the name of a column to perform balance sampling within this column!")
+            raise ValueError
+        if balance_col not in df.columns:
+            print("Need to specify a valid column to perform balance sampling within this column!")
+            raise ValueError
+        self.sample_n = sample_n
+        self.balance_col = balance_col
+        self.random_state = random_state
+        super().__init__("@".join(["BalanceSampler", balance_col, str(sample_n)]), df=df, fit_flag=False)
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to be sampled.
+        :return: pandas dataframe, stores the data after sample.
+        """
+        # TODO: update to minimum sample set
+        balance_groups = list(df[self.balance_col].unique())
+        n_group = int(np.ceil(self.sample_n/len(balance_groups)))
+        # print(n_group)
+        sampled_df = {}
+        small_groups = []
+        for gi in balance_groups:
+            gi_data = df[df[self.balance_col]==gi]
+            if gi_data.shape[0] < n_group: # for small group, accept all in the balanced samples
+                sampled_df[gi] = gi_data
+                small_groups.append(gi)
+            else:
+                sampled_df[gi] = df[df[self.balance_col]==gi].sample(n=n_group, random_state=self.random_state)
+
+        after_df = pd.DataFrame()
+        if not self.sample_n % len(balance_groups): # for even groups
+            for gi in balance_groups:
+                after_df = pd.concat([after_df, sampled_df[gi]])
+        else: # for odd groups, remove extra sampled items
+            remove_cates = list(set(balance_groups).difference(small_groups))
+            print(len(balance_groups)*n_group-self.sample_n)
+            for gi in np.random.choice(remove_cates, len(balance_groups)*n_group-self.sample_n, False):
+                after_df = pd.concat([after_df, sampled_df[gi].head(sampled_df[gi].shape[0]-1)])
+                print(gi, after_df.shape[0])
+                del sampled_df[gi] # remove the group already added into the final sample
+            print()
+            print(sampled_df.keys())
+            for gi in sampled_df:
+                after_df = pd.concat([after_df, sampled_df[gi]])
+                print(gi, after_df.shape[0])
+        print()
+        print(after_df.groupby(self.balance_col).count())
+        return after_df
+
+if __name__ == '__main__':
+    # cur_o = BalancePopulationSampler(1000, "marital-status")
+    data = pd.read_csv("../../data/adult.csv")
+    cur_o = RandomSampler(data, 1000)
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_" + cur_o.get_name() + ".csv", index=False)
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/preprocess/scalers.py b/pipeline/preprocess/scalers.py
new file mode 100644
index 0000000..7a1e80f
--- /dev/null
+++ b/pipeline/preprocess/scalers.py
@@ -0,0 +1,47 @@
+"""
+    Classes to scale data.
+"""
+import pandas as pd
+from sklearn.preprocessing import StandardScaler, MinMaxScaler
+from pipeline.preprocess.preprocessor import Preprocessor
+
+class SK_StandardScaler(Preprocessor):
+
+    def __init__(self, df, num_atts, copy=True, with_mean=True, with_std=True):
+        """
+        :param df: pandas dataframe, stores the data to fit the scaler.
+        :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
+        :param copy: same parameter with sklearn StandardScaler
+        :param with_mean: same parameter with sklearn StandardScaler
+        :param with_std: same parameter with sklearn StandardScaler
+        """
+        cur_step = {}
+        for ai in num_atts:
+            cur_step[ai] = StandardScaler(copy=copy, with_mean=with_mean, with_std=with_std)
+
+        super().__init__("@".join(["StandardScaler"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+
+
+class SK_MinMaxScaler(Preprocessor):
+    def __init__(self, df, num_atts, feature_range=(0, 1), copy=True):
+        """
+        :param df: pandas dataframe, stores the data to fit the scaler.
+        :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
+        :param feature_range: same parameter with sklearn MinMaxScaler
+        :param copy: same parameter with sklearn MinMaxScaler
+        """
+        cur_step = {}
+        for ai in num_atts:
+            cur_step[ai] = MinMaxScaler(feature_range=feature_range, copy=copy)
+        super().__init__("@".join(["MinMaxScaler"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
+    cur_o = SK_StandardScaler(data, ["fnlwgt", "age"])
+    # cur_o = SK_MinMaxScaler(data, ["fnlwgt", "age"])
+
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
\ No newline at end of file
diff --git a/pipeline/preprocess/splitters.py b/pipeline/preprocess/splitters.py
new file mode 100644
index 0000000..d1ed315
--- /dev/null
+++ b/pipeline/preprocess/splitters.py
@@ -0,0 +1,134 @@
+"""
+    Classes to split data into train, validation, and test set.
+"""
+from pipeline.preprocess.preprocessor import Preprocessor
+import numpy as np
+import pandas as pd
+from sklearn.model_selection import StratifiedShuffleSplit
+
+def valid_split_ratio(input_ratios):
+    if input_ratios is None:
+        print("Need to specify split_ratio!")
+        raise ValueError
+    else:
+        if len(input_ratios) == 1:  # not valid, at least two for train and test
+            print("split_ratio should have at least 2 values for train and test sets and at most 3 values for train, validation and test sets!")
+            raise ValueError
+        if sum([not isinstance(x, float) for x in input_ratios]) > 0:
+            print("split_ratio includes non float value!")
+            raise ValueError
+        for x in input_ratios:
+            if not isinstance(x, float):
+                print("split_ratio includes non float value!")
+                raise ValueError
+            else:
+                if x < 0 or x > 1:
+                    print("split_ratio includes not valid value! Value should between 0 and 1.")
+                    raise ValueError
+                if sum(input_ratios) != 1:
+                    print("The sum of split_ratio does not equal to 1!")
+                    raise ValueError
+
+    return True
+
+class BalanceTargetSplitter(Preprocessor):
+    def __init__(self, df, split_ratio, target_col, seed=0):
+        """
+        :param df: pandas dataframe, stores the data to split.
+        :param split_ratio: list of float, each float represents the size-ratio of splitted data. Corresponding order maps to the size of the train, [validataion], and test set.
+                            Value ranges in [0,1]. Sum of the values in this list should be equal to 1.
+                            e.g. [0.7, 0.2, 0.1] means 70% train, 20% validation, and 10% test set.
+        :param target_col: str, the name of the target variable in above data.
+        :param seed: integer, seed to be used to generate random state. Same as 'random_state' in sklearn.model_selection.StratifiedKFold. Default is 0.
+        """
+        super().__init__("@".join(["BalanceTargetSplitter", str(len(split_ratio))]), df=df, fit_flag=False, target_col=target_col)
+        if valid_split_ratio(split_ratio):
+            if len(split_ratio) == 2: # train and test
+                train_size = split_ratio[0]
+                test_size = split_ratio[1]
+                self.splitters = [StratifiedShuffleSplit(n_splits=1, test_size=test_size, train_size=train_size, random_state=seed)]
+            else: # train, validation and test
+                train_size = split_ratio[0]
+                validation_size = split_ratio[1]
+                test_size = split_ratio[2]
+
+                self.splitters = [StratifiedShuffleSplit(n_splits=1, test_size=test_size+validation_size, train_size=train_size, random_state=seed),
+                                  StratifiedShuffleSplit(n_splits=1, test_size=test_size, train_size=validation_size, random_state=seed)]
+
+        self.split_ratio = split_ratio
+        self.seed = seed
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned splitter.
+        :return: pandas dataframe, stores the data after split.
+        """
+        def split_to_df(splitter, input_df, y_label):
+            after_df_1 = pd.DataFrame()
+            after_df_2 = pd.DataFrame()
+            X = np.array(df.drop(columns=[y_label]))
+            y = np.array(df[y_label])
+            for index_1, index_2 in splitter.split(X, y):
+                X_1, X_2 = X[index_1], X[index_2]
+                y_1, y_2 = y[index_1], y[index_2]
+
+                after_df_1 = pd.concat([after_df_1, pd.DataFrame(data=np.hstack((X_1, y_1.reshape(-1,1))) ,columns=input_df.columns)])
+                after_df_2 = pd.concat([after_df_2, pd.DataFrame(data=np.hstack((X_2, y_2.reshape(-1,1))), columns=input_df.columns)])
+            return after_df_1, after_df_2
+
+        if len(self.split_ratio) == 2: # without validation set
+            return split_to_df(self.splitters[0], df, self.target_col)
+        else: # with validation set
+            after_train_df, rest_df = split_to_df(self.splitters[0], df, self.target_col)
+            after_val_df, after_test_df = split_to_df(self.splitters[1], rest_df, self.target_col)
+            return after_train_df, after_val_df, after_test_df
+
+class RandomSplitter(Preprocessor):
+    def __init__(self, df, split_ratio, seed=0):
+        """
+        :param df: pandas dataframe, stores the data to split.
+        :param split_ratio: list of float, each float represents the size-ratio of splitted data. Corresponding order maps to the size of the train, [validataion], and test set.
+                            Value ranges in [0,1]. Sum of the values in this list should be equal to 1.
+                            e.g. [0.7, 0.2, 0.1] means 70% train, 20% validation, and 10% test set.
+        :param seed: integer, seed to be used to generate random state. Same as 'random_state' in sklearn.model_selection.StratifiedKFold. Default is 0.
+        """
+        super().__init__("@".join(["RandomSplitter", str(len(split_ratio))]), df=df, fit_flag=False)
+
+        if valid_split_ratio(split_ratio):
+            self.split_ratio = split_ratio
+        self.seed = seed
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned splitter.
+        :return: pandas dataframe, stores the data after split.
+        """
+        df = df.sample(frac=1, random_state=self.seed).reset_index(drop=True)
+        if len(self.split_ratio) == 2:  # without validation set
+            split_idx = int(self.split_ratio[0]*df.shape[0])
+            return df[:split_idx], df[split_idx:]
+        else:  # with validation set
+            split_idx_1 = int(self.split_ratio[0] * df.shape[0])
+            split_idx_2 = split_idx_1 + int(self.split_ratio[1] * df.shape[0])
+
+            return df[:split_idx_1], df[split_idx_1:split_idx_2], df[split_idx_2:]
+
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
+    # cur_o = BalanceTargetSplitter(data, [0.7, 0.3], "income-per-year")
+    cur_o = RandomSplitter(data, [0.5, 0.3, 0.2])
+
+    after_train, after_val, after_test = cur_o.apply(data)
+    # after_train, after_test = cur_o.apply(data)
+    print(after_train.shape)
+    print(after_val.shape)
+    print(after_test.shape)
+    # after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+
+    print(cur_o.get_name())
+
+
+
+
+
diff --git a/pipeline/preprocess/transformers.py b/pipeline/preprocess/transformers.py
new file mode 100644
index 0000000..63919cc
--- /dev/null
+++ b/pipeline/preprocess/transformers.py
@@ -0,0 +1,24 @@
+"""
+    Classes to transform data for multiple columns.
+"""
+
+import numpy as np
+import pandas as pd
+from sklearn.compose import ColumnTransformer
+from pipeline.preprocess.preprocessor import Preprocessor
+
+class SK_ColumnTransformer(Preprocessor):
+    def __init__(self, df, cols):
+        """
+        :param df: pandas dataframe, stores the data to fit the discretizer.
+        :param cols: list of str, each str represents the name of an attribute in above data to be transformed.
+        """
+        pass
+    # TODO: add column transformer
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned discretizer.
+        :return: pandas dataframe, stores the data after discretize.
+        """
+        pass
\ No newline at end of file
diff --git a/pipeline/step.py b/pipeline/step.py
new file mode 100644
index 0000000..afd4aa5
--- /dev/null
+++ b/pipeline/step.py
@@ -0,0 +1,49 @@
+"""
+    Base abstract class for every step supported in this system.
+"""
+STEP_NAMES = {"Sampler": "SA"}
+
+class Step():
+    def __init__(self, step_name, df=None, focus_atts=[], sensitive_att=None, target_col=None):
+
+        if df is None:
+            print("Input data is empty!")
+            raise ValueError
+
+        if sensitive_att:
+            if sensitive_att not in df.columns:
+                print("Need to specify a valid sensitive attribute!")
+                raise ValueError
+            self.sensitive_att = sensitive_att
+        if target_col is not None:
+            if target_col not in df.columns:
+                print("Need to specify a valid target attribute to be predicted!")
+                raise ValueError
+            if len(df[target_col].unique()) != 2:
+                print("Only support binary target feature now!")
+                raise ValueError
+            self.target_col = target_col
+            self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        if len(focus_atts) > 0:
+            if sum([x not in df.columns for x in focus_atts]) > 0:
+                print("Some specified attributes do not appear in the data!")
+                raise ValueError
+            self.focus_atts = focus_atts
+        self.name = step_name
+        # self.input_data = df
+
+
+
+    def apply(self, df):
+        """
+        :param df: pandas dataframe, stores the data to apply the learned discretizer.
+        :return: pandas dataframe, stores the data after discretize.
+        """
+        raise NotImplementedError
+
+
+    def get_name(self): # return full name to print out
+        return self.name
+
+    def get_abbr_name(self): # return abbreviated name used in the file name of data
+        return STEP_NAMES[self.name]

From d06c7f3737c1b9cbc2bf58bbd53b9e4cc319e7a9 Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Wed, 8 Jul 2020 12:06:37 -0400
Subject: [PATCH 3/7] Remove unnecessary files and rename driver python script
 accordingly

---
 pipeline/fairness_label.py                  | 74 ---------------------
 pipeline/{label_pipeline.py => fairprep.py} |  0
 2 files changed, 74 deletions(-)
 delete mode 100644 pipeline/fairness_label.py
 rename pipeline/{label_pipeline.py => fairprep.py} (100%)

diff --git a/pipeline/fairness_label.py b/pipeline/fairness_label.py
deleted file mode 100644
index ce74e6d..0000000
--- a/pipeline/fairness_label.py
+++ /dev/null
@@ -1,74 +0,0 @@
-from sklearn.metrics import confusion_matrix
-import numpy as np
-def get_static_label(df, sensi_atts, target_name, round_digit=3):
-    groupby_cols = sensi_atts+[target_name]
-    placeholder_att = list(set(df.columns).difference(groupby_cols))[0]
-    
-    count_all = df[groupby_cols+[placeholder_att]].groupby(groupby_cols).count()
-    values_all = count_all.get_values()
-    index_all = list(count_all.index)
-
-    if len(sensi_atts) == 1:
-        norm_cols = [target_name]
-    elif len(sensi_atts) == 2:
-        norm_cols = [sensi_atts[0], target_name]
-    norm_values = df[norm_cols+[placeholder_att]].groupby(norm_cols).count().get_values()
-    
-    res_dict = {}
-    if 0 < len(sensi_atts) <= 2:
-        s1_n = len(df[sensi_atts[0]].unique())
-        t_n = len(df[target_name].unique())
-        for idx, tuple_i in enumerate(index_all):
-            if len(tuple_i[:-1]) == 1:
-                key_tuple = (tuple_i[0])
-            else:
-                key_tuple = tuple_i[:-1]
-            idx_denom = idx % 2 + int(idx / (s1_n*t_n))*t_n # only work for binary 2nd sensitive att
-            if key_tuple not in res_dict:
-                res_dict[key_tuple] = {tuple_i[-1]: round(values_all[idx][0]/norm_values[idx_denom][0], round_digit)}
-            else:
-                res_dict[key_tuple].update({tuple_i[-1]: round(values_all[idx][0]/norm_values[idx_denom][0], round_digit)})
-    else: # for more than 2 sensitive atts
-        pass
-    return res_dict
-
-def compute_evaluation_metric_binary(true_y, pred_y, label_order):
-    TN, FP, FN, TP = confusion_matrix(true_y, list(pred_y), labels=label_order).ravel()
-    P = TP + FN
-    N = TN + FP
-    ACC = (TP+TN) / (P+N) if (P+N) > 0.0 else np.float64(0.0)
-    return dict(
-                PR = P/ (P+N), P = TP + FN, N = TN + FP,
-                TPR=TP / P, TNR=TN / N, FPR=FP / N, FNR=FN / P,
-                PPV=TP / (TP+FP) if (TP+FP) > 0.0 else np.float64(0.0),
-                NPV=TN / (TN+FN) if (TN+FN) > 0.0 else np.float64(0.0),
-                FDR=FP / (FP+TP) if (FP+TP) > 0.0 else np.float64(0.0),
-                FOR=FN / (FN+TN) if (FN+TN) > 0.0 else np.float64(0.0),
-                ACC=ACC,
-                ERR=1-ACC,
-                F1=2*TP / (2*TP+FP+FN) if (2*TP+FP+FN) > 0.0 else np.float64(0.0)
-            )
-def get_performance_label(df, sensi_atts, target_name, posi_target, output_metrics=["TPR", "FPR", "TNR", "FNR", "PR"], round_digit=3):
-    
-    groupby_cols = sensi_atts+[target_name]
-    placeholder_att = list(set(df.columns).difference(groupby_cols))[0]
-    
-    count_all = df[groupby_cols+[placeholder_att]].groupby(groupby_cols).count()
-    index_all = list(count_all.index)
-
-    res_dict = {}
-    target_label_order = [posi_target, set(df[target_name]).difference([posi_target]).pop()]
-    
-    for tuple_i in index_all:
-        if len(tuple_i[:-1]) == 1:
-            key_tuple = (tuple_i[0])
-        else:
-            key_tuple = tuple_i[:-1]
-        cur_q = []
-        for idx, vi in enumerate(tuple_i[:-1]):
-            cur_q.append("{}=='{}'".format(sensi_atts[idx], vi))
-        tuple_df = df.query(" and ".join(cur_q))
-        metrics_all = compute_evaluation_metric_binary(list(tuple_df[target_name]), list(tuple_df["pred_"+target_name]), target_label_order)
-        res_dict[key_tuple] = {x: round(metrics_all[x], round_digit) for x in metrics_all if x in output_metrics}
-        
-    return res_dict
\ No newline at end of file
diff --git a/pipeline/label_pipeline.py b/pipeline/fairprep.py
similarity index 100%
rename from pipeline/label_pipeline.py
rename to pipeline/fairprep.py

From d3dc17d5d694bb047ca0ea3f22a456f6e00f7807 Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Wed, 8 Jul 2020 12:11:02 -0400
Subject: [PATCH 4/7] cleanup

---
 .../accuracy_vs_di-skyline-order.png          |  Bin 43040 -> 0 bytes
 .../accuracy_vs_di-skyline_formula.png        |  Bin 42840 -> 0 bytes
 .../accuracy_vs_fnr-skyline-order.png         |  Bin 43584 -> 0 bytes
 .../accuracy_vs_fnr-skyline_formula.png       |  Bin 43356 -> 0 bytes
 .../accuracy_vs_fpr-skyline-order.png         |  Bin 43619 -> 0 bytes
 .../accuracy_vs_fpr-skyline_formula.png       |  Bin 43487 -> 0 bytes
 examples_skyline/fair_prep_results.py         |   92 -
 examples_skyline/helper.py                    |   53 -
 examples_skyline/missing_data_results.py      |   99 -
 .../results_play_skyline_formula.ipynb        | 1778 -----------------
 .../results_play_skyline_order.ipynb          | 1767 ----------------
 fp/dataset_experiments.py                     |  126 +-
 fp/dataset_experiments_old.py                 |  341 ----
 fp/experiments.py                             |   90 +-
 fp/experiments_old.py                         |  383 ----
 15 files changed, 81 insertions(+), 4648 deletions(-)
 delete mode 100644 examples_skyline/accuracy_vs_di-skyline-order.png
 delete mode 100644 examples_skyline/accuracy_vs_di-skyline_formula.png
 delete mode 100644 examples_skyline/accuracy_vs_fnr-skyline-order.png
 delete mode 100644 examples_skyline/accuracy_vs_fnr-skyline_formula.png
 delete mode 100644 examples_skyline/accuracy_vs_fpr-skyline-order.png
 delete mode 100644 examples_skyline/accuracy_vs_fpr-skyline_formula.png
 delete mode 100644 examples_skyline/fair_prep_results.py
 delete mode 100644 examples_skyline/helper.py
 delete mode 100644 examples_skyline/missing_data_results.py
 delete mode 100644 examples_skyline/results_play_skyline_formula.ipynb
 delete mode 100644 examples_skyline/results_play_skyline_order.ipynb
 delete mode 100644 fp/dataset_experiments_old.py
 delete mode 100644 fp/experiments_old.py

diff --git a/examples_skyline/accuracy_vs_di-skyline-order.png b/examples_skyline/accuracy_vs_di-skyline-order.png
deleted file mode 100644
index 8673899d43952f6cd3a8a90887cb93ee847696aa..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 43040
zcmeFZcUV>1mOZ!$3W9(MQ6#A#QBcW95)}j_CqaUWWC=>n2L&-9dJs^u<RC$U<e(@>
zM$(ZCN=Aa@OwWb)y?Ryk>Q(n2zy7+uuKV44kDjo@+H1`@#+YNS$EwP5R21|S7=}^F
zU%8}?VZ_ZCwkw8wH@xHA;?N2&r1moMn&j}02l?$s@b^7-S9I+$jPd~bkEo^BYXjaC
zakzZVLBrP6!TIJr6U_RigWVllhdUOxSe#7m*<0A!@bjMIJ<G#l?%-f2#>e+>pYz(@
zGvkY!6YIh-7EJ!q1x=TTU%jr^f*rTEe-kboy1egmy`SJ0Re4rgjq3uTJaVTQg~i`=
zu>Q2oPE3>-Id`2Sk|#WbJML`c1i7+_O@fA)njSBINPt3E2rkxGA%%zO!pg?hz+_4s
zu7{i<Wb|iI$-q`mw0-olXj?elav@##*(!3leCY%i*CKeuVkD1ClfdiYeKADvLP-w8
zf#G(Mn!`&4F|h!=+#n*NhnFlIMg=d|q;Y7hcv${F@v~&(6DLmO+VqOax9~TK$DN-k
zX`K@~%xCz{Dx<J!?q{M>c%szyY7;)5>X;A*7gz0^kmb_`)7CQ2b#a5;IN7h=t-CO+
zcKN$;pzE)~%+e*jsm`2g<0cx+k=8wisLiVCNbsvyE=7CWV))G(+s3De%spU)Qolrq
z+P4+jo2{;|r@1druMIq)(Vy&d`AOwC+0l?Uot9aKXJr(zlQyn?uGRg)x;;_pSVM%U
z2428@=}xcPWO_!m+3nk3O>+wGH^p3d@apyJWcZ0*Sy|cIpg(i2`L`1UdxwcO`ONaI
zqU)aA+V52F+<B*$XPM9#dERQOE6;7Y*SWpe**4#muVnVVP1)+mJ)_Z()7RU|Jl$#O
z=!&Z(R|ZJFeExg||G}m=Rnl`USu0K3l8_O(wU8ubFk9-rK9x_)#FQaq-NnJlS!3N>
zIP;|tjhb?#SWSPwQK*5_8wP5*H}gy$vlk5P8wA(Z`?n3p2@xS7i7#I4Biy@m$#<k7
zJn8l8BfCgR&3}A;d`~PgV=+x9QyVUnn1n>|Mg_LHT)v$?>sbD~>e!QK&oot4U&sbB
zrC;=?F6`*MBUHLJkv7qi@U+Nb{Di#Qbm3^*+lyq5zrL&4maQgse^8E=OebS<k-mBJ
zCc19Z`qytYl9UTepBLJXrh2X~6>TncbHAg+c)w!Of-BWhTQ{?Fa+L1geJ`)1Bnaa;
z(){+~2=n%e|M>U!ah*kuX`Gy#XW>Z+9zA+=r2Y;6sZ*z_dfaD9Px0|JMTj}p-ua&O
z@rZH#&p;`gUrxnV-RsLU!uYFdYA?@Lz4jp?3u|sx&9NU<*3i_vO+iW7FwvgY0CVW7
zy!>N%MMeJ8dbwxdG4od)$uX%r+?uMzJ^#C!IsC_uTlTX97v`t(`{|QfE%fqjtXo=J
z4Vq|=9kc3#lT#R3B`HVNS2KJ6(qj^~iEv%&Gz&OewJ=F&Dp~BfUUkIyp1pko%wevc
zjtqn1t92o#8_c3TO(T?(3TMjJuLhk`s%MY(P#<ysZRP6fTGij*|EsUG8}8%nMgMx3
zD{wF#z1*+*o{Q~?1qI)GTt_2E9v-tfv%0!U=g|NID){*EV>pfE=7PpS28Lf@bWBV(
zKYskU>g4pHrl#hsL9vq{EQ&A~gv_$F+Fa}IjyruNoiO>UcnnLv`}+GIOKR0Moth+M
z3R`#Oo<(DzoYYz$A=;|%K9!fTh_8p|prow)oRfp&j8(_Gv!RA=DV^_){cGy#_<#TY
zEsO6gcFu&u!)Oqm(F@iN4Gp~xVpj=!c~UM+j>+}&qF9)K#VsbcvEx4-#FQ>vxZo#s
z5G&ZXBu$ch_-A#XPW5xf;o;$g#|PO{6+?Nu;6ZOK<(H4Y^=DGo*N?a(6n^65Ni@^b
zujfy|rR+?!Y!gL6+rFL_dE3@RB@G9M+>+AL_FM~%%@N^pwb8~%lblWgy8-E5yGp9A
zteX#5Ee1dN75#jEUf|C6Gce4YA|lsdV1fe!15YVOBs=%H%HeaZ2)rh@Z<`N&eMu|n
z?BP+OuCC51UszbUHWn9nh?$vB-m)t<WofEA2_iyGQ<KW(+G3jB^_kM;DB0%>DU%t+
z%94_jZWAd<L&I9D9lzj~BF{Tdw54dkRX>Q0i!+#f^5h8|C^On?O+HOK9iN$L8*DVM
zyWC?RDCYDlIVOfBfL>6cJKr{xm5rT!64qX6M$rV9phe3ic+Tizg>3uuN1Pa_{ih3l
zQZG)lE21;PjLWlVJq6cZFd_~YOF==Q{`Bzag+T9(i~0Rt#Tf-ZDDv)fs^^$BIW_QC
zafpiQ&W|?fz+&Fm++2jP$tft<KIQ55<1zct&!2IJPwQzeP3^*++aE)u`82=Jt#0wR
z=ZzaNPSZUpaOdL{J|q)8h4ZfSGz=m!v$J-t?mAQYBe1yghXt)Vl8wr}&;`H`B-h4H
zGM;y;U(`2gXOV|HXvecD-hTG2+qOJ?&%sk)tE-bfjYf&PW<%6VY)w+J>dZEQMH%|3
zvXU+M;UcUKWq3wfFbA|V4GjiAP&0}RE$QifxRU}g@0F-?)w82>^YcN`(a}+^BS+C4
z!t`mszl#!uAv6)bP4^ZX%rC<eqvh5fgVn5)W7@zwAsF2bF_y3lA*QyjE{N5{)O2EX
zG}?J{gM@^{aiD^TQOw~AwST47vXXG`kXx@=KfgSjNF!Xx`c-7)6#e?*`n~D(g~@6r
z2{+AMyLL_V7UxeE+Lrv5Wt4Dh6HK!1{$SZx;+kN164NIqqOZ7IS6|<@Fy6XY%bm%=
z&MsFS$V{Ns(LZkdOy=Un#m(jZ=;hZid?qF)!LBko5FoTvRR{F8`nR{7`xwF-M%K3A
z$7@rz-f1YB$~|ZJ_WNhG;H}T(M~(HHw6v-4ZJ9<=F!WcC3EgRiAr>&HeMMQ;2RF|x
zF0KzN_8WTBb?!Eyh}WQqgN^O7)YeKIlh?}4fMdeS5J!e>(jcUZ*$v>V3S(lVMMXtT
z8V#~;_{`1Cy$TH-EfTAq9jsieVUHez*NLvYv_zHY>9GFv>%tH;uL$=#1a-Fyq|Sx$
zu?h%iLNpTd+9;Sbx^Thg$IqW8`P~MNnYuZ$__p`Paxm0IOz^-mBO(}fk&zWwd2RAb
zE_>G2*532<u3-yJzWnrXSDsbqWT8rg$RNe?s12lt!#iWx)D+rassYi7O&%t93q(dX
zd04keD2Ks_*Bpt1l)}l+uK_Pw{?v>MUrz>dNJ$wfCutaY{Pttz6cW;o_TDPY&dzrD
z@p+HL`mY>XN$;CC-=Z}2!MeM)yZbsCtDQ)`UEIH^j83z*IH{$s9{cOpoqUPB{9dPU
z)5eGti00aBXYe&a?30<^n`x5U8}<RrQhkD3jpt{^#=gbDxS{#as^xN}s^IR=T(?C_
z>z)ENKR>@PK`ZTqOs;!FU-ze9doKfHKll6h%joES*CtvB3}#WTvhXAv*H6ZBWfq=*
zH9>&soD2ErsPQWpH5#c6RR~BXjSvb5v6r9bdagS~N_y%AN-Vubk>c{DORq$|Kl$#_
zD{*-bX$mG%411Kzr@Y3xIz@Ce&*k1EX73F(JWS+t@2&Z0y3Haih6j)$T{>jeqV28q
z_Hu9UJS&3od8c2^?K(!MQx?9Deg1q&DN0;-Yip}$IOO_5tBO>Le9$QXGr=evW@Vl6
z^Yt|!Z%M$%->iIS0Z`(}<Hu?<eWiRZ9vS83<%VwKY)Y@sTz~-Hfu^F{f;nLgQm^H=
z*vmz*%n$9`7Xt~pXlgU!oXHc2GrEKAyp3o;nk3h+J|Ls%U<=j%L2p#-ln!BZcyu&L
zYU_8Pf|64D?Cb8aF>{1N!XS=_IgG`0q+c&KSdBfw#>VD0=trkt&FT&D(QPV^&=@Z4
zIHB1PCa4HUbYe8t4<_ip1Y9*1Al%msaH74)(dx3StmW??Uxpi*#T=5O+$K0TArtM)
zCRnD$j5!_|;WkC`%YE*dyJ?u-ef@fcib?#t^=40@J?+t>DPEgvDFD*w#T+;YS-Vw@
z?~MF1i@=(%M2a~oprl#0Ho;9`Yo#x$S_vbHDas<FmyTQRvdZM-<;_srRrBXvJ8oMn
zKo7v#v$J|-9);a5klea&GMSZbtxaZkPlZHP4%&@DYS?uBIXF0o7E9j!v<t48M=@1V
zhEE9pU#mFI05AV@2*Cdcmh!(B`9G6g{&{W_=@zYt<4xXMIYM{7$0R3nsDzSWOpdKe
z!s9dE+r>$B$@q8gcpx690LE@jz=CCPvZc-tH7uK=B%*lZsyx@mlZs#o(n73y^ym>~
zBjB9UBBv=XBO@aN@6E;9`ucaPCx~3?A;b$fPP9Gks`}NL<2ad~udQ);`01lZ%|(ur
z`gOpC0CQ*n>bKk|^qG*x$v$uzsXwX`s>){4-V3O<_ii;KLb$vuD*(1EetmzB5amwD
zRf(1i6S;d!=@msxNE^Xxx4<P*^R{C~-kW)DixX_#-rh<PB5IKC@UQ-&cqNY>kvBx1
z*ExIktYv3*LSuwzTtm39RRxZSV((r}xUt}nkdZJc^JbRK)h6!_91+P7#8$O9Sx!JH
zPQ?c4dU*(epF4L>f4+QUPVvi^FPtna(s%FPU7M@pW0MECr4=pd*<%?c<y{8E@U*;`
z=UOh4%K#Cix)i`}VP}mW0v#P*q919DOoLo8xzuedHM0niOh+T}sv0mS-W9{s*FRkS
z_U#+JlvhcZux&<Xj@e|mZ5aY@+{>mU$p;&!%#I1$yvqvR1Po+<x_osce4-;g9H?0O
zV5NVtCk3P{UeE@v0<}Oy6QRL$gJSI_@%d*VyarrSQc{cXEIxGsfQ<hZC&$gfapmR9
zmu^1;g!Ly)o1)V1{;WbtYi@a2C)2QWp+UI+3O@1bn^>3;2#|Yud9BqMZN)x&wiie*
zfq4&I*OKp^gAU!+`G{!4rLV!nvR9kozww4>!UytfIHbh+Pn3MGKYfy!pDv!Ex3#qm
zdi~n`?;PzIa5T52UlssH;HrH@U1nqUP*9v>_F8!h$v_BF^n3Jogwy^auYh;L0ol9(
z!W3@PTl5By+-nud)z61$Y2R>aB)*2}ebo3FB78nRJ|pS*edjJ;zHC~9XFGM~Ofv(Z
zaCo%GU?k>yo%;(d$6J#E<>lqiRvobi{t^a{{0$&7`?(=`{(D1L<^eh#F@ClRvrY-|
z(~C3`w}Gz$|KnILT^R~ex$5HL;x|K(Sne?fRzE0on32&Mk>T3%aZG%gu=(y>=`YBt
zXMjt4tlnz{ktHZ9$_Cg^y@rZ}$=dRa?X$BZLp4E(a7U&9l}CUip}(8Xk2C<4W%Ay*
zYn<QBt<{}JFth?7yzag8jmgY%p-}yMk9A?`S)K3e>l=Za5rnxmQaBpLzkmP!#H_65
z#Ujbop@W1EOEdkptZ!mtV->8stE;O$RtEi%XfX0pj_It9fn)oZvUVKXFCW>`EZ2Ec
zT3Xt1xB)F7lY%xqA~4Z|&u>S$Ppg@mC+p>!zeNP`r0lciNBd6>0j~p;lLD&|5kviX
z+wx5(N5CvZL_|g7iQ!06qGA-&gv&!Sx@>FBO1{OP*(({In0f6h&bg%}Z6KErI#K~E
z)DQ?d!u=jOJ6sbGSUb#mLBL0V_aY+Rt<I04$qKm3?brM2S7BijsUHE!sfhpf_3-d$
z0|fZgtsKxM0H^_>!w{KzdU~FA@#4CiKSXOKhDPSy_(&`PSBaJj5M;#lEY_E%AXc>j
z(LyY!*L!;t$VwqFLC1D29o<w!@78`i<r8q8)(3{I|Ni~^qWQ-2AZ6GVjx0>~UjJDY
zpaH|Qm?X6+55XoJR&0p15$H?wXK(oe(T|1MSg>S6lIMp7xD5hdr=X?|X)v`9@bmW%
zJuf97(6;DlYMR*Rxoog&_wJ%CgpB=Zn4KCdVChb-1N)4~j{w$@`m<-0iiSo9WDHh|
zxuKd2tn_xxf!=#{r{@+I+hG|aaNAevUTj$a>XhTcm>R*U>w|R~@Bu|U3IPzz5liPc
ztxw)Kw&=aFvEcyoJO4dM6SbhS%zeOIIy$|y-xxY+!}V-|097;sYz~M7pnW9#0ede?
z4nwT7>!@8vru$&#WQc<EK3`woF}SVv>?sIn#bwKji@~kqLr3nmaS;nl_`%mX&YuU}
zhj2|>Tf1=Hn=$v|#fyR_wTBp`ym}V-fRO}>&pjFz1IF*t*UkeG;1B}?k37B5o!8Hq
zBvPlk^SObc!9|9!8r-X41B~qFG}SenI{5kV!HgSTHVp{u|6yiHyC@G|%itFi>#`2K
z>g82dw!P(kSx)YH-(jpttozuzg>0ae2shf;t3-+=5RvTW0D>m9F?-Rk$?-T)^EMT!
z&7q|fxVN{OMeM8Z@c09a;x4N2Ny)&xrfD}WX@86w-@;745zJ<{@q|!KP0e_dx)6)W
zj`U!+Ul3)`93GF}-bi}#^y#pJ7$lTLK-YR1#-AVUUxUb(3(qW4)Ls^)3OBbR5K1_A
zWR$8v=1ekX4^^qng`kI2$4;d<%+vwPb?o<C9tQ^Yiq*u-Y!Z%ZmJnSWAAboZ8iccR
zRbR&)QnRwM>I;plo_&LmxVW`GQ#bA~0o+vc>eXkks4Q0nNs`g6ALTb4>okjQ2hojH
zo?SVjxp=0GTb^F@UOY9k<U2Ye549wfXl)ytj1PA@Poylg7C?9homlglBCPq{dymA(
zF?(vldAB!byz}e^W%3mVfMzNIJ#T%(Z?*tSq_(<R4m2Qp2^H7lW#_tgiwD4rf2*5g
z3KGQ|q{M=tbK0Ot6Y_X^n=#CSyjx!=B<Ab+ExSLYBkJE{-()aLcS*M^*CMVxO=l8B
z2ptd^7DCH6E`tuc!+OH_fw%*4Qii}EnxCI<<uTn;h_u*pugx|PDIwIfwJfGd0SC=9
z|0V-LN>y7s1jVdlHU%H^B*nz^sQoXhj}NlUly8?EICv1z9P7^PFH<g7*kWL?p^$0)
z5s8i2N?5vhBu{f^6vzM%e#r{D_+*fZ<TrT4NRxIPZ@CP@RSK)c#>#KJg1Z5vO0IY8
zFjqm2odAUv<d`@hn=**NrlTaC=P<6GXVa?(0rCTqcz`x5;F&xZ)Fd_+S^<HymAKCH
z*dCh|2wz<^`vGMP!=mY;Nr%x$=ZO#9HuDagVG`~i5W$4#TMd{@@V*2THrniAks^uR
zq0=PPwdhcc8jE!3O%T)}y{O;59S=d?zHhC*{wmecbC&@5&<mKW20u^Og{j051oxth
z+Ldpc33PKJ%qk<L!#cW@zPWH4P@!si6G@ve!>7B?LGv~6S{+`4On2zOf!Klqk%aj8
zWQfigqVN<5UtgXS1SYvTX|&w}A$S2WjeC|Qmi5q(4AZJ+8neICTcia51EjwoK!b@u
zl`XzJp|!CGMlR?wdqciukznNA=Y}OujyYngj-+Gux6RCo4K^fLSwBzxn8LT^n5iI8
zgkShb#>^ostOJTno*p}<zt@cs(~Mm9OB&FJOxX(QEYIclEd%FXM+maokg6NQLm_Pp
z4c(S+fq8%E@L}!lgxJF{t-sbiEtfx51W6t#a;+w96mAAnZ@o3CX0A6_TrJJuB5xig
zr>1r)c5nIkfUK!_T@6%O19uY$Gz+lCxX+xKEwN@}W#u@2`~n^@hsk(Yd9rs)ON#)Y
zN0m??`i>+OW<as0#~d9gH{&c_`Yn|&jynqaNp&~wr=v?MbzipJ-rD5i;nB3TOa-hs
zB{cf|y>SL5K?;Nzq|Az|abcQAOR2Hc=a?ro0H)HpT5fttj{?BWi69qpz)S}DuBA4Z
zn=N?j%Q0HoggU-*eUQV{ZrzIYPuq=I?Tn^=RYu6{d(+BlxyJz)0i7;D6sd+A)?Fw{
z{p*I`611Kf#}HA%jBx;MUo^6Dz5{9?V<1c}N^2$W0g$Y7<;tTlA?t~8b)24#T4tK!
zSTS9op^hM_dBuGQaF9TZ0B>Y>-m#ekeSdgnX69|JYUH!CwKX;0U;qmR&(SA5aWgQG
z=`R(56aPG7UO}AHiiFFyaU+fb7ol(^5ZhDc-TQH2GMQEkV~$#KxZONI=oEfdG#w3K
z+@=wr-iSzhh*7oi!*`eaz00j!0V$y+F+2sV;36dECjkMF1YhtL|2p@js>%|wQv0<j
zKJ-uk#L2h(`0`ZHex&|s-^e@N9H_4q-6QZBmFbU<j|WH5#7xH!Ht!qiEZglBP;FXS
zSuqFo)otOMoFm9^ny%+1+!mT)?Xk&2u5U#%-lnHuf-6T`e03;@Hp2*%2_$+0TG6qT
z#1%a-T&8q0=&fWjkJ&BCpnl00B=EKN!GaZQH+Oe#m@^<aOW7L8%05K836z}B;?&jE
z#VdGQKopdqKXU^o<~R|Ll$D+-8hU}Rj^nWqH#tw7kWn9xdiCD8dZOx>t!5~%!9w#z
znnEU2n@RKD-a>jY$|WTw;_@vZhP3Bd>3}X?=O|bFq-6FJWzml(+zwx8b;iEX894+<
zEw*XELvIlz=*O+BxWIF6S}bZT3xO)drQTv^9Y`UFy?Cw;hw2sHOF)$~7}$yRnR4gw
zZQzO5Np|m51>V%2Z)-F>GGdq|iJ8o6W={5(OO-x19NquRXSd~}%HwRTS9a}6sC)A!
zp$>mL->B!qOwlASl=2D$o#czwr;AbaP>K}OLS-_bdQ7L`so)(}W^W=aRO~#{0UBaj
zE}1DnwY9a$%wZ7afwA_2c!$tHl-HUCN(c}k+CX?(1fDSlwUQR3FoPVa2$2WTM8EF7
z57yxTr{dv@UysD4WfUopkWsgx<OM4lQ7kp6jRAE`KqyK=2sj_J$E=UjOyBA7hC~Q-
zu;^9e<N|r^Z8bn9MRk${#ZWF@US)QbC`}MYdCHkF@$Sj0=ZtC2{hpzQa+zt4Xaug@
z`oUMO+Q`Mh!B~C`IAn`;es3b^Up3YqFqx20$7%sUxU1l9oK^4X{GJnhe0+{GeFm_U
zQ(ES!VP*i7MxeCFc_sq_xpiNOE=<}YQ2C(hmIl(-3_`Fl_Z%SwqyV&lbyx%lmAhxD
zD&UwPBs5eE0NO@>-sv6U`+Q|9G6-b@z2pU+45V(Nq&g~n+_`)`5;D^%BcpzbXe)TI
zs8R#9GuGg;=)S&RDp=}$xGSq?<jvAltK}+(xFu}bOPBl7{3z*wRhFyIdwO{2_v<)p
zEDM9+_QwlAq7JbOBJPzI*HEFgMIi)<q3#S%0BNJ}NW02Biy;#rOu@}J9P8@tZUu0P
z9%_eB=14zgTk>8zn%TU5%rVd3P{$dRQThXiO|5%U0Zo92`Z3*y=Exc8l#~=atAFKN
ztIX1lIj56QKez4{l24so`IrNebrQN^I8_iUh^3PNd^89d2Ei|1z6A_mh<}C0s}DM5
zK%pE>JjkZCO-%#<V=-_rZ5<=eJneKnHV9U*rqT^d*Gj5KhKI)ieIkHA(iEKu(^eKg
zG&5r~^4klhkeJ6x))s&yMXu7Ll`y+_SM-P?0{=d4@(TFIO@{_TzC%e-53miuy|1lc
zff;o`?mTqxU>uaFu6si=J#9PhL&nq$7m>)FS<+(NJ!}Cu3E?PMs<3*JVYzpey5}Q}
z05aT8r4KNO1HjTE2^&ytDfaBSyIH)@q9|zHC9nt*1JJzS>5_S61J~hTz^ARStZ3nc
zNWz8irYbFs>;3@ITM(ip9DfY(z&A?15*B#`g@F8tqTgj0+2GsW>GC5W2eUjMSbGXQ
z$h#+rVxEY1P6C%z$vjxvcyh1h7r3YC@dK(<@Nnx4oPn~;R{?N&J!@xt=R-^8Oo@hl
zp5See?}6b$IY|lh@dl_nSPr|^t+_74Y}Sq{E{i2-H6%rQO^hGt(QKQn{gGm*qurn0
zH0HQSCWj{x{AyFalG6+UhLuA?!T``IRMoU#5X21*;EW)x(YP6GUcl{E9X+j^{kfG5
zlMaX@v@<i1{43Lkv;;mVDDM0BYY^&CSstK-(`h<@4wLp!QXG52o$-P-AtAv6>Y{X|
z4L~`@0QYk_d+O<hm2J$`DKwEapZPYaO1Mi)9m51NM8F5zeqMQ$sDN@URFKfP6r|RM
zpEZ{Cn`~`qK?)NP5lzS%4<8%)7d@ANwRP{89mkI^<gOBf36JPb4t*a?UK)KZWF6LR
zZN9m&!5S<ycHzNpi=KkJsQ3i6{by(K@c_mx1}c0|@$DfEb4H~<O>vbwJe(*A_X((5
zln&S@0cev8WI`!mVYU)JrsLI5MwkI^vK^u&B*&}r3FYLY4dLwah~2c6x;w)}FmH*M
zul@S98A>k#-di3hwN^f)ve_vlLQy`|tSJhmo$g*>kgqF`U%qt3&@jP>aY4+@pt1@w
z;6o87kylw<7j|9Vg^#a4_(tJcFpGTla&qK(YLCrD0+Q@exfi)bP-O}gZcuF3dw<Ih
z5m*#t&z{}!P}qR9fM5{9#S7q1(Jq(*bUhB>w*zKQ8z6a9;6`PFwWX=Fq$oE9BhSTP
z;PMD&g61_}Qr+~1Uk;2Ap-?e;&McKtQ6b&%`fR1ON-Km3gZU|^VkXz2{Q*b$WnnTw
z&7^(Zlqi|r|DwW`7ZeKak>~goLlg9hiz!N}^8x^iy!gEMIAhjB`+}O!0Rnr69&Wlj
zKT}3VhE*Qm2#lHPUv*+wA@X=wa_vxw1>UQVN}i4c19tf^DDZT=)76;%fOwTvbIb)q
zhT>Ziw}Jd9;&)Rspfb38%gXPciy-bqF0TMs=$vvwydTw`xI7qP1=Th-cDz1o%qkBV
zJbz9IOO}89xZHO%hk`<(+KN-UbyBq(h3p!!mnIe^si1rqxN|5dDD)_2fnVXSXMeo>
zK}1TcjsRP>f}CX2E$X%5lGovQ94T<9o(E}50R*xxs9B~<O5rT=TFNTiwk1!-9FrAA
zcGD*xze#j6q$-UI@ClS8gILq`3)O+4(l!PHyZ~{+pj;Dx2C@`@$iW4P6co%?tfIVj
zdyV_cJQsnOf~mzAlBu?W!V`4YKr%d`P_+ib;0S_nDOBP{B|ZT4C8K6&2ELF9rkM!{
z!l+^mB{c|JP2qq#Gr`L92R^JVSkOikS5xB5e1ltCYma4~aATY}KrEno5&w|2CO;*!
z`?alx5gV0!YRBjsNzWq5^<VrjMNm~+9Y`&AL_b71NN(<Kpp>-(1A%#8#G8bSOug{v
z0Bbr%`r3EEw4pt4;Ky1hWD`;Qkukt0ItUSgld0EZz5#+qhjE}pB2@P#Cekvsp^imN
zOx*JUs`&4&+m>n}SQ)HinBqQNC<j6%y`beQ<Z~iC7A#)^9yOT|YN!q<4DoL?LE2}h
zi3-f+<?)&BnL(Mh<1T`+gtxr!-`H>0T?lvtLk+7POklh%kVMh|;RYA-0$l>V%*OaR
zBQ;xwie|@S9=~0_oEBHq(;2sa$%FR;VlQIw0ss?`*q&j$HW8HV)#tgk@C_=W%rsh>
z%iktbM}7sfPCNk!|Aygti*Zy=w!*SFJ6n70?<SMQ-mXiTRwZZ}Tv%dSlHLEhNf10?
z2$R7g67I39+@%shq}tRumFM*7h5^9)6681Q?VNl=EPYY-&rbm_AT!>z&lV&BBvNJq
zP8(9qBD3}x^uz);E&#a;4hp(hX)&<FC<agMHqRY%3?M1Zn1Hipv9PjUBBNnygL&3;
zebhnBLWb%J{KUXnh=e};FV^7#tOHzuhXt&=G{Fj{`(_e&fx~dHPRG5=nCd3igrnmg
zPr9g2|63~jf5`_K&<24Ybmj!88aKS*A1zq~{(~w?j_^>pPMwlv_FQ;LmP(2%@&}J*
zfGYUlAnNm`gLMMHF9ahs(9gOf037HYJax4h6omA)HqFgJ8t+-vJWMqos&g^>k$DQ-
zQtc@k?8p!Tia^T~TJD~a?{|bE#a-3$$oT*Uy>AB-h@~TcPK^^|SRNn9m2cm^%@~GI
zf`lRKMpU?}q!JzjK%)(^B9zp1x=r8*m%ugpiW3wc#0jAwo~aZrw1eae?UPDMdZbFO
zDl2Q>yO+&6J{s+<4fud8HN77c1eg9)+%i>^3=Z6y9N@fykY;z98HC1<A3xgIABL<9
z6&wXTU@1O;oycR<qR8hM9{S=%jIS@Lyt=05tKC%enmJ#Dr7bKhN^G?%)_1E69x=zv
zeF3cJI?otR^_Lq#9RMr=?Evz6M|}PKEI^vW$Ak8y4T}=M!s4YuOf_j|lt6;-0Eih2
z)uXiQ`8Hu^Zr%r90Sp)q3gh^NjrpcYC|@9^hfD}xp3>Es=C~Hz!)C>)2Qi$iC5W2{
z40nGAeQbQK43r@xnOhd~V9E2}|0US^XLb3ay;g(_Ex1W|a)M@!5%V2pCMM$$SP3Xn
zLjLmvTF#UV)@3Y88Ye0D|7>PpSh6)JO#hW81@=1SC~;1CvAaKf0hNqBRtF2g08vlv
zIzwk?Cj}Li<7s~|cR+@)u}8HsP!%umm2YH26a{UssEhX?To#iIF5~{ISBGFe>_`<L
zbHqa#z#NQtb949l-$wsUlxmXNewUjoK%@Dz3m+O<-E;&-6Nh;|9~ig+ihD(#2gotV
z7u29zVD}dMlHBqW{pICQdwR}zK2edcw9VR3M&|y30|!{;fq5Wlui-FFpn{^}PqV$4
z^vQ~|&7+S1=mVk6^WOHNJ$#r;P*4jvtW)^G!-o?WCkcx9vdyLIqLIpZ;DbCk4D&xG
zWPeQ4;M7#Qj-l&iK#@B6HoXEjKanD607wPt0IcJXl?lMi-a@+Jl9V)jO2-E!jd{Ca
zxc6i#hEHk__n(wChgo4`3^lp<$#p>UlTZ1|lBcra8>(dB^nlm@Gv@2$a0Y756Ebsi
zT$>YiE=4EcMDqCjBucFDr`?r3d(<yn7{6y~I-bqiITe*g39fOdc^SNVy`tzMnHCe{
z08WvLO7iAp_w4L<0bn9;I{{UUsH`(qKjA=mkFbCuc24=ZMMviO#nX)CSLB|HfXmXi
zrKJSDJZBgjx&-EODk6Hvz+PT9cE!O#e1rCypy+C>3XAx?&u=B%4>+tjB(lb}Ch_N)
zsSqEBfw%^>_1@e0ibs#C%gc`wEDP4EXrx$E<3Dc>ea7R7H)l!M*@-b+?MdmlZz7&W
zzk>GF9V5dq4oV{WfQNx9o$SX$gsEXz4-*T}2PjA4p(5vXFE%tZQ_QhwCFmd#CjIjQ
zPB!jaMx=q7TFu<=J%pl@E%B$71jbs44TA9;)zr$Yq!<=_5x&#{Uy@p3v)EcL6Z^pv
z^757I35`8z2gtv{#84-1Hg^D+n;P?V(~<$uyn=pHy9kVz-~O0yW49wY)>DrSqcPi{
zhbvRQCf%vi%V7%xIzT|b`~$nz+eJLW*SIXmb<lv)$-+5}XKw1uCExZV^Q}7-{)yyM
zusE0Q5)++w{zVLHo^^FE!YU5_lK4ex3ik*V%75zAH-GVOPd)j0_}-4MvT)y?mGZvk
zcJRsZ3tTJZ;Z&xkA-szC($rC^zdrxWWG)~ASb3;t2P;o@$P!5>c*o3q|Cn@LS^}QS
zgCMO&#=^Q|%RRD00&Zqfy<9x6o~;hIo|X|%g<{aR0pDPIx#_toiF`FIfGOIBXz>6M
zJ(aRLerVDw$!2ve<T)WdfY^Lt#ep(86h^`BmlLO`E(J{}qQEsZbgmgj0e22b$5q4m
zN<V9fG_!P03BItlx*dnn=f%exWsKP$hNC(%p>_oo(RI016{%paBHF{+^iiwEjLEI2
zx4{M9%bSoO&&C#&K`PC5ckkaC90J0O{V(q4ll?lmhwpO9k>l0W8Ko2z_RksanztsB
zCnUH6J0)inl~Ey{I6*xq9XI@aQ1w$PDea+P!v})%ml6+X2S?Q#&wTvtk6994o}`}Y
z42r?yTcXemUi!s3B|p{mEeKcSE75G83-hAr)4cAVSwXcIYb#^+y5itBnxmjVu~12~
zz#5Zr9OiGWnA%bRMeQ*%To@dsvoxvXE*6C*9_?Y(zkM+DQ^RS7)0P1@x-aINIUsbr
zt;TY<{kqViR?9w{8xWyHkffSx9loagRik)OsaM<Gbv0Fk-R*!X0sW-Y{X3dNgG7Vb
zHHou0JGMZD5mFf}uX2#1l=7x&B5O*`M~2H1wi?yBLijtdAapb}&#~_#cHE}}qhLC7
zll=_I?6c38QYtp%FJx-?6?wdm=8g8Ozj&D_c`w+sv4K6h!x0SE`b2PI4*;Ct*kKmj
zyW#p1o7juGOQK%9VqGnY{>o$2-*Ojogw+7gE~`6X=r*T#iGwKo*V*WLRNA0g-WML;
zlU~C$zSByzUuPWzF|Nm`rt@EW)igg3zBqA$l`pd;lUlA&<Ae3Oi2@rFEQtxuj88<b
zss~>jIN(*-Zi6D+P*J@mFw26A=VZCKxR!6i5@s@{5m1f#{MiWqTNfnRpR`vwc{bR@
zL+8)H+6vqs<1=^L#K5kkgi~7xBFqFG(50xXEXwDh!e44fAXaX|lGs%H!ax5X=a700
z{~8|8YGkzhK2a&5L|lhT{-jWpaDr}5i$(d!<n<y^IKehlPo*LxuJiHdEbJ}KF&i;p
zxx_5`qd7`qkC>;J2L9W4>zl3|_P6(rciDvEgA!)#YD|N=ayd7{f_|oc+*%$~h$>PW
z9893uLlfHPp`n*M7V7WkA8PaPAUP3cegvX6GuK@iBVDhR<7#Yd(}CQaoXs>+nsd)2
ziw$bOzEDtLl`3!HCv8gkxb-a^-f&yyjkL@!miqPU2=3vR_myvv*s|ywdcyG83u7`u
z8Yw*`1b=^ph-RIT)}%T^qdpg<HmptSKf;rK5W-OH0s2IE%Qx-;!*z?PEvkV7`fBZ|
z6@STjGVYJ;A^m7K{hq%a@-TCBHX&5l@Sq^(sP<oA7%)6o{Uwo3m#y0bFrM$Y*xn)W
zc=Di@>d@BsAKk&eR3YXpi*%e9$E0tWLc|+FiUuCwC-dm=aGG*ts-3J6tra#A3w(_0
z+&OhHU#dghNWh|n6)d;0J*&_~@+2@Y^9#k~hj#}}yyrAwT0aD{!rEZZ!8I;dQ^R5>
zr^}bw>Z;lG@k)}CgHN6)XoC%1gro2Yx^nO%K<(C`?)dTJAVcul!_-RzuSq6UeF)mF
zj)ss1qQ!%Ssbg*qYWF2eHAY36sTDgdrS2NBd`Fbrq^_ZX946rDkFQ?@yN-ggatP}t
zAWi@=X<sH%J5{9QJ3ABD_X=B`+C`>#;likr`|*igCQ)BCA9LOfaP2HP#I(;afe;-U
zs#D%COf%^D;18%p%p0nj`gOnzHSO%|-q>qsXe<H;4=GLu6b+K$6+F0Mpk5banV#1X
zQK~<ZnYr?F^7=Hp(tqU2_^PGkuDd?~Y%eZ*&a?ChyHLQIrhf|}#S~P0Ku^-Kv&({8
z$-7;HpC6g(E4>Z{=o|p>Gd^5ax|c)icuq=pyMH&C`N*y=ZT$IcGuBKFLiZ#9+Hc??
zZwKo*G6aKl3JRnl+8O%eVWDARsZc4c0i>PPrS~a;i%9bTv~R_{L`Ylq*^xn^C8)<1
zod!@|e3*L<TmUV_5g~%qSotR=^EL}=|3mwaySfH@onCJ24UW1NbQwz~6lk*H6mHQ7
zOuVyi8m+14F$ifCWq99m&?=i8TkPxHPC9r{k*1rRlAEe&qnvtx0iaQ&*0)u0t#I^v
z^L>!KPG02V<lIzn*{!-8Ch|Ll=q8OI_ANX0{m?#02Czrlb#y91*l_%@pe|qva~}FT
zI79G7gWAL^zTZ(hp=mUw=AgGo3i%?oNsU?RJfLbuUWmU6Sr(xCM)8tq-N+B^!+6B?
zvpW^aLtHOrDX6@wIe|63?I)6+vdq#Mbb+Eca<4!g+8o>hsNMn1Q$SAmpq!x+>aUR9
z&HYA;TjKqxr<nKyFET7I?c=mhNIWR{NH#@+HgY}CNUlhix-GVV2>NTh1&h)9HiN}1
z34F&&?+q&;5%9roDs?Cdp^h`4Lu1H>H{T>RGd_pKbTz|wodmyM6zTKc_5^Lt6RDCt
zRi2)nsFXpYwF_$x{B!2rW8mRKEd?3FpihrO_3#P!A7325D1PlIrpdR)h_z+xb|<d}
zk#Bf-v98XlCH^rK@n$|g!D3ARIt}ItDi~ZteO?dfQh?;$vQA*hM;Fdx$p%zH()rfi
z+T1!B4yQB0MvpvUWZcBKunXl@B@E{op17QqF;sEk1W8%cNd&a<W)6Yx#xZm}*I0^s
zGeG;PJrakUF;Fs!;-$n=88<WTA_@AY)_sa66s+=Z*v|!nev7|7^D%GIC#>xTi32d1
zJn)G@pOQQtWNmT%{aD-HHG`k83M9YLMm2G9aH;-Sy9DW@5+*=MJTmGRJ5AwR#~lK|
zabEX;mPooGre9D}mpz*&fFxOxYU#)rI8=~n-VKT&CeOY>l_m{}T`Hk4j?cXaBA{C7
z(vzur!tu7iZHWw6DbShdw(+|fb^Cx|?ELqOSD?qU_9vFv7?&w|1`Dv(T%;umkb{{9
z@@S~08+!&kiWI;;s8ZvfMvR+~E4PXOgsH$3{0`tk=WPua=>ohQoBamADkc4|v(0g`
z0WHvDvIvrMDp(?6x`lyam{doMvxJ|P?10k>68hUXUQW($@LUfaIg;EDO<Txa17r;p
zKXC(k`XklU{s5D>CNG6-bxmc~_PuQ2<8R|akH2iUE4hDYXed%R2-%=bCm5aj@3V}y
zVsElh;~%W{>>0&VN_a{>$BxEm8{8=;!LAWlNOD<GVHr7Ho}CS0(gD%2c)CatcpMT>
zz+?9+Dr#EDY-?i`72%*xkpZKA9r<$_ThvrE9jLTpgtYSd-@ddHiGM_1Q<t`o_CRXR
z=V|*5k_9jyX26)#f$0atVLWKC;FfXxSZ)1J%^t&zDxlbaNzXClx!ZlKtk{fX%kqmS
z2X-i#Y2xjx;{;z)83DVO@OT>=8nmDZ5qc6GVTnUsc>=7EbRPQuBGc$x?@g8}*v&oh
zBz5V$0P*9;>OSl2CaB^{rfNbp$0{Sg1}UYEp&SM{bxQPKCY=wXvQsPL?40MX1J(}{
z47&dyAtt89k?&Gz3}kVNdOkI~49zjxSnu8S8aLNJY`jxTvyAJ18&%6Sdhq>R6_KU_
z%(Nzn<%mglP;VgXraI5`tr5z%v<83PfA9b`R1xMNJJK>R=#TV(W`)`{hHa25V9%aC
z_;@I7Q$fN16E`j?a&u^&aNgF#6QE<wmN#A3cac$=2|+6#kEhRC!0-}7#w6l>K(#Wp
zYo?jp*DsjqgIb3Oj>{^G$gMq9%cc%#<w(N*SPc$N3n;L4m3x=NH3UO$VN(o=5VnPQ
zizIeva@n>_JL;%-Iocztp^Ct>a)s+0tqQT_HLZ0<dY}0C>}68cyZJH$w@~0{0>wKP
zl%%Em^T<UBZCvInvxCS~dFtxx7-&F2O(iHAH7J4r6)zujJEzk!`;Loh0$5Rhs<mAR
zo^m!P{k7Dh^7&IGTX^~cck~M)x^#%pW#9&Y&Nh(wyYsibm!>7eQYWUjSIlHs_8r)>
zhvkK@?-4527ru(vg+dpSlb3!y&}Z1E$d(W*X<ho^-R1q4v(zs=2^4$OJaH)biGHs7
z{-d-{Vl{bW_G{u_;O?!~u=&1t5z)_VEW4X{$hUrTQ!mZx-cm$Fr+v$6V|9b1nXML#
z$K*mWDycLyI9i3nD(_y(wP>Y}kDfJhTnkm%ay{fVpgzGUgKJl~v77!{^X@UwK#IJm
zX=n&gfkB;T%uD1xiOSn^HBy0}XjEsgisJiD6xHCA6>$1>4f>)CwbQgfTvzMtJg7HF
z>*GeV9p=4S!q=D~p`ac7X~8q+wyuekb5cC(sL-}J1#?WFAVXkV8Z(V${<UBR!*^$%
z-nFQvZi}TLeHlZAWY1F2tH!$|UFYI$3I=yGFL4O$T9epF&$jBT@nB0+y?O4CV?l`U
zjXZ^2yVL`j8s$FmmppSLoDo;g^b$`l&B~lJ(SSg|S9ikumo#6h^b7JX%)rnv1EwLt
zNmKLC$#atXW|wDaR)le?(xp~Mm{tq6U79)F)J-hC#;X%g5qTXqCvgC|eGv>H-Fc<Z
zoQ2YqzXXL6k$2D&Tw+wS+uytZ-X!F^+KL|zz6@LsEZ2|k;^PtWykmJ_fbRa%iBG7Y
z10@d~z`G_#9>PpG4@Ql8LNA5r&T;<nw`5Z>Cuq4=TLP7oh&N>58Gw&#s;5u`Rr^>+
z6>!kUw{$mCMx~-Z{=`5(CPTECanoh!=hn{)X0J94SM>&j0^$t+y7nYHacDS2av!v9
z$AT$2IX72xd)a$?bkfE(qTC>2>b?7l`(!{iIT5jRa!2vQ5p?xrd8alNg@(CP)ZX))
zoABPb37&^HkI?m>o(1z7axmCHBy?;A<##%0D?TBiP}W~OVdTfJ9I{jQGd`@1A9EC8
zpck==X)<K)`v&DDRP8F8et66~)wwU(@#CX4pi06`zpM2|Rqi{H7Nw>tm~G`>(9#M9
zF6Z(6WE16^r#Tl#0*D<CK3Sy3JUGoqrD-76qg2J`y{ahRg7h{3ES=S*bMIahik!dQ
z$Du0x_8LscP&nXPN>Wq{)UlH23|%guj(hNmPJC#VJ|otD`TGUuyG~zA)M9L8-uL;j
zdKZ7)MSk<bv6#wFpW2|V1--ad@Vwo&H<#PN+->As7iU^;r@Tv2L)9>X@tp0U<Z#t@
z&d2>}$%)DjIB1A~T{i)ujX9X_E!$IrWdrEh<n_R^9_T*eDoAKjdh;Wq0emL>8oS$f
z`KptbsttUSDeUu$pfp&&Fhx?rJ4WmPjS&ved@=!*c6IQ}v5vnni~4;xv!vRR>ZKk%
zoKmD*HPiXHpOqB0bIK4tqYOB$1=Lf)`x6U_eG*Hl1~YD-Tq)&o{dJ<1;26qg5H|@e
z35t2p0}lcBiv4I~I}nD2G^0%gXu`|LEGEOK9=g%V6fBX<a$lf}u@V6Pw{y>3f8<L6
zPb8oZB^8OqSDD4TFskQnbi~(u_Q4mA#Wb}+;|e%UtY89^O;cblhZcDs>87h+@X<7w
z@?aGwkvs@05zu^djz4=+kp9_|f_Uf6Gd??eQoR1NC&kr2ds3YG(`<AQ$vqfu&)<IL
zfc=pbc<NWp^Vf^y;KR31v8ho2>QY9KzV=qGzK&LTv>SKBwBJ-`uIIL&#NouD!kTQ`
zxG@45iR$bj;@s%)bC<ce?QOZa?LUxU_X9n{4OY|UwOn#lAMT}PI<2`i;USYNzgPNW
z5P7p#M)Ftv^B7B(lCJP&li=x->?Xk`VcE}L9^^lQyU)HqW>epvdZdd8^GP2wpE}7p
zY9%-I*9Se%j;h~|cfWo+K8JDfbG!)V_3q-+F1ag>gy&snv0W-Bq}N5$w6Z(n<5_EN
z-IeUR)9BPj67&OGDeQT4RDD?YBzy2nc<U%!G%GROw_d`1;*_GU)hjM#q*tGi42*95
zex;>n67-x^{^6I^g4&Tga*^n`6w8ieSmrHt224svleH#Gp}Qx@@Z}fb%{{lZxqXgC
zeIyr<p+x5qtvxibP2Nm?dgo{y8o2F~(kDeCXOxfq{S9Mkf%_^AkwTWdhu;;pK3r(^
z6)Ag+t(=8l_8;+)ogGv@26E^FIEZAA1qh)o`-td2`YV1X^5qKRPR~BWttv%oP8zMp
zdy_BcQmvlI13q%>W8IKK8c#h%8#8=&^qJy!DqT&mtM)!r@(v3A^Z6pPJ}>dk=i}9j
zP3KQ(ao2D(AJ}vdJ#VxRyN|D#lq%3CO_m}SxK9sn$UOYdZ?Nsv{k`uWzwy@_+9xhm
ztZlg$mu)&dQI6zNzjTQ#A-=wOIJ8)2D~#_xr$$7pkeGwrL$fj#jFc@%@ksWM<2&~d
zKU1DRnU*=HWiVCZyU~^+1CiIhaJoxI0ImRwd3(l>glN3mHmTr~eZa)TInVXe`2?7p
zQI~PD(qfLOOWo!XnT!>MP*8}2`bR=yq9yz>Z+adYISrt^hwLVd&OL*}OuKotS4i$3
z(-7GQx|i5kTgoTBZX$hsh=Zq8qTWHBj0<oR_<>Ou8DLtgXz(psf#(^y=}>DtSo~J9
zs(s>2{mR2GD_9=3!@pMEO-d;nv;T#<x+w=w{pJ$=&6_V3<Oi4Nr43*w2wkv}q3R;k
z3Q=P&RKQxSqrv544hAOFHO86(Rh_ai%Zi(RQY<U`;W)hIJ*l|@;U`YuS#=NXe;J=E
z&@D>0k2u)?Kr9%{4Y~o=0cpP5Z3eD0<aR-hG-$L%FsX0aRQeVXJIl&h7_iB?x^z<0
ziU+6z8*Za+&<<=6*EZLe3E*x4YuL{Z(G8|S@Gt8o_;>D!Fi4K2Z@Vn%ro{f}SvsW@
zKOY$eMeP?-wu>KKA7nzrU?KfdU2O%pFJlVg75uq)DhP^*P!39i9xr54M`jx&SHK9`
ztV77+-maD}1Dyk9rt0;XCNPPNN-H+*f^8?pJ(8*h?R+;}(`XuxdH8WMVEgXhyLT_w
zec1*!n3020u3nKt3Y0f{mX3l=*~(!Cx<b(H0?+Vh?ymSm;VnIDOiGW*sJgfmAmbc>
z;55i<P`*_KCu|~URj8iC$EO00;ts_B78V=?E!(a@S4J)rHa&V5aTUbM(sn1ME4MqR
z-hH8Rk%oQ+uzvJ*p$`1D&Rg%zp{Xy7-|TsQ@ovlu96H-oR9ZwBF8U_5z{W9QP1qR6
z3p~Vsd-u3>Wx1nBQ%ldN708{lXNUzhx}P0wJeC~Ad9I}sYAORHq~<HaHp#HJ(#Brr
zLEQGwhpDi_1J4j62iboL(fhOh$orJ`l3i0})1Zlv@6x~Y2=|$yAAY3+Awv3=pDi4g
z8^Qpq_6q2>WuU13%Qt&qVmzfSgPY*{hqG41OB;UWe6dcwLYZg>6(<=MzM(^3J;^rq
zoNS&w_Us8pPU=(gLW5<QahCm=bS2*zE@Bo{(m5GrpC0FmXgE-LMKU=~djD5S(OviY
ziM(ZS4~=eUl1j&r5BWuy*1w<^cCxkV5jnF)y>bvng*fg6MtnJjZ|L@WeV1buU-rig
z(h<8}B9k6_|F<}Ngk#XP_E1Xfr_%J;`{Va;x-fEn(#+q8aMBeHqA@q^=%`nyLPmCD
zO%u%@T|?u#x`^5sPeM9#k+)th&hJjJ?iP2Z+2A+I$zAAqO8$nI?6n+jH&#JuIqht8
zjpOLU`QM{bp{K6eHD1iHsB5w<*yp)q#U%MF;uIZSwS(w@I)oey<4d0MS`H2#oz}6Z
z*?35vmse_f+3yf78*E`QUQH%LyK_RHH<XL?6S0!K^rYfD`aRw2Le<Q-yc0i})^TPi
z)6U-Sns)9lc;6QP*!}uN!?tRPC88#ji3fJ&_0D*GJ25^S8hO0yAl>5KK+g*Pf=3Ya
z_(c0Rrx~e7Qy=zuC}E4$<U3`Uzvpy=rvQsNPM=nS_=2`QNMGLoLyr!`Ea)e7sAhC_
zDfEXqHA_1p9i0HH0y>^LVD$vUdIfX~LJOyaL~mXbR8o;o4>_Ze;TZsn$9H4+oy)<K
z&1Rh?t~tPMgYo#-Q%Ygx3uE6<xwc!9@-R;`aOie`CGzTEaPs)31wUEAh}L+(<EC)+
zs)IE<D{FFm{AcTEXj6sGGlL02X7t*^IC25F6o;b=+n;FHRtb%gd9dI`8K_}TCS*7V
zhhF%tTeo_aaysvTE`pB-$JhnfEQGAoFCY4D=Y)4-ETsElnht<*0Gb|(K0JN;6m+_o
z4=PZ&c3YlK2ZwSRC=Nvvsc9&b3=a(zOohVU4jUUD^6KE305jOw?=Alm#Rg%3Q@HN&
zs+>1|XglpXVJEV3j0D?9{QA?~5QP6Bih<uz?eb+mXmU*jvu?g87$K+WEyNwhULXt#
z1$pL$!%oGoqWNeKD?A+ZsBdxx#DlpI!#Kd+0b$U!i|qK&M`Z=(PLzY74>QlQolCxD
z;Nt_h;Uox12rB+9Aoa{Q3z$67)E?>6)*iWoLI^1g(h0C!0fV%$hR#>#%{%X}379o9
zfW@V{+n`bxdOq{uocdglfttk}&Jd_say#=ogT;k)ZwC5ty4qy_>V7@5t3sSufVk@)
zAQRSj8xR~|F@R*Rj$ibXqYN$=i{3|7Apl<tYv<~`qpps>{a%FGTBsjF2F^P?+vZ6v
zS1(@}M;cG-D@T0+q6&r)>5~JxkYX@89f#bdrzaZ;)YxbxOjH(D2XnRcFZ&@5;^xuj
zX)>RlO`36JzhA$6@MXYB!JlrtVb~37z^D5TyfaA8rpn<=xs2z@=79g-sl1czWI}xe
zvwsa%5AvfO8QR4BIK`@enEy2YhEhkJTBVzBP{3GG)!T&5%_n3t+nzVLHr5kTFm-C*
z-Xl=njqy=>GpFH_p@znE-wJ}f0$H*1?;7rvIblA2fR3d}Chel~bm8(@ve0rhzrk;M
z<>j>12M!drS!Q9U9(YeX?RE}$7V%$xc;t2#;P#OphJnE_nk>CN_0n0lAQNHOKZ0$R
z7g?W=57^hkC9tTXKhmMA`RkA9Do80pL`XN#ipEKQJm8NrLJ?sf2k*_t7c(SC>Cy1u
zq}7_(sI4+2cmRC=OtPZc^G$jaT_QQFcABOti<ZuL=1IF{D)fEW#|6W&ROk6iSDruz
z8Y)=YXMPPq`};j?`tIa0hj;$0FOexhYRAMg4EIgGbL|&z{=>DGJZRV7$NC!WbRnK;
zDi|7eOr7v<<iy#0QX>ETaI68>cO&;%z-t8PtJ;#6`~*7aKYNp04e5fg;*dH9f3%M>
zf?ZQkS2^^5qV_}ZBF<gl+b?jfVb=xUVmr8+)YQ}#h%0$cQ+mjP4ILBUCsa^TaqhNl
zOIAbcDbVe4OeLK4z%ywQY4War8lMn<B0{9+I7FaJuwPEb6clKnWgsG_ADK)*1r9iJ
zRt8)}i>7o$6}AACFx52FJy8GpEK<@D(*-GV5-eC!W>Vn|r{G_C)DkyjCsNYWTA`z-
zSL>QahENMS3x}j+pY^3efL7pswORlC<x8O#5}p<TR1w_XLEstQE{N6t6MMeTGE(28
z_J1(1O%ox!LZKl7;S6GwIS55vIc9-C#&e;q55jZdR1R2d2u-k&l{%>7rM*hntm4jH
zQ-C(Ku{w`D$XZ%j7O)-7U!euLO5v9l(8q-Kj)8riP(KIQtJaIUu>m@ma{FQd-2{h+
zPj(1-t~n5z(1v}`<yP8z8k=>$ItUXO`_}~V-{+uz3R+;|VE>U<{_SVxHy~U_@c8+O
zGEX@fEt8Y)^KijPAK4i>W8|lHB5822Xg(!nWH<hH-dctcuK#Bx2@4MJg+L6ahS%Cw
z=FTFvBSJshM=a2c+lhZG`<a<AB3}%?Ma>BD@E`tult2FKmE{j#ub7?3Q4XMkLC44L
z+j$B?;jdn?>hd(og$G#0;GX_K9|5L_@zG897$sT!<WDAzyYUa};y<Ia|9GPV{Qv&u
z3Ph%nSIy(opSW~x)amA1`|CeK24-#HIbvA!>vO|Nv-)xG>N}}d_z`8FeH?qjehOE<
z!_lM1nm_8FAK*igqJkYT)QOY5-6u{)PT{0Z?)*GUk1bqDK>^Qmcr>n|YM1Y<_f9-o
zc?MFV?7%;+>Hq%PfzanG*3_t9dp-ZnX4x^2m}waqW5~7t-9tGbJWxiubM2&Q@Hi-%
z1Z!&}rjmuxwM%Dlah)SP7u(q?lQa!Ihd5<)?bt`ls=a`JOHV^EeeSk%?WElQj+*OB
zyF4zSgcvS;CI^Bk)iV@mu!?^zV-Riu-+`*G1_(H&4PlOr^I#j(MxE2J_Y#DdQs`8X
z9Akq6dH8gHkNht&kaan1&tL&H3(1L{FZ=Bu%lYTLjnE5zHe56d0IL5Rfd!)C#jAtK
z{{LNIK|PyLbS>@GazU+)sDIeVnS|*B$Vf)cA2T6tsZLJ#{E_cqi-=iAI`BiS@t|09
zvKDlBD7HIyv-9#cE~33B(qSuvf(~Ouy`U9-r_~iHflOm(H+FgdY?c-&$$}k2ev2lP
zO~`H9^$l=e9TZW7%?F_^EnUKWDLFP)2Gvgu4Wqgxp$_f5Sps+kwqNM%?uGg*4w#B(
z2y9B%5G`c{tV$d4X3*%Nw;MIw!2vYkmuSl{=v(K6-O-?J0qvEQS&Tvj-*IWezXvno
z06k&T&(^lHR*CmmTz(5vQ*A`@J)75%ipXCZEcvK|G|;kXI5Xr$1P69VlFbbs5~dz>
zUNAx?!{z2l+8%j0UB2xF#GebcDBIpzE-&r5DB$~aGm??oAG);%*BGKjzi?}+PS(eH
zwb4Y&t+0_$jX;x%Ec6Hbh8;<cfFyYC96x^mY!Fa9l>@GPV<6~~6UTYmp>^xV9Wd^G
z{rWX|aTT2K-AB&dexzn#V11`6w`L4lRYKh#J+ymYJ`Y4Z;JQuWm%7q9=C-1>w|~xb
zQa0w;zB(LP`PSw-w~)|MNmC}+VA0-cJcpkmOng%Ac|q^eKA*u`-lM8=>1oXHXl7k~
zrA|l-vxE@I8Ei}8D)P||gkW183AgQlzERI!#?06b*WHV8TQaNj<oztt?wwgTD>Y%y
zhV6|}sUq02K@eIljGXu2<Wc7(G#Hkx{Gdc-Wz_SVSv=&N>^OT^GH`Ymu>(o!<hcQ8
zx5Y=);cVeX(>#66+K0DE$UV>|r(kMe6`eU|q^p|(b<WiSoxY3d^LybEeb+|!mtDZw
zNb`_92Y9>*-I2jJkF2fy3~Z0RObJp-BcWG@1QOxQ?F|>oO(>S86&2~h?lzLs!ON<X
z;tya>zuj{6waq$kljPRX*#Ti_82~1Lwg&`>18vEnsjJJm(=G;SYZ2|Y2xJGE^U=;6
z9zDiw)nB$6(f+Sd-dnDI{{CrbZvfB?U|0j)rzHO2322`{Q|RgrPQ-5fokRW8{N>|j
z*_Qkgwn{|}eXvE%=D3Qtx}#$rYt5jYIrIQaP9Mkg#Vi@^cp<#EX@K`7*Ktx?z6Gcg
zqFrOqcY-e6qlX_e7ikk4(1(CY6$2s<D)Rz_DaaGU-ME}i5&<nVB!DN?U=2|-hzv|-
zV7O}4N$ihy3x*JLmVU4GbZFs-hiSf2#ElJ{Z4-l@27yPg@+_f=st8JPChxC#^fF?z
z<d%#BpAAHCZemclfs5CI#|7&Y?Lb%#`}%l7hBsG5S9AD5ojGJKoR~2J&HvqC{Kp*k
z-vwnzYuJCGE&mMzz&HL8CI25r9RGW^zl;0-`vCq+T>Af?S#n<=8X=HJA20^aAGMUB
zy|;c$+O#QxF$rS5rk2()gd)dVXZKdlI<iBgm_cm^Xd5&`&qb}kGG*vMK|4;O)<`B#
z#tIFPA+LzS#$-B%{%VIXs)p-}lsk%^BiQdyn<`{gln~HXU%*cA@gRP4K-&Q7XsoM4
z+l=$Ni9h`k>pDMzdS6gOJ8Xy*KQUpsv&BL?l)!g-kzuDa*x@7*Ii6tK4o%pC)B&75
z;BAtJkT<l;2=;A6yJw&V0kqEwv@k<oBVisk?+Eo;Of%vH6$Mfd0dn+W|Mr#+2>EDB
z8xYHK0pSp~py|Gigho6D>P2Xy7lOWhKpG9&C=9iJLX!w2Uq}E2UbfH)jr4rjHWKx7
zOjM#yA?Ub8TdV$6jvRLKLfca5NWn4R0Uo#vK0MS%3ww|sBb1?9Ajs;F!zg?I+L{4|
zGvVnL73LxfV~e)Mgne97VLV`K<x1DdO4u^$|7!2c<GEhfc7GbwT9Q=83>8h1A!N#s
z43VVFGE@=~k$G0Fq)aVKg~*gKv{Et?LQ#f{5yHwWWXSAXkG0pk*V^yiXP@_T-gEvq
zpYzY&OMdtc&-2{(eO=dm-R;s(Nmq!3)7Mlq)LV>wE=JmDFO>y@(st*b0H{S9(M^<2
zoUT!7*P(AF89qY$cz9IM1;sSnhNldx*+g_96K<sV`r7n9@-RtT1uE-1fq~j6K&N2h
za4Z-AT4HSMPGb0m9+0^n&gG=**H`KBGB$g08w|G~S!sxL>PUtxgS0O{uaVAsJh374
zBE{%_NXP|vQ!@hsEJy1PS_dRp(2$b<4A%2p$WL1`O#Mz78Cerg@p3qixz+V39|Rm{
z&L-{NvX_@vI6eQHiV3a_IbS&<yg!&-77)WiJp1<tWMmxdf#xMs5C+pdkT{mZvci9L
zX}LFr^1CkpI)_<<U>Ua7*aV$Pir}sNI~RII6<u6%x@^S!(G_NbGMfBq$l?O3`10Ph
zz_=R0*Oa)oWuuANAA+%!G!XQznY@BR&U8`VBR7A%TV|x$<$x*!5KdXKh7}>NrH4Uz
zK170w2;bw#aaggQ%aH^4fB0~k3zkt1{%C_{G4`B|KgaAN2<E5f_MS%qKsfw?#IllR
zw6@ugXt#e7MSc)3?!2b9HY=fn1uYu(mI`d>0Y<2ZRfbgV3iG{t_sU^L5rB~1oU`V$
z2l56*_z<1a)U+?2JBvSX_7$bM&jB_#<*+_J2LqGRq|{Vl;`G)7jq0Hdr2=U2xk%i}
zL$M`oA%yBdqEZeosLRYg=N8cBCXlLa89C<&TAa0LuJP&0WfDVQ7!>U{0rQ;RL%ht1
z)j5{Mxx?@nBC8eu-Q{WwZJ7f*F?jIdj)UWtT1B9*uRw?Ayj2fU?DDuEX$x4J&7VFg
zau#`w(E}`$e6(GskKE#LeKYfg-|LbpoEMy7vtclTR%THk7Xhdx491*;{fRkI<|0f?
zSegx;BF`~bQzAn{`(4v>b6b7={e@RHT3gZ56hD`^`0Vwkr?x~NKYrZM3Tb+tATW+n
zfrw&zNHo*?q~Xie1ErrR-hqJ?nwAbNjop%A285!Be!Bn&s~$~AAe%b~!;PUPM_5;~
zprAa23NrvMAIv1rJOx<xWeqHc{b8og3IHi>Y9WFp+H2g;IQtP97!UyRk-R!6Cj9E)
z*wbv+*x0BjCx?u4ECwfw32f4R&KYD&GyAxCbD)YCM<-s+G+3*H!0<0_-MW?8Yrd`1
ze|UIU-}q@dRc=z!oAop_LU?z*7H()$3ji*a7if1eDGfgYtlX0L?xlZsD5MxZFu@r;
zmEH&0oi(~rCu2RRSvPBlecpoRJcmU#_~&9}qPl@_PC0!VavUt`<j0SHi>){gz<x0z
zB;>b4Q$6qcHgX(J!fu=DuR98RCpZlg1tNTWiv0=|(Wcf2S1E~tYI2}pBm;~Xvlfr5
zsMz<{?zs2h!4TXG#QH=awVQ(PZq6Jd>>zNEn8TWg7>*E^Dxwd8n^ZlFNn+surxHw@
z;VK0-5Vv(WT@kfr&AhaQ<ae{_9G|lDr{v^H6l!QFH7T<OMoH2UX}|uezC2&FJk-h$
z!)c#$1mYtvB6BJ9*_BRdG*$Cp^EGN~^i+S%;zrNnz8yPa)at(tULAfur0e1$Zuh!N
z3sh17im1Hzb&}BO5alimdi<8Yg2*(KVParN#@0=@@+<3}%RwCrDme@fH615kKeV<I
zu9Hdo<&EAzgBrbMk6Ock12McJTtx5r=3!3o5>N#bIjFT=kR|u--4<l^6|#$8l_xG~
z*b6+$tcRU{MfcqkP_a)C&xoXiyYInU;zf;nD46N?%f-oKF!5+81@sBJ*d495`E<Rc
zeI;sPQB^ir8U`A83LTNTBW~Ep#EJtbXJ0|)moSH6IdV12-<Ox7;79P);z_ctNOfuz
zmLd5yjqhj*)C>b@XV639X#%(trSXaynM76Mhff$3oZ)yDFriK(+zTFVBM$i?NBZ`#
zl9Cc(+iv+4?{^$^+-MQnHP<Jztd}UggcC#^TnRCyC&FK%pSEm&3aKKg*Dz&5NlRP1
z5)~BQGpm&A3-YVb1xWm#;gdLztRo(2#sCNsVk$!vF~mL_TlF*!a2WE#?0{IgTYjm$
zN3>O#Xabk4FRWdjIIj%OWg*yQ=<-^}bvoEq9mq0=X8^H#AjCHwq#9iBB9l!hV-Yls
zx1OH}7gi>?=ZgCh{3{yah_YK5rqW5I&48%Ja1q2}o%mcTe;{EM@h@w6=IIF-VUcVA
zyqptg<z~{(e!w7G9dwh!2$^a@oC4r_*F^lP>FHy@$!8n&x$h|CtDtrVd0b>mx_;rT
z)V#b<vz&xiJBM-}c<%>^#jnKd4~F-|+%W6YRX<OhDG^Sc#B%~Q?;!j{i47IWxuEWR
z$~VV~U9Cm7Bdp`}(zFZU#YC<<ZT=9Z84z<jjCqN}LL7k6N}ciYoKEu+<`ZZWpPJv@
zNTj~K;YvwQ*Ty}eKOmW84mbkIm&5=`OGk$k>#(;v1^pa>k%_jQM04%|tPCw^07$Pj
z%6?+>j(jM&KIaLE6a~|*mc-$Juv@S~OM;mPAawx;Hp(O0m^m95&*hyU290FS34*Yy
zVXl20E|p6eISAn=;k9dl-cfCIi^exSHRe}NKXMIi5%4csyuPTHap5+G9}v5|cE>g=
zQf7bpI^+%$U??d}iJHR5<qHFO3kcL``}pcqRKMlq;TWWE?;lCEO+KSCHrc_jhGd%n
zMfPJXgUM4Q2M|)W+ztPrZfHn4Igm9Hy=-$kl?^@7N)S+u%*<&v9B;kO>-`j9^s6s0
zJxdH^6OSeP2Ihz%=a|#IuCD!D1gHH;j50~Vl7Lcc)8_THnYzy}1|Z*qyJ!mViO30e
zU@_@MORmWAEMTy*Lxg95hNJ_?y)71vkGg()yT{Ta07|=@BgPd`a4}QXF|fT!Qd)Qz
zS|Y-e^9URVT};#3TgYOQuo8&BS9(rwIJMODti>pt6&s?Tu3p2MU=u-LPvX-<jvL$@
zp&Fo@Bb>^J9Rijm01aZJN_q~+uz{!f-|o=4(NcN=L_{w!Uz4r^m`CKjAUnyN41~=l
zIK;J2lK~~^d0#P82*Y`*T{tKU?3o)WgFzIJ`*69;dRPPkIO6$TB_t^j&+#d{PP`X&
zSeUV8N3Qo|agF!!BWbI>=V&0NT2)8-;bD_YoZ(43GNu|dCY_42`-1|G;1W0U)WeE=
zSa_0u)oQP{&yMi&0!oB;G<L@)$-$x6EWujrBvXaBg;<Q{(BKVR1U!wh8h^p5QKPq*
zM2_=QVF-T(WYEs-+v0t?a4&@|(*tf|(Gle_9X2Emw<nxwuAkXoHP5tBXF)nLmrFxi
zawapKZ{}SCZQnq*m=f(CU5m)Xv8YRCsBMUYa;8}g2U^QXp8<*)yS6@PRMNLK*I<wl
zWGG__I#kh6UsBY)*7QF|`eX^1nH>HkkH`2H9xkq<SV7QMKj9W%rV$03Pu>;r89fnp
zu&knhnqs%|)4-7x{vu6r`CE&87SRgc(O*ZII_gcWij+?isVT0cChsI1#3oVbH2W1?
z-$Ps~0I4Aq0g7GaY-nf^DHFRCPOwJ!VaVZU(4c=WB7)bi?DUW{QZOAPo)6r7d@eBi
zknjk#&wWM3V=2eeCX^bCaZOVdZ04_iLi0J2h_ww&KfZge%ZrCilU*ko9iu)Zw>mR@
zX??A?vU|)=)JxUt9o(tqR(ti$fgGSJKmV*BMpQOdNEgo_^AhrizzW`na}N#ir0osP
zhy=n%CwvT*+x!;17v%OT!NjQwmBaVoAaVG+5gB@)hbOiJ@sNr9zV5S3l7mc4(=2B%
z{>E<j##BAvZyxphE(yXtrK4g1&PHgv;Ze@TAQ@q#E8clMC1FnfN$+9cfUni#6BC;+
zc~O>b=zrg05xETdWNdq$p^Do5l9GrZ$N6|pA;8&KHLks3X4Nj~(b2LbA4P5J0MDT(
zNlC|oCyw{P_)Q1?Gej_fvT!lKf)lzY%UwADPWNFfOt2mI7c-gpFrN)9u`ozDjTGY7
zAtz$hYl!hfgB=BV#E}RluPVVr$%X<d8R0jbVTA7jbVz?21r>`BmD`HS)E2H$1H@Dd
zLUA%Qt9d@fg&!qQ5)88Lg@q*(|JP+S3Xu9%3bua4)dwzZJujDGa*=|di@NGKX;(r{
z$=@nN$((qgFN!uk3--+F{h+E3zy1qx6`b01Tyh%pEPnTZcFV2-aGdq1r3lVk0HZ20
z+#(YWtjZsNb7!6_r6vk$@V8T}*T`@w;`)R68AP#y)|x1!y9vb26fg7<aU3Io4$-Er
zlujlU$@L&&y$*4*3}AIAHUTJ&>iWP31gV{+0z@-R3y(GuX$hZ8%-B#Lk;msc+21vt
z2fO1YoS(@^FDtRwNPiUOWf&P5rNx~<)FP%hBol?m0})CUERv9iG(~+V$sy)8z4@!7
z397RLH1UdhVb2#&8#;b8CH~B?YCfDbO)``~$**_039O1~Fg8-6?lU$d$%I>2Cyvkp
zqwmD92iMww{(UT|CrDQbh=2$agYgw)m3z%r8Efk5!mi4Rgo#`fLI80|45Hz1gArEl
z$jdfZx7xw(Xn-shl2H+P4kiUXfn{b23U8#;!x5cn1NHo#<1{4P!IO~5Jq`1xRFGP0
zfX?CWo83TXc9<%O4sVlhD+CO5v*X5Fhdx<1!Nbs!WG%=>lj@fM`=>|ImWqvfLa~B;
z;QnJfYb*O+K>|BNZ&D?&f#3_o_k=X9;dz;(gt^#Jxr+uYG-87Y=H!rkj}^s?jxpzY
z(*}uA>_WO#!k>4rON3gAg7MHs=4`+JGqDAOK6;ljtnX3%rw#P%hVmOZOt5|N><tmZ
zgy6Xjq7mpmcchrD92&e%gOwqi#)>5#ypDruIQmdnG-A)9T24Uv-4E`j0bs%tJl#vk
zEs>Y$T)LFpfdO)Mo>`FIBg50e(McSuJ;uwJf*<$y_v`n~6UQ}FC3Em#-zCmuEQSsW
zh0guncBjZP@TJf?zwu}J@Nqepk!9fWmA!O_$GlgJs*#f}r(!LVK{zB<KJu8>tFLe$
zD_%E+h9Kg0VqGB}1pvVk!HiYtqCYtxrW$oDfjyA%R<HxH&iTSf>=+T*ZRZQX&ZASO
zc&Z2(e?J21X}Hai1x_5xNCylQkp#GbW`SUJ#Gj3@S&pgjk0i017&Q{NO;|-83nn=g
z@vlL#Zs(Dsd>WF|7neUCB%}}Wk`J(4c_X(<R*Kx>j<?=AkaLw<xdALG99Yly79}QM
zZVgZs1DT?^lyls>SbQLKai7wvnK}?kytM^3l!_tqhA9%A30|6!OMB~5hnd&?u5{_X
zTdumI9fwakBs-XWd!X-hagkQOA@^0W>eW{_aB39&$PTy2Lb3-CF-Xe_vajzB4=54W
z^qU2e9Iin$_iB_(ADD1pqP*%M{~}J9<G%~j{{g#Ra~ER!{lA9{VsM6(!LSWmn|4m0
z42;6eyg%5Q2k!4k#YqehYv%JyyJ`UyJe3%w+b(u$=<_H@=S{N2<7S1yL1PYzig=ux
zwG{;^dFYlk9CJbVL9L7$D@$#D^?J1ii5W@7XsVMLEA96yI(gBV<Dby^U|m(KsGy*H
zm9Zab1hKFo{=});Pd1d^<%aK%0`WEt;*b)CYy);WosEy~-HQjk1e1R|CXW#`G$a8<
zA(u`Z<8<arZqg9P_+<14lKSv#!FMC&YW@;*ejX1d1S_OtB!k3DL|a-dd=;a~bR&dN
z;+p?ydKu>sF&Wy6GagUd>5LVPchIbI+E88Gy3CD6P^o|3_ZMzIxLKQR25UGIkYeVQ
zP9v;~dO(3I#$#AOMEFOvJH(wF=AxubpTb@+Ad}9pkEY-nYdF?|TA`jR=x{8ggh}`$
z@#}wqkU(ntAep6XVq>`<KEot(0w$`ST|`EF33&if-5?(#c9nSQIRU~bMXEE48Z8=8
zFQ<^20jL{JC+n79)<%0SBsCHDm^XKKG)OqJ0a^(m+7P*9q7C`};eiUaP{`f8+m#XU
zv@!hwa>7Yeam3&@gd1=VnVcbU^*3LVzavx~UqL(pGxg<HEa5m1!CX6Hs~!Bj-wTCI
z@M8dgdvS;b+mn151ytXHCwkKHW{EE%fKKBv4>ox4p%x}bH@7mPgoqt0Tr@ZvbFob|
zhCVaE))>aP?=qV;AXT>fDHh<>$f}1JAJSX{So{N^MYJdpC>ktpRu3E6Jdeyf2%;tK
zB|)v=5~>fI{aUED^XKL$tm`wsUPnvMZSh*~MEm};=>?DAFqw(^Q~3ijNpPkv#fz|l
zB;L!-4^&J|;|Y3!8i;r=BEST5&nxr_OyivpKUy4Xk5f{3U}z~%Rw!ZzF^nUEa0y`g
znAaJAa}=c@!2yu!Pc~?r9DoND&w&G4$Ps}tcvdd~-&Q$xtQMy-w=z!b(;q)_8ByGC
z=qshs93i*<C-Dj`nS@DZfreoD<4Gk0=4Sd-gtzXBJZg#z6~l_fGEQ&XM+y>vrP^rJ
z9NHPd3TcN1M4m%l*y2bZ?!&X{F*XnZsz_|a88Agga3Z5Q4fA#~-vPK?Han4oBhG5!
z&5^!%GLeCRT7cFkQGMs~rHa=h7h9npexGjEh~?f)F{696E3WC(klfEyPjSDu+NQxA
zc+}Gy1S1F<*hk@4F1{EaeB$$S$OUFbzlrfi9jPW!udAqWfB*so{lE-J;hFpFRnG58
zo)+_NQHy#$d2<n*gXh|vCk{8pH|31~qTi?ORp719lnZTxb8zEerl0x3TP%V}49kE&
ztN~`f)hqBvP;L60<j|122|0f4XN#|9KT?YQfBzvIl0-GGdaye8^P1w@MM84yQP*l;
zb>?^?`~CsXw-F0;pV*OVfeom<LKQY_%0PU(rmy1#S)%H(W5<XO8W2C{4i8if`xK8v
z1Gq}pV!NiU@j$;XD5<FT)X*Rd(?V;rZB5{8Hhz6xjBO)2mEIqhTIKDd^ZSR!7>Q{7
z|1IWshs8h2i;$4P5$i7TOGh|f1TivpgrDlw@=9X_QQ|=`;tC^y7bX=O$#%koH=DLT
zp#HVUK&Q|ZYl$k60DJBYXDDO*&`qvCg7F^ONORHB&;ZlS01hdA(&yS%fwUFyeDGt`
z+r%@ru%CxA6G>!pl+PZ2l34eH=s_O{EDOj{J4v880y3gZ>Kn0M<!wxKpcXKZ*i0-Y
zT|Zx2M-CX0yb@lbq~z1`_=k(nVF|HfCi=xjOtnA_$<sjR4+;^9em!i(XNYF-I(Akj
zOyHUwu~zWk?b9pnVX1mt%EU7oTA`H&cr3QxTR#5K9nrFh)oL08yZh7cBO0C^MA%7e
zxKUzjo;npn3|TQ^)+!1|kvVdg;KxM2LlQacc#k?m${5-k@T~rgq&jM8a?b(XUu8jf
zk;Dt-BC6T$o=2;_M~T8yPV+AXv^X+$c!o|oW>H>^3jXu<i~m;f17z7Fr>|ygW$bl-
z0r?IW#gp$-GdomnX1_Pr?YFaLVwA|*_10p_u!u4vf+T|yV_6vzg^s6;gJzPH2-smz
zHs!7e9=-8iyw#gxZsa&p&=W@4RQabPmSV$!=Ox>p@eley3ZE}jwiLOCq}BNBkEC_s
zUuRf<J6Q045QP3`sog){3s{{r06GBX{+IRwB=g&#X!|eh1!(%b@2#)Fwu!~;#AZi?
z1%lRQc@R>Q_)in?*a2pY1SaVCx*NCQ{u9NJ?)ZJ)I*fCOCxbrGx9|Z}4oMAoc;9)*
zcK&527}TH)&$Z%m9?M!w*`5s#yZ_X2P=gdL^Wu9}<qz;fPD5r(gi9o=ZzEG^02Ks3
z#tJ7GUwR*QLEEA0>%7h^EKh%xd0G7}f7^^{1$SBePu#m0?7Fj)mG1w<$#O!G{^=&+
zhjIl!KX9Ph9C%1V{=jsCs)_#Wp3{n4rc~o{m$qq(+4DT<Ewhi)DAAAAj_J@;WD4Lj
zG?HVqJQ2|nudaCfAxrq@nZ=$4YDu-L))jq+`zrdlcFemhv=q9(%bA$6`#PUXm1NZ`
z8}$5q*`5B8kpPH^2$>q2mnR9>U8*;{si~<2HvjVS`A|@f!xy9MCX-R9LJoxz5=N0y
zgY6B>4n=@1Gk_lpM4>H?1_@SdCVxamdHL5j=%ksE?KvA@#s|3HTxSfvjg5)f(Njr9
z*}G0dY<F#49X&fcyGb=3)}_fsq`t2~gzJ2TnO_p+D~>GL6@?bSQwrrSX<PMPQG#+4
z-TU^HZq^k|%P}I8{Wl*qt8HYcYhtWx(RFTXyyu`mk-Z<D!AIfI)6tPPGkg8{Y*+Tv
zr#5W#KUbORFj4NZJ4JU1AG++k=Wq}+8*juq%HS^)+C<wKn^qSXx%^OT!<v?MsgKIL
zb&qW6rOO>`b)ODT8`#-(9j2&9tuYr1FyPx66{P#?cHHjRp{OkHO_aefJ~kWboQPK&
zDEX!AkhT8xss;3>XQ;h%FVOE$EgzVkvA*;+osSaz+ox+jZ-1>2r8FF%V%=TJ(Fy<2
zHRhok+}t%*V|*tI@P#|m)6?0I5~$71puS^ALFR{qy9^9I8<T;}5o)c2gHz3EdEP&w
zC+z~`>eilzJfgzFTQP+?2zA14kRMDSN&FC70&;qvM=f3k2)k!`_~RDnC3M^udN^y6
z^kbhq`L!wbDgSxrnMemFIpGU(-U)>Pyws|TN^dzjHFHo`D7m@0@fu_cpkIyyxiL{s
zMtz$Do>zu2S*g3j<Z{a<+wRvbc1?z|7uT2%wteKKe)nbPMs}WsILH3GW2#y#%*?*9
zr3&Jxc__+HO-;Q!|K-!C{76ZdIT!nts<b$I^qi;|6t+nYKJ*It?4O&LXM(^ih^6y`
z%%hq<uBaL%d3m_>EErsI_(KaYIXTJgv545Y+e@u|_ETSX|EHX~39seRmWldACFfzq
zS+Cny>7$dcq@~@wv~5e|E!L(FA2wq?0*lnw(b1s$8j}K}U%v)oY{hRO#!sF<7n1P$
zxXxqt+ZS7~O}iU`ul=$h?qGIyyQ}9+Y4&u@(r~2R`}a4KO*rySEg8oesCGpc>TNg{
z2@XnOa(cQPvR=Zr*R{8^gPM^Z$OW%ZankeKFWKl)I+zrED#zel+D!*0b$UFahh^Rk
z$q5MwEf4}@pgj))85>|bJ6^+F;X~G+n2}a67o=mi0@B|`OH0e5eMT~O+EJmc)T(YK
ztlZo!`#|pWZu`1WskU62q)iDnvV47<I|olIDJj|8+xv?{Ntz&7fphT+%l;HAL$|!x
ziGgb7!$!K<pS~*^+e;gU&IRW+Ft2HRjpe%c25sIUf1f`38x`wJO8L_Iqg$~n)v0fF
zhGys0Bv-G#cifxW`=EKGxUEx^M&vr42b2?wq3`auQ#3A2RKBlsuWm_~dEX~o(R(j<
zNj4MDobH$QjBRJ{{KofQVadmxH#BEEYwQljcLtkgw|!&WB**aK=8ZX<+yi+IpDC5x
zPM6fq`mlc9C_Gh+U-=3+Rx({;Yj_*JkljfYK6kdtnsWBe`r3wdXB=05QNP{p$@6S?
z(p5(K(=wMo(uRCEmKF8$+A6Mg&z%|D+{m}6aJ;<I6f?|<O4U4nLLKdxl+YjVwTvWg
zw!F@Y^W>Fg(ras(d%9a1C<v<wa=RSolV68NQ{zwUeo}qy>XSddYbcf^r5(9iFZuU>
z&=hxyPA30A*~KlZ;g%|~j-qnRJ7HSaVIw8vF6G3?K>uFe`z~3hQzf$0=_!N33O_a6
zPfzzPS{m~0>T;?U5tjet*w(-oE8Z?wcK?3zyfj6&%ywOP>%qEJC0$*<SKQ+!J1gX1
zx&;fl%y<Ku^#?RzWxPvP<4k3ycdeX4G+ReO)A`Yk{yCXW@w|AZChJ!ro1&8sd})2z
zG@UJ*9FUsXKjg>E5!f6Q$rNe4F!gh3T|AS5*=Jj0{zV!J*BWz+<2f(WGNt&`Eqa>t
zH^?8~Q0`V?5d31Yq?I=GV6Kh2TV<wI+KBD7)2vwu-MXs$y&17ZMsrLQ#cbqs_sl<q
z-x;t?4Xvu4naVwH_@b=X+&{+L??#=-yoQ;E>C5t(0nf#$)cvufCqVsnd0M9V?Q7<N
z%Pe-vA3pfKe;79uaZ=vwFo#DSV`&{T{x?#_D@h{p$VuVEyr(r0)|JHFSVh=6t!#Ke
ziECMDS$Z~}DaYWl*fPzYnmFku&B*%bJ>m5p-9L%9hoCLzml%ev>Kk(mUdJ50dDeL#
z{^$9sWf52Be)YM7{-sZ&M8{_>BB-s@L|HaR@3^O}{z|WM@XA=Y#W`aQnGKDK-aeZ8
zN2ttpR23ClrrcbVmr*q_xwj;sxs8hTw9Cg$Vl;Y7v_={ShWE^!W8_Wx!0BVbC3}>I
zf7>dvE=@1X?~$DhR&-ekJ!Y(NBV8~1ThD}_`7ruO+atxL!O*Qrqe^eva_PHAkq!zC
zKj0TOHdsc>=A4+xnvOTD8!M8G8x1IA4(#f8i{qDw`udJL+i_&%PIcn`IL0K+Rl-{P
z>UT4XOtJ1&x8AZy+cEZV+xhHE!})>5JvW}6)0u77zTA?zgE~k@@ql{r<u4~1S@gVx
z4&Ji4K&2+nB9v;Ce@&T{MdHxqy{diYRssDF#mDmlORt1@S(POhnlWa|#$R0*o)Y&1
zwfac;$#z2lyKa%KX^CRSDv5KC^cq&(Q}Iq1G`~*O5dMC@sM3VENuPsoR6W~6o%>!^
z3cDKr^!-oqId3}e2CWrhl0Pc^`g$16POqcVS0CK>O3@{st7J1zsqTGm(kH8Qp?*!=
z6U$+?^7ju<_)+iXce<2*Ue1Di2pecn^#vvIQ`eL|4a*t*N}Y8Sb<YGF^?P?du8$7k
zy|Bi6ZkKE+x9%I*%^mmlQU3KVw$=l@{IjB0C`ra5+oCa1Z2+jHe3PmAkEX2^^Y)#+
zA$cxJI<GqlSVjcZ3J!Kqh3LGe?gz7Q&~r-1O9b<R9<>yCc{P0z&{AwQo}5zHKvz`1
zbpFl_3&GCb$2a}Qt4>NdrasOrb}pCn{+9BQvMA~`Av!*~yfiu<L+7zqdi?gu*IGJ;
z472Xip2$MY3|Y<MR7d!CPg%X_-Og1m$)gu$B>Ynbev+Nuc4@Y+0$9d*!1m+NCRNBR
zl*+t@f{WgGri3YjFW@bY#=M`h$F?-_H+RPaRWuWrXk&&g4#|$TJmd2G?&KwkF*B`V
zU|@zj3M7GXhs~gL;sjTJw&{zYo~ebGVDgnXk##JB9jB7>1(edr^#Z{uBRK;Qoq&^G
zbM=BwUqAW!np;~37R@~=s?3(yokp&A?ogCi;{K)1x|bS78^!DPD9E!(4LZ{7GR==W
z?Q$+qU+ei9mrhigwsi_j0UTr?-FW5xr}AgcoJlpvIapdLwjU4NZ6Ry6VL08nzb~?G
zvUAIDyVTY3H*L*E+vdxb=8H}iGA5AQYHZbK=G!Ad*O+>7RJPN;{zP5NOm&bbVth8r
z+JmXSFuV51?_9K!_UiC4m(YGabT`MKYvYu&(euYY`8#RyNKGVb;cY`NKN1SnMgKTt
zXC&gXXv=FIcVYjgxg6#4K6o>E3ThvPI{j=lbHZtB0SptRoXMm!aLaK#IGNHrNf!HP
z)^svlxHHFUb05Dt<6wdLdQ+|ERC9GooB>5Menq~>k3(g=Jjt+~aZLO)HDqk$`Qq^A
zKI2op)7s^d<@IhXDo-vl(R=nE<-7w1GS3zut*@`xX#T{(bl;Tw8P2xT!zGp>yNvqR
zbNM^vw;db(Qe2<vLS3zfmFVjWf@WwO>Yj28bP#d1#z0VV8&zNKEI04@q93}dTv9P2
z{d>l_9}*Pbt950Ao$7%E!Jpr~i$?J9VKy%;3nq^(Qy-|3u9i!7oESCTTY#Hg{>Hz&
z*?6{6=RByz@jkDmzHy+v6~L-9_-M{nvWz;5K8%$cXuRz0S>x8Rh5lw_Hj*38a`C>1
zEs&mGE!Uz_mFi(1P56E=#OQtXwZbl+ZQU4BddYpRJ}MAX4R^A#vUdLxwQ!-yluA_`
zpVUsEJrk4O_#1mac$(Kx4fI^6;q&dBFX;3uEG#6O4E;07Nj04?{o&PrvA4CYtp}W<
z<UkJqI49mSFL}YzF=CRWt$|JPT8%|6^(&JR=Yy*q^WzQ;6l{ps`h_a@*)u_?L}XkK
zV_;>dxZTB5QIh|qEH~=IT(S_CRN~y+PM=$y1LC{nHGX=|GFol_eEZIBIQLla>Sb_~
z@&*XwP)NfD^%0AMvP$w~g+r&C%*?@|HkbeJ^5E*127@05T-A=y?-uG{uzMbzl_iR)
zn44khW{?lD*Eqn|lMVwEt>lZVCo$cF5wM>I#93U`aQ$@dobO<Q>(dg8%$2iex5MWG
zcYCEHpP89CbgUH*QVfKG`(zwvWJYfCQCBtF`@jE@&(IkK!I8ZJ!|8VJJddfCyXILa
zDa!Bpb}nEVwb8dxl8!<wd>i@TiJ5VCX-=G2J9o~_&&MSunoXL3IAga{9XYnt?)lY@
zz9KNv-E(eo{c@1{PjCI5wzTukG<z*J3pV?|D9Hl~WMXc<7aT=LHYP6Ar0Xg=AZ8(T
zeBC9wQEBL1Q%n9;yWjXk^jK1kV2xw(c|96i0C8p4T4Q<X(rplg{v6-Ge^)wwyy~ZB
zM<=HWun1QxpEiE{$P!ufNyg!Cxp$yvDz5FFU?$hBlca!vV`sasTurjObjdQ;)^iy>
z?O{ZbTi4dMO-f3tuCtR9f8fM$d*1iQH|rDg^wj-b^JB*59zLyjawVVD$>VdnFWuJa
zg1PP|<C#lO%ZBaDPR@u=&4|T+V?M=)<Al@s^R>F~ouCq1FuBMbyCKjFT9Fq~S${8`
z6`?k@$MtqDjdVV;ZwH^yvqxrD*#O*)hLNO$V$Htl4;RYw&Ra9SLwy5<bEQ~cgkARy
z&V{oE?&D?TTK8+GGTXHeSTk;o<G8Z-h14fTk1G-RXMRJ_?5>Ss3m>x^9Jf{ZA{c9c
zgC7H3>rDcK*u*z`ecR)8wY<jC!r~6tl5!yb=kX38Aeh$8wZC-vQ??+4Hyo%p0)0!`
z`_PW1h!LsEgNfC@U0o&g%b~FG^w5WEMLV2LF1Nirurx-uEMVj|k|_rE0|lK;<<L_l
zB>Bz(r6kFZ$HH`m6x$DD=8}^vB_$-F7cqI4GcKhC0>2}fhxV)r!^+w|3o*<1`zxM8
z_j&a;V`2FLYS|9|9F2si*q52rmm|kE+7G>(Tzr|AoA15$i#85cu%4_1lJ4LITF}63
zU|Yt(qO(?DYPV!Z<iN!vxah+WFN}j=I$>4(y7%70xRrWksk-%6uU)vwMCFUcmrAW~
zf_P(?FL811kUzQ)XP8l;`)vRGi4!LX0FjOLDU`ZCTG-#I(d=<s`A5etF~e;xRIDnA
zxG#j?ZmCh18gsuKjOk!zWApFr?e*Uu$DQw^%r4ZSm55a4x%HXEfWsH!E{dMk%eiWr
zk6;&90aE(nLhqHOi#`me4!$f&sAd@*)EU~Bpc*xJSU?2tstCNeEqEmuxr1I36kBrO
zv$&D=YqDK=-dv`OU5wR2o`@(@_3C&HJcWW$k1Ot@Zk^ktJiBf(bwlv{{@0x<<(8oh
z;ys3om0x~z<2#_csjgPt`C-k&V(NJ9=XTN1XLgYY*{$RGMd!S;vzRB&fihsiVGepB
za+a3oTFd48Vm3054n92eP2E$(ZJjFwGtXeNG1JXd#0;&7`3lAZ?WWjJ29FUykuzuH
zO-#D1J|@ziOz=Kf<P$b)vZjZ5lv4Qwj|!V|^YVnu@)kJ4DjQ&WGKp+6RYeO`sZ}ZF
zVNc5*UZyg2rxlv0Sik#h{_4F?5qDOQAe`TE70_1Sf!_RZ=^Y;9Ud!?71aAYr?%EjN
zEqi{va(vSDVs8+=>MQ4U>PFalw|snXKhE#QflM7tP5Ne9$jCpjsyg*YmB$-dMpL4a
z&#4>Y2@}OL5q!SJG}iMVc*xtHX{nID@1VzUV8SytjLGJrmikVIH_kfks<bx-$%BSQ
zCnyLHY8*T%k*2d^O%H?yP>bpHF3Jn>e}0z|EkAg7JUuszmAVlN#NNcj!~m1(>YADu
zfm)Lz;1Fu^h=27b>^!3B*u*vwbcNs`prn?vF<EQp&hEC(CCBo6TDRWK-@a({DopFh
z?S>zE46SM_IZ{r(Z@;#bX*{3ljAmUh3*($UlAW_^gjXg|M{3oo(5vNnt!FKmHKW35
zYK`Yxo4aHE`0hZ=#3w!b_GCF8KzXX#x+I1_eTr;uayF4e39=_Yv<8czho^BSPG&A{
zrdPBm%)y1CQSPiGS=mro_|=^5SaO4bH^+F#_d-!?h&r6-1`^|niex+&o4usaUb~c9
z#Tv)>sKs=d#iHxTjeXj*J?tS5r@cc?<4kvY2ooB6M@Moc8GNP<kijf($s4}$OWmgG
zq|FtQZ?2OM8m{SA5f)HiJUN^}r>1z6a_6zs_IPetv)?rHhPCVe_W5Y@De0c~)z3nL
zDUm-9)bEitKJmBDd&h>3C*UV`$(q3eaW1OGlrzHLhoT_ZbHYtl<+ykCo8ga*%dh$7
zKVJ}Dv-b|mm9?#_PxFXm-E$I2zIh;~YDuV}<Oa3+`IqZ%eYlFDAr}5DSJd^<aoNBu
z)Disx4hY_?!kh1lUf%o8gF{btuYBppEzwlb)B!ItwhfdEeyHS=^YwXdiq6S~?P<f&
z_J9wPj_7|pGC+#j8)}Ui*3-jt9o^d!uEQ8Y#5`8p5qQ(Wu-Q~WnW*YoAPN?>@@d0(
zzP>#b6-H)8!jHE^pUcS2&ENQ==8N$&dPPffubab|XfFZn^Z$8C-Dus~lw$tn&t7iX
zf*Tdjr20KGX$+nHV7>7b%MY~zEOSSx8oLd>#zf0)UMLtJ=7@~j;rfoT(|V)qDFPQR
z#&}x@7IQ9)1{BZuwb*?7{X4_*>wl?h)z20Zs&nyHxq3%txQ%sVMa<DM!=YY^|Cf!z
zs^#*FG3B@<^&M(cw-)95G}3QM?3O+Hb9w5Mk9Yk>KOIV!X$WmTLd7Fr$+__RAKP0Z
zKV+f3=)jkIh|@JuCS+`bX6$WNxu`+9r4Y-Qz_8V{3S#ZKtjp;X2L1>8J%3fnVH^yR
ztW{q#WI^$!UnRV?IG2@j77h*yzot`Dn0Lq?rB8iAMN!n1U$2mXfAO^;BVL040})oS
zc&C<bb<-s2BIJIQT|ybZ486Ii>EUs+XZ?^$k`V!sciH7e8pP=TP)m`0+w%6FJ~#O{
zlRY{XKjYnm_(vKfE36U@doHmp+VLa4{g`H5SO&hBCSUYy6eeF#k`Cb`|MWkiRdmN`
z<JOo5GYI)GN><-L@9<KP<LsS))S0gYX7MyD%c8n35l^#D`OZN2&#ihUJUV7WgH-CP
z63>iV*~D+mS$!h-eD1{yAN7vv(0OnEwR>CnrfB8U@0i2Ch*|pI>V|ku2=w*Vk&zKw
zB43{zhgxnYA7<|AzPzkQlH<b3cz#p$T?{+Um1SMBZ&*zg&RymoqvzOE+qn9U&_B<D
zBkkMKcd;41Mz_&l^BR_TKOc$L{As6-`$L`+1%JB$Q!`pasc12B)&XSJ&9*tG8YL}7
z(8ji`D>b>iJX$~RFFX+!XSBYSd*8(k2NcW#l2hq?k&ylc$W_bz3Qy2G5P9>W;@ARo
zb)VU7&G5gU`Fogk$#3Z|Ig)`Vp=xRtpsDA1C21EXyd6zFlf`%OT_}B!)U+;uF{|<6
z=QWS)4W*-kkf+kqFD@?D0fe3O2;`6oUl5;4V&zDz@!;!jdG<>Ukmyr^AD(mx#va0Q
zaY%}fH+{$c1vBf+FI{4>gWG_eVrLNsVs}FsA2w#`;OMB>$3L&YXIyxg6w{COU0q#+
zWfD2GN_&I_1YX`Vmp{j9D9!oD?}Rgs`Yr8LUHXE80$X&&h{1)I9I_Z$Sz$1N8a}A7
z2-*PMd}rI7s>hX~ib@4oyU`m(L`2G<3o}7}B#4)T{v@|Q5M3x$gP`rM0DU`)a#Tqe
z9|asw8pht_UQY2|dAqF9Y7X}H@=8h-p?09ISUtRe`9A~ZfAzsy^>O0e@-JBu4F-tm
z_doI~JmBie+B4hN)_qSWH-qm7N|Huv%c*hl&7}z$$)4A@OPaKcW)zGU`n(t&rRHC)
z6W+F#Qdz}r&eK8X`Rh61!$BB)EZ{kIgO){W4W&VdD*VoOKVvpE2|9|p$|_-oE9L%8
zk0)er4h`NhxAZf=%$x5)FuzsDcFJmuGxB|63kY~`QY~`mP({hjhe9dqLtg*_t*-Vh
zECdKz(pFel;^NJ-1D};1yJo<v9}>hO#3(O+^j+w)?TiO1&&aNCgnq@h_?aaul11{_
z7@BojU8_0Y{P0elW@FU(krE`Un%kYCsrE)SX`~v8HM)5Jh<&dv{ZLd%ILNSmQ$WDY
z35pi5#Ax@cIlk`F=ZmlV`uLQzt)=;~k17S{kF3j=tMs=vGBDC{NNg2!b8KRu;)Z(C
z`$A8f%aHsrsX9=}#}0m|c6-cq&e*t1tmHD=;0;NfF~PT6UFbGi^qnB~Oy1nBE*hhV
z4%=@#b~H%h;6HZol;mjd#gY7ED6T1YFT^@_%;vBCsez7s?@(c6)|f%@BwNJQ))p#C
z1CP|Y<0U}fz4hb!&MK2bFH@zaICxX^o?qcAtNX(I)QhaF3Md3cyF`o^2X#y!>3@bc
zl58ZLRR%jKvh<T|yg%4QSFo_M-XS$$R^e1ysOS}=RNZGgLGrPK2W2Tp_){7BE?i_k
zNewpar>2PAeE+bb`dVPEoz0KgtEH+TbB`-1kir}}W+*~~P)U@7KEh&$(?bPBF(%mF
z^z;Np;#FBo5tG8j%XxE{{X?<JZ1)~mLpOR^ZJ~#(P?G-ka|;UpoT@dHy@8KzXjE?v
zRCb2KDh^kL9T<zegLE{sKnqFAf0Oj8zKNe}`nYzI@8AQ9GLA?-1uGgh7M;NewpEj2
z2M!#th0j`D&wuQ7w5oLjlk&H#7T+U1Q%qM*=rQcuXQ<x<n}6T=ZU_F@`E@~4lav19
z363-Et7NT&Zkj|e?AMmZv}v}(&ZK*ivAw<h8RmHlHh+RDD$a@7`0w?SzzqwKo1#>%
z5&v`M#@E*u9c5)WgwErg*+Q04hNk09P(HuK{5wiz@z%f}z|#C_n-_sO<#6qfA>EOc
zh6sWzb|e)3caWnL%D>`K)42={>8-5jv0btZ4fS-^P}DuZn|goiAH35)><?~~?fVo%
zAlJ^TIm3gi5GKFMHs7Y**6~9P7xg-tK0Jy?6vI6mIn^ZTD6d|x>6fT$4R_?Ac;Eh^
z*5Uj_uWxVfv%c_$|M9`UY(@8uRdN(>&0L#?pAFQ`VmY4qZ=-~h#Qx#7#sU*kX(--K
zII#ZnZ83Lk6s9nkmijoAO!yWT<GfJ;L!^Fc<+jwVv5gtxGet%Y%jO`*&<uJ1kSqLQ
z#Uo2=zKC^6gs^7V^}z2<9~UF9FY5CA_k9Ppui?4H$41eNAos=d0H?<PF1`J)-{3zd
z%|mzo{s8#%)&Jnb_X(RCQW~!Reh}j>ZON&&bLWF95YcA`b5_%8oc;R~phUl1Un{Wo
z_ZO?lPsN%bYiUWiaaW|HYY$xsG`Hq(OV7!%;h-cPSS74_?Qev5v`zlmBV8?cu}M<7
zHm|n6{qp1T)TBI{M2Le3MK+l*NdIyz{_{KeGm%~${bYY^+JV?-&&FOp**|p0n^y$)
zMA6LnXE(K7l2Cw`F1L)SIdce#>59=c?7<nVhNkit6S5;n_dg5kpQ*22op}<BQ+-Rm
zcx;Z#a87S;iA%IJUbGEsj)A;1=d-g%<c}WhvXZ_+wWC4EP>k|<!&UKXEY7$7OK)`x
zIRgKReV#l#E^px6b@Lf4he)dDK5*Q)GFpW~$vC^G#MTFOF_gLEX;(b<Lj}wP?8uj-
z4#B&Qmm4^dN*FnYP|9YCcj~`Bx}Rex3lRnc2?+>nf`FG(>JehYh7B7c?AtX?N->am
zwm=PcLm+#)dNUQ}&E0)$zo7F_^fgHR^X2|51okeo%4%n<!IWd=<}VYoZ~7+a=?So)
zOMsUOPY4Jqpc>JA8NQFRsIB!SzLR=9$sDE)tU^r;rFIEo9Em~`%qJqz8DnZFr?4fu
z0xrQUc1k%o-SzeL%MH+?8BzCrG`lA-8<M%*3;L0fk#!$Gig#5ISZjIu%JQML3>udf
ztS!%2h^306FJ1pzpL6xriXQumUyaI|D3o<_%JN4vIbf2tHPH5dHDg;t+0OM<HmJU#
zVJpP^0pzg8D4S0p7D#|?feMlri6V)iXaVd)qQ_o`kD|3QkmTn>894g3P`R#xyal3I
z!QM|0Bq?@w!YCE5^Gj-BMRJ0p`XVm+VdrSWL?|P<T3liLIe;BvLpv#$8AK#<*bM!f
zpC53xB5kxLz;12d?s}j)GY$#dOpnY`kNK4=K_sy*{9Z76FGsGrJj}BQkf$K&py?aQ
zOLVriC@&QfdKnScw};_eXs`<c>n)6D7h5o1o-!^inknP=;m;e+=pyL)I}zihZx3!Y
zKQ)dzbyKxTX0cv|wg~;CrEu1Y9P2^zX!9*}zt436#@_%$dZM&wd|T0}-XF!cnawSF
zqF<c@;OElFK^?7j|665(D}VLgEB5})83&w4a(lxq&qG*}xG;mPp@(kiGaV5FA^Zou
zUSXd`=#nqv_XJlv>#~_tg}Hd1CO3~CUF4Z|IyvPEij1`F2L*Wyn-E&$qM=%u5QiGA
z$#L7mGW$fRBuq%bz$Rh8cL|S%%X28pbH8=o$jj8@tExYvA063c;R-88lr1E3>pjri
z;(tqZZOzHbBi?*5=d;#E37RO%Z+*Ge3&>lgK3DvyccYI69SAq)(ih1@f3t8LEpbr|
zj7?LooyA1_5194NDvPhL-F4y2*c=*G2WKjE3OARQmX5>ASM*|a1Pha){i!BXL4_4O
zZjAp-Kl)`Q)gr{`R|0)X<;BAhtmXHGCKevE4a@7wy<Gd$5L!pSM~@ykLt%0r29D!l
zHSW8iuI8Yp2X<(NoZz)+;dlUf@qj01zC9?4uq}HmRzLckFpoWc#tY|3*#P!(8`6}z
z+S<FejrZRF{gI;T$Ey0DmI<ZYkZ>L#oC{uj0A^Sn=0IZ$VNd1dp{RL1x==lO!C=R$
zk3vnwZzr$$C`+Jq!364kdV1)fpO&M8nN?=~V@<bE)U0axJt|dp6ZVLA?B=1XkhBJ%
zZo`J6O>4ZSZ<=}o0<VNO(9aRtfaIO2$Eo(Rn6R(Z_wwF8`@R7n-;TY*h#!5QyVW~b
zunEF1jqVDWCEHJUDdWi}J$OJkW$>+w=Jt0`kEU6t`T<fj79xa#=w1mbwS}e8WzU6y
zsHIDUk=logNj=wcP#=>{!(lf5e>?R4Ibi)C&fflg#Qgsk@c*B{zkdQ_%TzaFRf68k
T{S*QLEk#*DO+H=j?DhWv+7o#k

diff --git a/examples_skyline/accuracy_vs_di-skyline_formula.png b/examples_skyline/accuracy_vs_di-skyline_formula.png
deleted file mode 100644
index 271ba877f30856507b9b09d19f5444fe4898ddd7..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 42840
zcmeFZ2UJyCmNk4xf+C0kRH6YzK)?VZNK^p<B`1k0q9n<e94;!NBA_Us1OdrelH?#L
z2#92^WRT1y=ghyh^<H&%Ro7S5<NwAty2t1@-g{SX+-IM)SD16I^-M`YhGrMzE)2tH
zWUpOS!7$PW45Nso+5yivG}$%71G(*GSv4y7$Cc{tQ~3K%t7|uHF^qa2`j4ck+kFF`
zyl5w-Wv6O+&(7htjS=?Xww;x^rJcFSog?-}Hnt{~7Uy}-@t!?>#MsWxN|=xD-(Kgn
zw7Jh0Jty3OVMj38tCDJtVH4d>T7kH&?OB54fva>a-@PVdPI2vwxsw>EGyEkXcEH^x
zQMU43wo!*6Z?sotb@tb^vy3t_mmZ4OcFDzOpW<mgEF*t{M^&ORBl0f2WS)n|%wmhO
zgatWm?Yy;~zLm$1k**Op+O?&x#oD_o&XB|6Bjvhu_23a}cu5yY0+0KtU>7j_PI6;-
zC?hB3hlg9FB#iKI#uKA~hqn@Vv_;pCNRYw9$^RZ7YdM?Tnv|-OD;MzIW6P=Awq7)j
zkm0&&)9pCm8y6vHJA8d}h|l8{`-9#Ry{M?DIKu79{t9pU!-qN9*w~7fI?Qs`w1Z!V
zg{9!}Sa2H_X>H|zBf%Qph^`L4=5?72wJ4DF-ouvGN?0KpY$rus@iy%(am!f~#*~78
zlF~%#pWl5YHa0emm#p<oP*B{*j~`7%mp4}D1$1e!mffWMc&>Ni9wi0#lU(p~i{q{G
z)5Xj02QqSdoaIGbmh429dbhVU1MD_dtbLgT6(&1#H4E)0`S1T=mQ~T!4Vx;OSE!m-
zRaG_0&|>y9`S$YYb6?-qNXbWoLqoCAGXAFRnMQtu(WY3R9@qIB@oR1A2I=3vUCXiP
z*48VuZw=zn$-Y;|xH#EKFzG4xqQ<RH=4!5pIZSrM`mtO}ii!DhcB#i@no~qXHz3<(
z*<z7MP=u}2@TO&17!luI&5Y=-3FM|{W=^r6?BEm=>lw@N5%=-oF(}d1E3kXz@9&?M
zKg`%ZxnwgUI`{2(!E7bdV!g*!R?YP4uPclx8p#Z&wIXL`tX9|7Oq*h4xlW(96Dcui
zJ}D(FT~lALNJdH9LNM^q8(U|+VE$HYb#Q;(yYn(VJw0nnotA;&;i+TI@yTxMOS@=k
z+u**AtWcP&&J8j08^2_AosrBjBPdu~TZahQ#%(Rci;sQkw0NDHo4bRCF=l&n+QYfe
zld>i52Ginb;}w_XDO^$9*E@}`j>|CDmWtgSx)Zn=z5`zaCafk)NYl%4Temm5f1j8b
z|Gp{Qp(jx_Ucsa;MBurPPxEtn?l-SrN7FD}NPsQn;^b^hP>wzTThh{^W_0(i@!Vjw
zx_faTw<g!oqmudg`Dn4Nng*#iLN9P~aLCjK@iaf$BQ{DkUg$90wlLPb*d)(4_Wj-Y
zwTZ0yyE)nd_r4!uxnzI6*LB{v**RV@LcgD=@3FaP+L{z7wzXnd^)pF5nIAW<Cfl^R
z+?&2WU1}gCB(%1*J}qGV>x*2kjGWx_t(7Wq)1jI`VW+v6%*;%?UtgYax$Rh8T`ia{
zS<iNwzc(~IJnq9+qICCeOmkRMlL}m#yn)ANj@ZV`l^gFx+8125a<-SZw<127w|_9K
z@*5+jm$uCf)t-=roonrO=uH*ubydB2^X6Iw+g9XR!^+XFd|Uc7;()(++YQ0)n6Ewu
zx%v1My9*r%@B+GXV4LI3isG%Xd)emg=c;VMfMR*|3PxcEz=Uxv3thJLoafK0va+%=
zp1FlzTNsP}{{6Zyn|O&_>6bV0A4;1sD<N=2>gR5kljEkkZdzDagz%dL6yS()@H@n5
zkL^NMw*_S>DXHo^8K>CUuYUggxue)6SK+-#Vz9omY^runKzw|B;_IG!Kf)3}ems+3
zSeWcIKfE?+>5-hKS6Fi=!+56ebHQSp4!ZlYO--@#xaDpKvoBBg+SuB@E-fu}UKl;P
zXV0D+A^ftIrJJAF$ND}$lDl!E6%2sOZIV+p?$p_{$^maeLzDN2tw^ZGU4O+cVA-Q9
z<HxLg?_OL{aWSrfMo_IF&7sFhCR*CJsZOx>G<=`v`b3tEZL%{FXKUN+Nk(b6G;tF*
z-cmhg_4L=TJ2&2)lM>%tczwa5>l7bfx@qYM^;F@s7}|mdou67q8>3mOcsbbFNB4NF
zJW4CwauzUayMF%O_m|$@-X$AzwOpma0Rj6?oH#M&bx{8$98oxitj?oHQ+4y3&c@r0
z{0O<^kk-}R-I{46FWdAp!>~2i(!j~t*>vJ(#@xa}3*4aAl$+<_qN`1)vk%jwb5eU$
zf-E6`o?GLheee3}>R6d4S<UzFEwckvt#AtEhuOs4bz%E$9MXI8Ydk$Y6>38TE!$`W
zyCOe5=tL)-&2`4}-1+k?>FWA40sMxQyW#H0J-KZRMjLX``h$pw2y7^vI@8IH9ASs4
zu(mY4WHGn3Qx`94sod1s26xaJEwcyiwMlc_jleYGVD)Q-=~(q-4O2M0`L_4#-}6{n
zS~|J75LZVcG?O(_vz-@Bd2VK24Hw2WS$eGNun60IJG=eu)p6}o_u}`?<D5ChjlZ7v
z%u~SGvvZvWZ(39&B5PE2t`xS=FQPOeBEpYNJnc1yqJ~Lvz?*mP65-ZV-RG2r%dGkK
z&FZwlhHY>REsOBQ63h0?p%b+SP5G1mYP7+D5DIQ)-7}f}^|hv^rs>`J`*xE#gqkzE
zDR%4_hcjF{C<NOir>MxceD$M7s&-;zWLd^WDfq1bAqJnXvzz(0BMXJSn+kixH$Eyv
zh^3yt|KpUfu=bPN7xj~i7}>|3Fj*-1Fz{r9<)W>GT`8PcpKMJIe)#B74xIS-3PJWn
zom>mK>(^W0J0h!CU9JyQ`D?nnyL%DGTM}Qx)lF&fo;szZp`np)8+`ia2iSn2p&_BP
zA7^a_uemKW$_V4e-@p-wf(sEEUY=-IMZWAoZuhXAfbV$GNbSzuyVcCiQ+|XA4Xmgt
zDfRt4XF#v1hYJi5v`ov+=C9Inb}rc1++?XO-OAVW4h|+5ZCl6&ZvSCVc_|wkt~XCJ
z<XHCV2OuB8a?wg6z^-0&+4%EAijJ)N{<|w0z+a=Ny0y8U3@+JuW40<VA|k!K{Ic8D
z>PSEgggUpS_WQ7pn)dde^d0-X<P1s^60VK>2v=QSnjD&%N~;SMBv$MZw-eb~@7>l%
zR{$odrmday;>GTGo}A`*MM9X6ZA+2UJQthvIq<W^YvaktH}iwvsaol@EKOze*iZ)V
zGdDl~I!10!^&xThq6-f?V!(0|;J4<7>yqcxrO-`+@X0MMZU8QAZfU7xFemQ%yALA~
z9{RamHa^XKdt06A1K$al1qH!oLnx^}vwLV{ga}?;6TUr=(1=c5*hS@xjEtJPI(2RB
z5C&ddtufg*_aIzCI8aBi2O|2ySbRj#rS!}{J!EievL>JCLc<m0M}_T1BRAKV%^D&l
z18V$^`$D$RlBhZ~Fc5udD(?#ND<J}Aagb_^!&%t1YnO_I1YRRupGTJQ?48FaZ-m4^
zw1#wHIkRd+KW}H%muLH1A%-FoGpu}JEz(o!;ZgO>tE8laQN;0Mv&IiNkH{OuJ?wbH
zsft!+`t#%Mg7Kte<^i2g=E_=HZ;>0^-dyJ9<gB(%v2ZFcFJGu+vNQ%8L}u8LZK669
z>I660w7=ZT&oBU%sJaO*y6|q#i^J#nEqg>D@?K9&Ol)<p4;&+;NAx%oc~F}0pd(xP
z+O?+x<X6r^yu{_0w#q?H(&#JLlK_rn9DHWlE5;LJ5Q<wXeVJgVISRYeU=yOdoPNjt
z+KfXJ>7)(c!bM#)G!5KxkbNY<iAm#)SK$A;yc@%L@ibtJ#ig*C=g*&;sz_#XmfXh@
zxVvhYB%<%-Xd4}R(w3?-!q3XUFl_g>io&vFHO~A(Xd>Zb=Ol`gR@IJHRv&3uMU2z>
z7?m1b(rCuU#>U}h<@?9R#GE>HDxu8`!xvN@Wrk!LzmvHfCE@=qQ|$i{b-@2F@_%Oi
z_K)vd^ylBX{akRVF*x&GG&DGz5Z0%X+if4ZwOa2{Q1HT3R#7n+a(+2EIbFS7N+-yT
z`5gtj?OxNeiA{{3GwtSxaQy%kTMdDrsgb;xIiLHWGmcf%`CDzR{1tCnaIxBx2X~ek
zU9jv)Ieh+}Zlmz8U%%`|!X4U_r0BD>Q#%T9$%aCwPF-_yax!g8xtUseK?$m4<CY&0
z;?v{%l46SrbaG5%3#UtT_=*>rj3Oc+Pp}@S9BaQH(W)rE8RNdWW&(9jjxrZsxJhaK
zaKoX!k&U+|x(hYk))r9tggh=pGg%cIn@mztQs#L`*7RLxrM$eolOeki_Sh<doKBvG
z`Jxg;?&nXRD#05Kw_YB-vJhOlAq5508@POFf7aBLl+*3z7f|7+tgH-4D(?ICFCV@G
z^z<hT3=AM~8iV}*C3~uVvG(TLqHU5>;|D3m^Y>n`k3qujywdMYf8YR0XN;F7+J#+~
zQkq>HXTKh7%`lYd#u4;_i?$&In}M?)Z;Guo5>JM#4~n_#`%hkf2{|_G+5-65v4*HC
z>gyj#TBr{$XF+sS>!{t@S}wg5v_0$ZkpMZGX>**fv>)?W_Q7$;q*27_`1BwdesvVG
zHUbK58}s#MP$Wb6)VO-}YHUuph?5Eg9_7wY59s&p`v6sQ?W`8m#g+bS>9Vg+R##QY
zz)m)T|2`oba^XQpQ**O`^MWx-1k*h>lU4pwIG3{PT6FK{P(ZmY3WZP%Y(jaP|Dpcy
znL3OS4@vQ1NlD2{sE~a>d^lzZ<<XPE!os~wOlGcg)f`Wt6fm>^U%q$m-rF(p@k|sH
z6qV2RbG(2|f^oWZyF^(<<&G=tp)@Shm(6|U^O@Zhhaf&S=9p;?K%A|F{O`qkQ5VMU
z0^E7AwbA3zd3itI!g9|7I#F?E_i=9S+L`(J^BO5yw+|dV$OQT8xs8pD7f=u#7Z>lH
zSz5YKvbm%q!vnYO{J#DBX)4aVpk#F{J0$9y{VFifXmD`w5Y!jvSUonZ;p+xoD~g?k
zFn>>EZ8Tc?_U$NTB_*Z)`?HNQY=<Fz4RqIsGAH7!%aVC~25iLuDtOntY0HM(yX%vh
zYEgiNZ!v_$N<((-`{vC7v|dL@w0ud(b{u+g`w&#1(p7#ez7*69hv2>n*x1;-;49r|
zg6no1PmJFkTBi})VNo!4a=_nXBib738^h_I;!1T*JtOnZPjwlr4h{}zm!*}JU(G^2
zd{`s7^%NJEO;NhP*xI*@G$=t22fP5Y;!5qTsHkv+{K0pAq@KxNd@Ij-bCF;$CPVYK
zoWYo%%D{a!7OM2cGO|4hP~0&>Zua5B2d*%WUbkh=GiQ`|bUwA99By>fSV~5wwyy5R
zap@OIrS6*xpDEcc4)=m_k9XuGySlmIIv?aFni0|phJthkA#vQf|D-W2Iv{4b*TVqz
zqB+~789fhWUxJK3Ye$}SbWu@}9Uv;)s0v(wDj6LBbf|V$C|$n%kby@#29-Zxql#jy
zF#!Pq%mR7&`C}-3gIZ1dlev>rgr>g!d-(D2O88%3XsC-6kwR&icOffjA(I;Ai5%}8
zF9S~W2F&rj>zpj$Ly9IQCU6j63)@d5c(XcRd;k9Z-0W;4Y>H)49HjWf`Fe2z9QIR+
zii(gpj=y)AYJ@7B*>-JdG8rni-!<@j#C8hGqF0B41v00YEQ^#I8ymsJy^FgMI{dSM
zriGrApYbtNKQ+C*22hZkWngC3?cXRXlc-vT-`CE%=L-hP>|sCgGYS%Op<%nxhR?HZ
zkY4lM{83wL+hRRS+(Ajnef;>900dsh;PrD%1E=x^Xu*|^Rk3-fNlJRcf`|$If~Bjz
z?i;f*Rn8r`mVB^i*(SIGzQ2@nX5fz0w<EXw!OMY>g;m1;1_uTVo%JwX;$`t<IljUp
zJUlnx7L^+6!|8*Z?A*C?2u;IHb}T+OMTm%@VHri~lvin*4p*lH9UWaHTvpq+f=l)j
zqbofS3B0_#n2-AjOW~y-H6*HX0H*Wp%x+Y?ID5@|uw;{wkqJEL;13tFt(@h6>>k<R
zm@pKnuGOul!QTu`PxJlGp3z01`}6GCvk+li68l(Ea+(V42yUiNQii6zz1{DYyLBBh
zg}J%8H^IU2C**_Prky{1`n!qGb$9oYlI<-QSj#*8gBYQpf$94~CTw@{>TsxnZT)-E
zWB>}dz@WkOQWtV~Q6awumK3&Rx>;~g&PXSK?-UPDD|n?AC>?7{6x7>j@)$`NY0B{=
zWXNm@`Yt!Ve*LOlXrF37-J=&f(&ID|j*NI~G2J6ECgv*qw48!M8+^lc!d@1ZR6sm5
zz!5dA&X43kF%2oJaZ7@7K+JGms9E{roz8==WEKG7F#)(FpypG_cYu@t1&MTD>8F%c
zk(PdfzE^!UM_yiDT}vwgs^`N`qQK<n>FEP6IRMrmY}JSFaa-1-XJ8nAcmKVzqhlTv
zo440Q$r9M$+S{}469K?z`22|Sge+_xGC6h=SOKa^^__<3!!J1~0NmD*V-_5fS!@jz
z0v8WYYIYE88VZ++BD0ToQByC%xiA}Rid6_M>x7aL;R@hLh=H2=uiy+?<n`~UBB~wu
zm7Hl&)(VlGPCMH;3WA#{;Ebl7pHiVTN`-)~K9&#V`fum}FlJP-!e5=oVuN|t*{(Xl
z65;{v8ygBX&;amufl#`o9(v(H6S%vfx97q7A9NR}qfiNl*l^?PK?60YThLFVXpL&a
zM53k#cvk>(V?RVk2flxw2C;CwO~(=^0>#D2!fxKg39FY5y>6ON8=@czVA_yfkpTWl
z)pb6mOZF+OE)!<~$GZvGZCegOKN*T%`2bf}S197yOa>}_$A4x-Bj5~-?`i-lY-Np#
z(-b4b0^44{eVYJ4CPJ6M>9)cRP1Ui;|2n%encJHNwMsHvn0CHREZR7@^YQ`j9e?c<
zUGB1lYPk6!1@-lGCMyA-S)$u7U=$<^z*=#rf(JM+jaLfLSz(*Or#<fL4>W^GTXuZ{
z-k_ljGL2C{u?mJm%phrNV5f_$%`mEE@Otuu3s5|`JEeoO03Xn>iRt$D_cNZmdo9N=
zmIBj^CJ4}qZ_GSGjt{JY2vuZCBhDa?vEi%7C%7fi>16!;%;nav0RF;(iL<#ayn>tx
zQlQu`U#>z{QG5127%roL>6?4s-$ubX)<SRq0Ni?70@&Lt>16nEL+S-7I3B4lU((^m
znxuHNvl{?v%5mRxju+zO<YWYxn%!iyAu3pe7F%Q~rp1!{F}GI;O7JPABqzV9SELtk
z?AWnv`$=utCYR~r_^vKpBO{|$lW@%H#3&9^a+?Z{c<95U^-;oh_<JKjtho6)!CG*#
z5Dg3O>O<Jrw{IW2Eae_nH3-^z!Ow*OtKbq8bmw?50njc0QDfuh#>tD*rW(YPyz5h!
z_+#}oS;$9yR@T=qb&HNc!5-YTnE>9wFZCSwT4xCLQ2RPm7u>mX=fM8`$7Lap7QodR
z@QYMeSHtE$O|F&TAwpf9ET6D<p<vBxOIg&_A}2m$p`fEY{lUR!&!0De<;8<_Bm&wY
zD=Uj4Tyq%YCRJWU1IK=fZnx#`v{Uq0lCM4`)+&Q9c?x#Uz^9)A;7t-_2MFMUsvni|
z4}N}>RB5*?0v=|(r??PdCO>~_9n_@3h8u`j<XYYfJ(k<K_XeuczQjctl)A40nJ@s?
z%j2xK_{a4&5V@+(HsE)HFB7)@g^oC%(|5@45q?gcbos%|!h#xL8h}AfjjP~=HE*UT
zD~=Xw`sr(mk{g#jhAi+K1Z0h5b%<(X8E|`6S61>8&b6>toOu%%7z?jkncd}=4O}c(
zdtBmTG>U8e{d3?Q5!6_`wTD9U$DIsB1uVJ~nUr)!aW}|{)7GdO@Uk1;P5d*q(yCRZ
zo^hTLwu^L;o;xb<WQ<06v)Y{IgIf;gA&!vcvF#3Vnd{s+6@=#s5Okvf#_G1I<q^=D
z<O_KHTEMjBlx!22hQR$FVchy708M4%h;6&r400{PRa8`xMA#%YCDYd_iqp3G4;e>#
z&d$s<fDc;yvR`o&VouVMv-9#)D)1(<0SNk5j+QprSeX@2zZdQ}{O&%Nn=529=oC0F
zj;qJ6V&i+9yZJ_Oi+gk;J+<%V%4;MK4r?s6T9nS=3JMFkPo9(pps`>u1Kb3HV$his
zbIpUucOq~z-Wu{OlfE)fs0(~hMA=0Uzx$1@D%J<kUZWiC6M+w84+~m^xzC68CIGnN
z3?7N4es!i|kHD=mY;mkv2K=HD<Us5ua2CfKqrJg%4CF=Y83Fc{V3uC20dXr8x%;Y>
zta`EJh6W|X>xG;%dJfc5pQFLSo?0Lw()5d+5s8J+(vp>5)X2{KMdqF!JAQl-@U=^I
z+smcfsVIUTJ9Zi1F#z1AjEH1PVw#HukkU2~M{&HL8H^)?eom@fe*q`&Q&^D3r_-_*
zIs;SJfgn?d;G}9890Q@%wDQH_kh6wQwxFgNEm@yx2EWQK8?O-F1dJMFK}UTLpO4el
zxwSOe8DH$OTvJhT1*{!m?vU^$j=XnKfNg-%E_DIQGlY4eEddDsv>Sk%yLRqujFElK
zrMC>#0>VTA4wrv&dvuMMPJoIDx3%F^ygpS32xYpOu5Jn-<O!C&B?|R*fH5O)14S*6
zA)~-9A@}_DtYJ-ehnkuieC0_KaDV_~sVFG;!9q;d=CQsGeL;(^i>WuWp6JV@C)=8P
z?l6@#pWxe$(goLZ!97+x*8(wcM2IN1jJ^4`@o@^_T))dKNc;L7C;To=y$lTv1v82T
z7*&&q1D=!@s!bphk~Gux<>cj?gALphp?;uY<iGan)vGlqx)Q;)0~mjeplj<3TX)ju
z>xWCzMOjU_>7N-_z{#Mk8JwI%QP%{j{xzrw5o=l7+^i}lCdMf0yl|qvSg9H4AD2|U
zLJfU={lJnoAaq+Hb%QcjmTO(4ySx28rbb!pr%yPr7i@aY!{tD-h(#OO9+Ajc`Cu+_
z@lv^YQ4ta5UQK&46(~gjt(R*$5qx27Q5d145Q1vY?Djo$_A)G>xx{V#O+dg0_+W?=
zx90cmZaqZgt;`PO0Fk*lELfV@OrQ)o=d9NqMQCLs2O}b<c*i7|42K(eK2<Wdi&l}4
z{Ikp27wRE!Vr)@6Rb#R}OD2Fz)P3e5lSQ876us#AKeYf69$Wc-%tFUxfj1%K_PQqn
zU~x)7K)u&vyO^Ex?AeT_#V)(%8*p>N0l$3%aQJeMiz51B#C!s>oAmLc5_o>75t>ib
z*VWdl0yGY31rQ#`+!k9k1K8QvB5eGZscC3n&*B$gCv{ZX>ka{?;nBXlt50P?*jZ2Y
zDjtl%wa9L^M}RLqb2enI;mRJM>zXq*z!uGbb>6jS4*^2wBA8-`{s<uKDF`Snm?|KK
zwduVA>o<cG3Z2vPiVAer(LIM!Mef?Q2E@mpqmNM6=J_B#gEXMtt0I!qOSFOJPy?s|
z2Q?lM6)sSo#U~`3f-@xzey3paQ)f$|gC)?(!l~SXf|^kDs=(2-D4Lb#(alo|;CGt4
z3qXuKS_boWzN`wONFaHw3%{fZ#XBGl<bwOKr0T$8i!waPf^Z{Iw_GB86hcw5yaWvd
zx5Sx@u&^-s${zR5GY=opRNUcQ2QE5CQ^znUG}H`A-@%a)Ch)8RP?xB3r0ew57s3I=
z&2`Yw7pFNR23s?@FD6r8lNCJ6+?vG|>2$M;7Xz3Rq`;L#c+ql#H-$U`u#*-*w_D;B
z*))U6`!q*!0*~|wR=`HLxPblJ;C5n4;FD-5s~}_g$4ZL^{9Pj>P6#A$wor=qn{0tY
zAh4zl>2LvP8&1fgC=U^ZKHauD;DmO@9Vst=4h{|?6o?6Pu9)xsV%kL(&3=sWCFGPQ
zplk?;$;-Ql@RNRTHhXJzbwQV<2{|@J1%)6e!FJFxYeB?d<ZxygzXaJ_&bN|YULv<f
zzUbshcxYelG}{n>m@{U_qzwg}tPzbfNET30j;Peox|0=W*k$1h8^M_&+R>L;NCjM{
z%?p{US5JX!wk(>hG%~mZ^)$H7-<d4%R^QK-*o~>6j1D|EA?0S)Z*_#x_Fi5HsR0We
zX6cFGZIC@esuma$k_2^?F~D2TU%Y4mI}9IY6SBFA(rie~(2mC!J|M9=Td0?*eL^rd
zGu_^qr8v#nkeXjf9vHOe?T4Zt{vwrC(Y6n{bD{EDfO#=-*pCd+;E;2+0@a9o89Jkg
zdPQ~xxsxXN2~;C}t*jhp+g|Yo>*M9oFVX_iJd4$7@Rc0FZEjFjbm^xdqP5^)M(bD>
zs}tK&Rl(L5fomEE<}V%cb$;`oXQZxP)tv(Y$p{g+**OU903LzW#X~`K>46CVW}${9
zb@!T4)XOleLWdpXCj#rYNfKBK^`1X_wX4QCys4=sC|0<za91%c&KIL;3HbRS!*9k_
zQqtP<PVU#XoA{~T(o)Ec*=50U04xMjErOQ$V&vZA($@k2Oa;n_0}5VgNI)79Y5m*j
z14^pfUx1_)W^hwM1ydZ88Mh5kr)<EEJ*+7Jb;t#ry#6*43df~@5vZ<e?`4fajA&qS
z=u#eA{|1~El!cXkEZnldGxN`%#S(60WRyN0q29HtrjmF{qM4{fplXodk*!w{6wue3
znws+CIRF=@!12@3(GiGNz$d^<@l~8Tz`zg(^`CI+2!uKTZNd=Xlk14I%eEen{#sRK
zhJ47{8fJRUnL;%6!>s8CLX;TFm4e7ucf4av=LN|Speok_E|~v}1ei%~#QVvUaeyW4
zHsV2clk$ji4`QtWa_z`4^k%Pie`k1TeR(<^FkQ8G@7@6sr+nwm2XxiooZO%eUjqR1
z&*&U1lAE2aR^`t|061t7qCD57OFH0?0i^*Naq7gpo2%;qfFU5KlMVP0ekuN;Pp%`0
zm2c!OXKks0)~o4P+%6;UyD{VFB(QA5IBz-?yV5@^uqK>!udXRR!aeEfQ>w|Xd{toO
zlI+~U#V!1}G(DP8#j1Ms>SIWGSb|j{#J`&=Ue<?1Q&^!q{}w<ayQrwFHe^A)fW%35
zqe@_BD0wK@T<Qb}BORb}6?DEpTLh|<AtI5Ood$(eH{^8V-G#Yng;T<SaJiwp0*bdq
zNQ#6MQdV<7_c9Lofhu@ma1IH`Is|6MNb>H?&ISbAroRr>Y<HwVyPt9B;C7v4DiuT>
zh}L>~JJK59ctk^<6SS^nVUY&bA6AJDU^heZ!h{{g>JCnKN7rzc?Fl)wcS}oca3lom
zC$vzx;b*wU7%=DF19)h&$JRXCL3Oq5Mq<)Pc_90^PmQx){=`zI;rr(1+D{@?RfOu<
z`k}?{f~)Cf1s8WPCP;ni`E+vcTRw(^2jih&MkfexWulQ{7)e7}yE)Fj{Fd8gcrVZ?
zrd?jdccc=A;Kk8L!4)GV+}kfEWH-pf@AKm}n2Z@$gf#b%FrN7g2?_{g_`{@soY@T}
zpY?t+{+kO0ai02`Zp>sfZxQeW#l<KXS<zqy03jht$A1AKn3r|9p+&8YIKF>vDlnR$
zj=n1jkW5q_*SMV)61qvybI@baB*z5OA+AI;IP`P<FID*eNTAqo^+9{46y)zHF^6-*
z=OiL3ss~wiD@sg&DtDXtOv#(D4;yC#xg@_NsBQ`%8$$vNMDao#gJ7><3z>`%V)h}5
zsH><1AhlY4f;N_504CNzKG{>ObM)v@R6!y1srogC8Q8Ug?P&%dKs(tH_%+^@pT#Xl
zfp@OzZl%DmaeZ*4q-10)wIn1YQ27Nnhe6Qpb|$7eL*j%K5(o&wy~C9F_JXu?2#G+z
zQ7%FbX9Cg)M3jKHKO!hN5+1yB=T6AUUb1gOq7C?;y<`rJj=nx?gpR;W{%7^y3tU}a
zCqlmKH-w7k-rio5A0U;=0mw$S3BW})K=%L)(1YAkVgBEN46Q&?ndYuYpAr&!SD(tb
z($&WW=K#B*q^9-;>ZVIb^ai;KM4KikRKkYIDJV?9O$EdNk1mbMZ3r;jHE_*dpgp~T
zu(N-RDFiIc)$$Qjb9+?410vPMP;GGSos7IZ8_Y_SwjaZv<O1;wK->z3^&;~x2G*fo
z90Qa&otTM5_CV~1ll@<jx{lhI|7&)EVQCWh5=bncA#ePkr|2da1l$faQ0+B`p=uEJ
zw56xmCGgFg81c<9McyO^SPKMgw^yKY;|4$sL=DQ|<56mxYEar^;`jBAG8A#_oDfwY
z2HHy{!UwEKEQWsLP@pdo4k9BZ%{C+O#tQ-9fb?yPkh3FU3RIq{#B(4s=sQL}YI^M-
zS_RG8-&;ijH@M-A7bKW5DOJP4+2v&da0x86913ACfgWoPgZju6O89^nsHYRar9zl2
zfI2mN`0JN1ry%PMIKar5bo9!T7K>S{#KFP4?v)tEPr^**_6mhiK&b%MdkY2X8`uaC
z7+$Q~xocMw<P_|(2-yLcr?te*Me_|cNF{`&{~5;gf3OD(YoVa+KP6*lmkmWxc;$Pi
z;lowS#m<XONU1|Gg58N^z4c$Vh<{x(tt{R?8PLj=pC9O!)2XMo7cPjs8XfE4kDak^
z4}lWQT!<&t{APgp1p+$kM?OhJi#_+<JRxF$%5qw`dw0l<Tg1o4hAUyQ<8&m9i;7Aw
z^R3a&T*XUo?q}n3auV`)atib{h-!|X?8t9U8oVnw*=5o?)*Kc_g<;>1N<?pEw8uOq
z=e($usZ~>Ql3jOp=i_+APTY#siyz_gukV7ulb=P5`N~ree8<s6!lKPpOp^eMT*}#z
zR${$ioFXS5uBdRDho_-2`uCS6`C+~%2fFg-6fdV3gQO)=1lah)tgJ&3kF^>vVwi*z
z;OplK?4AXOhR#<%v#`(~ZyKsmQ~)jd4N4Nm8`9pi7rdUF{8n*XOGi=uwSz$OOT}Gw
zZg)~9)`ectVVD{Xyh#q;#Ioo|_%K~;SMc%blbGxY{;%dQj$S3h@LX_J)z2V(cz9<d
z+4{hNgQx77GzZ{t1k2#1U%!5@x1}z~#Yl-^D*~kaAz?>)=h6$5(aOrwft4Tpvd2GI
z(-uDID6p)uVU>n9m)9Szsj96r?s4PJT}w1UQVbO9JGgJ#Jv1*)->~q8i=4FuwK$&p
zr*kLOclfI|NW$S4|4zlzc<SG7I^oNW9cD(c@b0L7_RIXj_Js%bAE)2wt=rKIY5R$4
zjqdqBUO#$=nx8TJ=T8HhDWIH<1{Td?vB*q=hUMw|(+mxGy2!@S7)>jgsppk?xhmL=
z9<O2@DXD<l8<`G1bTlnyu>Zo6^j)tCKUfn|4@kVO>fwo$JSGv{_{=}JC~Ufg2UYfK
zIq*qVmt6dgXPB>a)xFpI_Qmwws)oWh<fWC@$XaWS90|YEL^mlTds?}tboIStFLNDk
zZi<GOuT|SNQW4ZY3n^M$P|twwP&wZaB*>o}xKAaV7MZUVR)UQs0pL6IhMYo>oV=di
zPCVDILw_EX5dL-d!jyEap{49lcrfa%&(S2w`!UvR52mValNoA1l|8;-;=E6Z&6<ib
z<TSh+Dzb#gya~n>lXjEm?OOKd{8I(1*hmv}{xDrR`C8|ud|MM?*Sr)C+^B>i>G*y)
zKGl#^5~=DB{>3H)&tOW+eyqMYs66E4sqGd{iSA6W1mnzX%Bn3@mXsR6ltO=mbC)iw
z@za*!&<hV;|N4T6NX@h;y)T&+3m8!f<mGdwQ?4!+>BIWZ*?!RM##;8G$}(2erN!gU
zdtr04sfCZBp}_@qQ5+<T$by4Xb8qa7-hW_%>9o&juqlo5QF4VN<Vu2&vKG?C#mWYR
zgnv^Rt!2O|Sa4Jy+v9%=4uuuzQKdcB^tAjYpL#Bkp2rAJeFJxeDKgq!(utX>Ep8R%
zZI+TEN%&2$^z9m>4`YTShj)GikrU5Nu5M*>3odG$2DNmg(SdsqiuiUKZoW0P2oxcP
zX@2~RFUT_&x|nC)Ug>mL!mUd7F^1(muhUrK5V|RM1BVm9lAeMf4oTJiL5GN1rA2Ux
zMs;24L~^%&{~J*k_jf8@@gx|2$Ma+>j6MV0+gk&RoKOU<3h|y-?|D-=9klxj7jN2(
z37iHJRD-_)YY&+HZS8^AIl0zOh-l8hsvGbC<UNmxH?S}Iz<Eyt4lyGGDZD!}ver+N
z|Lga0@GljY@zNE_b?4_5B0iaoC?p5G?#{N2R|jNr_C?8xV|8t<Q;L%<Cttnt71sqn
z`M7Lj<qMO9qN1ALUnd5H_Y>;-#Ms!t@F<7=??ttbN5OdtyIzzY`4KC7&TdLKgqF=O
zQTyqqzdFts`^lP;dqzU0A2)4d_T8IMva)JdefTKWW8;eUr-|Tr3d-QrC;O=&W-@~n
z9ZIsgY~Tcjr*4usKmXhEjOpt~v)W@?)DpAR=9YLbU77j7XwKG^2v1n{W+q+;op0uS
zL1L^6bx~%?oxsj!3-sV7PDmyX*@YI&71p5n-gczMqv*qIM*>MgIuh^$NbdFcvGLZ%
zYOu&w-=jT<YCs4Pz+6Du>na&--tD#2{9bt|)+oOEWz=6?=nZn<i?7ftjB@*X%dGlq
z^6N7RV2jHCAbFAUVL*5hkoFP7Kf;p%ugy>BSmUtQN!j&0eA$BjxBo>8B}EIJQy5Jy
zf9IrE@C7s&t4_bYe=Otu1@){QkzySn+rNDIa!aiIK|m=lBYu8uk;oku0pujeEn04S
zE$TSiue!Vk_uY^OZtsk)>tzG_`tQVs&v6Y6<W7#{@|h@bL`Cf!Z-`=~JfVrc7IfN5
zw{Arm3LQBjarp4zp&vH@kp|2JP?nDcT<Ja3b37jPs&FzG4_XZQJ{c%)cRGn1y_Kwg
zEA2BQ{>f=hPFW<GhN`>9O$3@dASnun0qEx@U}`umJ(~3~4$kAz6Oa2Dc-TEW5+8KR
z$MNjeOMv|hWQ_GYE9evY(Ctz3;@14!9OKf+@G(~lE09${nXt9BCG5VDXSfP%-Q4mr
zpMwQd1wFuP>>|+b-yi+<?IBqeW8;J;yAP{G=d`_}4zAr;r1vN);_7NfcW&a7R_y(F
z?r(zEW-A5+(i|53_5(g=krxq4J^N?Qjuc;@OGB4C2-ST|3_G9&;{i_4+0#@nzMTSD
zT3%$0utxjPx!Ro#;8oh4vt)J|KSxz=$PvKA5q@=C#!n5T2heDO)Bz?vMNS<bZ$C2B
zht-}`e1F>K>#uTCB2{n_xsnu+2Y-`fH{0my>au7I-jS3vauDLNWX9Y<P@6!Qu&*`^
z?CPp1(Qq7@;!;ehJBINa>n81!pwFOSTx|O~G9MEg$RmG{#ZPPKS<X#z{Eq4J(y^aE
zGb=7Ud{|R{e!GoC_c44?F4v`qc<23QSyY(s!@85PadNS-Mg}|Uu^LIdbg}x-yp!qt
z7d8bhQVcr?nL$s|Xi>O)WwJ=d1s`x?H>9Fz8khtv)J#mqhB!X-B+adg<T(A>!$BWK
z64^^4(QTSd0>nn^cw1^HYBNFo1Fb`mjOL#*neY^p@(l9r1RF~#jo(Y`6JEH4CFuAr
zaWSJl5vbbUfaL;i2R$aEfSfdeHX2kb1LJc_QWtdSvrGdy+V1Ven>9>hkro0SJPiRY
z1voS4ebGmzk3e~VR}z7rN9CNJ9wXLrYK;YJ372My6a>Hu?x`sta9q5+ZByNasMtif
zzB6cf^z>M<mP6>hsnRY~RbT+glx(g_2#0LW3l0tr;?*1fkc34JH~!_9F&4l;6vBmt
zf{?-lNSs>J9MhGVe%_Sbm?^_1EjF$ZE&A9HSaTcO^`S}BX##wm)om$E>C&hJrj*P0
z051j_MW|c*wJaJxctHZMtF;$vq5aDmm?ArKEz^NDyck4)&a~vTf*LP+5{b#kUID$O
zP5YY$88wq8g6N^7cWOKUe=@3(m8|$~^*)d?B8~Tx+owSC!mW{Fg}aF<4U9V6c<olI
zcb3UeO-c8NtX!>vDtIy3fK)+38l|Z!+)<!qTM_4knmGahu=e35#dk=SnuP%DDO^5J
zLGhIWashdO*KWeQ0~Vl-iU?RVYD{|pqKW!HS0_5wJ{DI=(h?@^vU_2oJqw|`#anB{
zbGxsYwWTP7Hb1$R6wfFF-~7<^(RE+41R>uG5*{F#1c==WA-wVLo#ie$Otk=mLmY=H
zv$AnIB<?Ti!Vv#(YnMEnT|KqE4R-t;*yPw4XX1o{E{EcKHE1V*ULekcniQCTzdM1y
z-MRD48DYuKC7(0DeYn+o>%<F%zLootwEP{0kw^;!jLGOL8CK;8F}J%JTKqsHY6inc
z{oDz5@KLJocxwI{Ob_}U0ClScOtW;odzv2@W{};0ESAt38}%w<I+q*Vc2idKqG`Qm
zV21dk?e9wQ=>9G<4jpBN<^%+4u+&0dL)_MR;C_5AU%7(hn1IL`%&kCoTLLiH05v87
zzySbZ@~eWzkGgp~%m72*eVB`jQw8pl#a+BGE*4DcmVWf^h5&xo6)Qw#p4(l~G7lJl
z9kfN7!nMoE$|7hLXzsA#)7shB{$Swv9j0`F+DbbqFGHr>E}-REpw{i~eX{(o2}iiW
zoF{jyG1)?>Wl17P_r3w|3Lr=jXl#Kgb88>@&rufoABs=~!|^oDwBx7KeMG+~g<EF{
z!G0s@O-<|-r`8$wLlKmw1m*&2Baj%WgJuz(aOnC1s;E|s4>WF|2J#yM5(~6rZ!QUf
zAPvNk%^z=9n5J0pGynBl@jtZ^%bXP7A-J1FO($C9nNoll>IJI;$YYq%#TUFZasnU)
z;a;A7{FwX%JNvZ&_pR0YCpb6=Kh2^oHWpsLd3BVYUY1Au+SZmz=O-ioKOzns{u*&j
zE*t8GQrCkC35G82Z3(Kl#_0x$Sj~?zDIcsc^(NRIdA{Pol@ch=fy2~*=)E{w#iqF?
z;^E=JBJSRm;ri>#PC$EtF9|~I12RV1i}?K6Gv({oU$ArX^Q&EXvis~TZJ+7ECD$$I
z0pH-@H!qL0Z%tojx-`5k>o&~uN;4yBkb;_ju-}_6oV+D-eCZND3Algn2X6#vBRjxZ
z^nl>3c@B^RB-qqlassXL8gy#dP5e9yzY9GgvOs?o-0cC;_5$EDX28-3Xq+Dpu32)`
z|5ho;yDGtTfh*$Aub!P9vfTM}#KNmA&^WtF!r(LndO!>6j7Ff>&kTC67=>(NoR=mN
zLDYui%3xp3PiUB!=Yftzc*@PwK^>==c-3M>BPLj-MAcfsq{{AbcOGY8K7I4FkwN>F
z9l9CPg%r+HUHN`iI&g@9`a^zM-Pz*b;|K9Wd3kw^iTf8d1D9MgT@Eak1unpqUtsb)
zoNycb!ZMLiKU)k47J{ZEAMMEhc#NN4y<@(5O1XvXkI+Hq2fIQ9ARiHfceC2<pmi{z
z?-!Z2G6&<hzy{yMYFOuwG{elKs}{oDO<bW9!3w)*d3EkBpVGy<YucK|8f^B;O)`Y0
zOmPg}JDX<Q{U7|R&9Z741?)d7C^vq*7bR+mfyUb+e~96rMhpL@$I_J8=^C!t?Unm<
zQhP}y>&VHUU6v$aYiN?2lsz8Xk)9~<E;Tgkfktolq;)XSrta7S+dAEK4a2xXf#e59
zVlBh>+SCeyhVGdaoD9TA8m-k*UM26}+B_>H^t{IR`}{giPp`sl*sA_r=<xUW=KT7d
zO9C6IHpV}0hc>CjQ5}BX-uaE`qjqeQN}_={;duXHF=-FNxuPb6O-B`(_9N|gHE+6Q
z?!Fn~W>Iz0+nn2+tJC1*l_xbz!PTG2rJwF_J$;wV?#2yY03W-m&YnH{;@gU3LwY~k
zf+g0E-%2xb92(dfko8%3o7B3MQm>a?aIF>h-Zv#JS)<W(jlC@R@&$=p(2s3{R=vB5
z>4Lcui1;>sewaD1;G;&q$s}<xvefLb{%YR#O6`#8M&+HX5#HY36~3?}X1v4Wi5riz
z?$@`Wwj<P3dS<s9s1;Bw-t12Kxc_gz!LiALOY>dRkF~DuUj>Qg+tt1k3^EbqD=Oc1
zj6ypc5(=`MIOY4N767T6*$E1G@MKH2$>oX-d@6hY=BkoC_(F%Bt`8Xh$oEM<74*QX
z&X7^Yj%ZA8Ao=@y75W{4pTdDK1+p1=z`0b^)dRPtx3}th>*fk1bY{!V%vRRNeKU8G
zJe0^=a^R`sW?N=cG`+DIJdw$nqwaP39PGQ9W%5Mygq*K@pcF%FWo(UD4X>Fc+Pl-B
z2EU)i;0lsf#J%}vf+l4YlsfbBE}#JFj=s`kGeJAIUq<mV?=;0I6HYjsQP?h_SD&q<
z5n9O*b+%>k#9Lx~+O?73b!^RlxWB+j@yHU+ll`ttB#=<J4G*Rk0wMX+r#63{xss1J
z&m1%FIn}`5|Ff3_FYYzLJPMqpF~~{Rpvj+8)_E#_#I8<&u#&IdLw|u<`f(i-MaW6!
z^-I`H{7V*3kmy$7;NB2Xg9g$Q0rKi%&QJGT0VnA2jpxwb3zi3rL#!sI%T*$~2T31i
zyrgMBx&csdG&MD4gA5SGtnv1&_|niJ^{-#k63eQnt0|6`5Brkd*zdQg_jBjc13gc=
z{Whu>0lSPtH4-q<h@1b-dpbju+vqwseaU`XSNWCMg)7i*cKxu5Af4Rp1!>P_>X9wp
ziLj}jJ!@iRBbRw4mZ5bWTA34}G@yadaxDN!z{PWG98%iOjLBEPkH#}hZR%<D_v|?m
z`MB$VA28peMe{#U=;1eM`rYkuv_ORoe@4dRpmg3c8P%zy2P4ygxJKDMpuw%^jnD^x
zf@A4gYx)8-+O8KmVGVair^~69-7%6AgQN>*LDftE8aM?SfD=A^xQse3fPS8uzDFYA
z6J1J8Ht~EH`5V}(Dq=&_BWQsG)$HVS9ER6y?Daa6tj@&~$jGSc|M-5HW%;q+f(9Oh
z>HjA%89;9T+87F2Qd;jfK_{iI=CQKSdgKD8dBk;ZBw1L+QNSs0zPo!UO(xWrjs}bR
zo>3i@4PcSu#mgpH&4WL$#`iMsmiVkg)j*P#a4jiq>2z|MeIOZj+28Q>l~42+0w?)i
zAH8C<UVB&P_VKHsU$6s3&oGw0n%X*e(?gxc@nh-)C%-T?f_k#Cz;TXyXD|xOV-l2k
z7Tw=}zhQN&Qs$b1x?knlKVJBChf}yC>$Gr(y5I0^i=%jnHlDQEy#cC$(blkzP*}$g
z)4&OPP672yEnR>6!er**v$=E<k=#$Yg2ojN2_^d22S!&P!ao$;8~F&|$`u$SM?;e)
zQ*Jx|`)%u00~#jzdL6-BhZzV1Ht_Kqm9S>crd}G%^;y0bMyltO&tAdxVD7r}5W{vh
z$J?8!<o6sYk2g?pf9D~m8+}_yO%B*cx!f%M*XI_KLY%w>xa{ZYw^W$%Wo2D@M`rd$
z?hR4M*nE?dtG=Xcj}HGaAC&<WBQ~Sp9Kmh>JrcKWP4UvLye0E(ur*&#G~^y->vWe3
zNYS|IqO<cmM#^aG;8|8l#eaEEry|^6zGD){7d~ztz_6KPr0`_!M=~nS_`L0CuOzaS
zy(ElL@vn4#Ci66A2GP$`kQ}jNDcp`uxqj|N#MJghsrdVu!uSEY$i2>fg<jn+rG|pk
zx-EV#3f^3Oh|v{X@7ex5c}!v+t=PS5yU4}#k&3ShPm^M97WyjocSYNeI;HT#$0d%(
z%)%4qZ~yLzuJV^Gdp!T$JN|gWkmE9W>U4?iXsG3LyTNPuA3hXtiWX~)yeGELv#lRH
z#btA&bF{6F{ykbqz;%n;-~7BK=3${gb^VCqE6q-f`i!GqeRbSuZBCwG-|g#VTyg|3
zEBV{IC?7Tn_8!Qum_G~D_}Va25CINB4|F3a3ERZX8u2Tr)Arrv)=X1_kpj@$0L}9+
zq=_5AvukQ-1Z8GrAp!Avp4Vp|mMwfk<pEmq*n|WN_-IO_4R&9{z#ttFiqJg_<emm}
z832<K7#KK|rmCXy0mOY~pfCi;<PnITC#x7J{S7Z|uyn=@{@kIkuelnp^qGqPp^CF}
z#<61}!GxzzpK3S@*P3BBGQbmpLplXyD74`==2;IQv7)~)O0YnL(;9UJdH{fF+-z)#
ztf2Lf7=<lA74DLz{_dT<lvJ!-CN%rNpO;=+93gMWtErI#wlvTG_3PK_H*bz9-v=5H
z>0FS$!iTru8eIGW?);;DOczO_{5hGav`5~Xoj60LZ?`b247}O5F}o!2O2K;e?@(Lz
z%8H}EGf3^=-+%?srNPT<Q8}IiiSaQV9j|LUvtKFTOp=n4%C`yxC<!!MptZoCL97!H
zl6n5zTwL(yL1hjg-l4lr4MaMv#V!td&{~dEAFzQkqM&2y^ndMo{``5Qqe#ycf2pqF
z{KGRym+4~rWK$xqZ}Zshbz&Q{uHXu_>P74bYV%Hp4hyJ-H6J`kMKmVx|Ng=-`lQ&!
z&Js#*bhRL;-E8RSD@WU9NY03rqiG!oS%bnYxuirNh(a|`@}U$GwZlUWU*WxnfO?yt
z^$&GFqmBn?4%_Trh%DQoAYlx9J|o0D@s-9AdfPlfD=1Qfx_h?TVekbsGlJ4FC_Qus
zM&gV4mQ_AX0vB3#hLX|KmqMKclfWO9IQ}ssk_*GA&X60Ef3k@G8M3Z61g5DkjhDW~
z&1=sS>0C5wtSJP5FdG>9Pc0>S7@KrRqhF%WP*NSk=p+t#0(tcJ(}Jk9|JHXuCbKwp
zRq2kmC2(7OFThkXDg?nl93<s`xbKoHDBwI`HUGya;a3~}^7%b4{!KD>hV!kU&9323
zWS$Q>fsuMk_nk?I!nV^>qMVAKbVu|t?4jft%fSQRA5fC+1cylV;yFge@VV%f>X}>?
zDt1`NkuzMRN3_W2r4>GR*&lfcmq~Yso`ml7L#N~MzwRXB%<41SQXfT6N*sX&{_*B@
zX$6MbF@KV<!z7F(N8ifWVUOAV8UsJ?d++l+rbD|vvqL+smgF9|D!PYmS1GZFB+Wzk
zpwC>)RENy#Xt18v)KIg)=*k)rewO0^y02U_Bz9k7Zi>6nqeHg6LP$vxUOjmkzF`<K
zEHj?)YH*<JlYZxo)%w_H`}O8u`Ukd~^Bnx3=vr6oqSrjQZ~r@9$}pNZE7)Q(-&6+!
zO$VD|L$kizs|gP#1kB8Uc{HhipCv9%mxgXz8Rhq`1cHj-qIL6;J{E-cyOw?Tb>_jT
zvYAbg^+>`Ts4_}Mo#_&sS>g?*`plrBM1O%Ll4&#<>i74T-tt^1r%t1r^bK9pGGv>f
zwKztOmB}dN^b|Lbb+k4Fr|*nh{aXWJ%CH%dGrD(=3rcF3XkwiG1BN3xfS3kx-3Y|;
z*MUfndEBO5*f~xl3Lr`xCKCW03u-M*Xmo@Qi7p(V3H)wrc1Rrzs9gg<WSc+lz{ghh
z9dF>!g7F{VMc+V=4LA$qt~_fbm68bu#f=F}@<FbUJrTMmwv55_u?;-t_V8bGPY~Mr
zIIi~_KZ%y{hrvo$72;K))u~`2vleu?Cqa|u(vsCVnb|oR5Ca{ek_z$7ds4m9)zsBz
z1$Tgppo>&|`Cs%w;D%(;6c$jK758je4O(53l2UQXH*@Lub!R)pY1~eDm5B_a+wr#C
z`XSU7;3-iXby8|->#wgq(33T)JOa7^2Y>!4&?1A&RfTaKY}(Y`S2EjUGS5|?(e=L&
znheLLjESm&<0*(Kp;K4@V4FpdzM`JoKb#RrE#w29z=SMlvVmJ=0`ni7e?8*>1ax#H
z$UFYSvBZLotdxQdb;MlLzz$vg$KL44{x`v)+`#T!)Pd8NfFs^ZW0r8VHMcqSz&_*{
ztDjQG1i#RQv%S=KCvZwR|CU?PVpZ_2Z=cHJGB=IR-Ayp}cn5XjV`D51PFM?&QWn7~
zswaHDE*j^q->0X=JX4Kvuabx&mLUnh%Rq||>=vzU<lco84bt=Y^W0G1)$NVvqQtA-
z23JhSQ1Dg2-S$gj4N0FGNMi6Gm0YDVKIpjg@bB#w*8Zw#)l81r{(j-s<by!BH+qTI
zi2&EN<f*mGv>Q#<F`vPP!M9>`7gCBL{lf11?){Sm1&7*ab<M;?<?3T%<7b;@iEp=d
z!3OYr?|*52A#ttBD89Bvw`yOl1!)u>eMnnNxlyry_3gN+rVzeQ1`cARS75{)%OKT2
zGAdqlQnj^mRVV6fl5DWgD9x6j=H?ChAi8X>Y?qzPEHQH&j%Y*qf7lXRvZHidT3+C2
z`95mz{4SAj2QK<QPJKJ__T$&1eGcEjHyqJ|Khn{m!~4e{;mR>`>Q5KRNzu4C{H@On
z-VzeX70{jjamVM%lqk72?e7;Fzw^-gce`k5oIG5qxXilmgXb_z7Q*PRr)U5JW@>CD
zdg1mW4SGKVdOuxNvA0AmDjDd0<t@`0-$a*px$9<^3svOa-+$IUKP^gbwaa`6uJh3E
zwacR$_UGDNA9d@}9eax=YZOiYwBW5`h!Oh~>hq;*A32LOI8OX>Mkw`VD#U?0vK8hZ
zd5lKd{bx@exenu#R6x@LoKgb-Gz!k{(7J&JnEcVp3F48SJ*ii}-4Z<bm`<Y~!=F+6
zN2!;xC6ueEAtXZk*~<zGUaGwivLp_n?!v$KOr@a|m?WCMcRL|7Qw79BQv)ELQP<FT
zuQrWzm!Jef-T#ogtYkjJ=su!rMSYsl&>|qn$;BlH-FRV0l?&R<<(R+Ci}GSWrza&n
zC${ZIxe9DUxe{<YRZnfCq`Uwl6B%bed9uI*de1u%UdQju`#bfyR0BTR6uRJedi&@`
z?1b|z>dJ7BO3#1+mBW~V5^%t1E>UY+n<-#$0K6?EWPu!MxxO}-PZP2ox9;m0)wSOj
zjWpVSxRq7{!3+kM)%^I;26$`>paTF=o5FA*?Go2~P_+iWd>QFQeQ>R?9iyWIi=&H!
zlm=RL`jo+u^Doz<Nj}iu@78@9qjJ374>s`ksRumw57}#Y?eAn8U^oA;9yPxGAJZ7B
zeXfoe`Lk;tf*92aTJO7AvaPLY1DWRt8SFPxx6BUG&8*B|RL9`5c@h#v<EOU`lalp@
zFoj2Z_TY~|5~{7_l%H>diT|n~JMt7-#~4~tL;u6y)pSow$+!eb$rRR*gzaEs#UvkN
zG}%Xz)iXO|ThyxWSc`T{c9Btp?fw0d`b9j$y_~u5J^a+@?+^c-!};e@H+=6!W)<y_
z1AF_O?JjO~AM5+B1>Or=@%mj~<8p|3Dc^CYGuSVZGNJ6GjPLn#_g(c>YMO;Lo)XgF
zlQ4H>eTy_^ssreh_8sglc*jjiN`cPVij;N60hY;|3@npel%$X&p|5S_YM%so(cO$Z
z)=7JdBUk$j!AE0#f*RbOhd+D&mzIuQephicHdc1;-ZUBSie8>w&MBZe#ORWkX{+zO
zhApo$GD<UiMv+%fik6N=CBe;@alMYqNoKD5y>u$|p`qcLMs~Dxyb>g0`8NDJ(K?T)
z(I1p=Y<zFj*qCO+&-9n4Zl(rw^8)$E;6rS*7}-&t^@R(~^NIRydg4s(8RTUCu?fLT
zpeOEriiQWmC_Lb(p|h+~ww@1qSzvIP3iQQ+dqG1n)}sR9%`=oi8J8s8+=_vHO@T%!
zPAF1f${(CCHK+w;iV-mgP_?`133&5{*WS_nDF&tb@U>__7IY-@L%SCYMEKrY2c)P3
z{)j|G<iFrl;CGM5S@Bo=Gp7P;K!dv03le9dU@C?_T)~{f`934SnHC-PlWH3oop94T
z-{8T&hfwDN&0(9}kMYuYG0A`GaZsW&4h;)pFdB+4`;h0uQftQ}%<uL6Sw&Zv`N=kA
z=B>K>_^$U~l>8T+511U}`~N<t`7byh*p9uVc4&6i-#ech;qyP$Xdm-#RrY(9&W~BU
zg?JO&M8LQDe1*{%y@2at*h322KaGOt#e=ddooL1u^pyZSo7^};Q_%q3)zQ#VJTy2s
z3h8gIzs}PjtGLpv<}qmb3tP8!-8hhLC24S<6|Q^6>8ixJhC6<&qNZpr2^#5WJ@{?!
zaSo2EpG8htGT~@I2XM+us?^q>dOWt>3@e_{NZZ@nN7t7^t9Iqny~kfXd9nkI`ltjc
zuWQ%k?8oOf!+Tw(MF*g-X3;vCl+sOa9^JDD8S-P&kveppc<+*0-ooo3hKP4{2Qm-|
zOkp~yU8B@Nqz7TGWj%H=Y;2l&YLpn_y%p`WB65sw@9z&rCvMXoSph5?_3a~lI~q;t
zAy^eZN-TP`6}?)%I#0EIL}L9XSQ?tpw!JYTjz)>0hIJ3Yop@PbUjQI@C$_y|i@E>`
zfRNqnj=^6spfu>8r}8)=ksBFjLB<L*Z>n=>R)Fbo?T~iVRNZ2cn7K3(W{cXNt8*+|
z_4LA_eQRJuW0lq|+yO35G0%gwpN)K!;md`-Q_z}au)U@U(+Xe~$u-o%K3z1w(S17m
zBbta%=`TN;4u^_{h6ZH33%1Fw-X%XSjf#hGbcy@V?6E&)E@@82x`t+-2!$)4E<dPF
z(0m<rO-&Bjf&srvU^S5*8v|@eodP5hYGWN?h!*;>;NT7{{^IxZ^`}6u1F%dpr0KA+
z8(pToQR!x^2pgY&D)B`DbXgF^{-p@la2@I+<VFA|g6Sg3NMUShZ2XcP)&{*-P4ztF
z_$(O%%CHx>NY8uF;2e&oPRb%Zjz!7pJv5OIwbVxF?7(=uOK2}1KHeZf4bsr83Hw5T
zCb)w%DCv?$i5OJs4VMAHP=}lcWXtP?oLHa00uDni5<3G~X!c8_>PCw}s8II-Y=&`7
z7!S`sMV`W;%ldcE6ifs35TwM49+uEr-KM9Fd<S!!Qh>^DZEZzeZXm`Xpx!|bL269<
zfE$a|IiJrDvf&pbUY|wD>woJp#n?ZOy8kRt{QrZ@|5GgHw<)1O_djUh^Z)K7s<-~{
zC9?l>56)3uXe>bO0i~fNWwmG&H5&Cj>u^2<rn%|@6P%QuzU-iiKbZlYyDNntwJ+O-
z9`>a4k~KLP*cPB^zyVqfki!+F70t<|m(cd{0PK*$Eb83CkzR3X7d~h>d5gO2-QLK4
zTbO!;rqO}&115CEL#CA!7bhKnhFTyX9MplJkn4c?0Eg@EgA51ecgM!ZCqqAb0dyv|
zHZ&ZO)y}huLWp+n#;_o2$blY99K>qWREI=nKSY;xyREBOgk4u=ObjZ<?6@Gg^A}xu
zi^iQm&%6*b?eaVtedoTIYyeFH5{Q&TJ^zTdMx5%efdMpl8Q_>gBAb}27D#r|Vg4CC
ztX7^k|4a816wV&)piyykErR*RLw3U5rMXu9mtmdU7cOXk+jhw9haT)E=-WY;2;&UU
zNMKgN2s~+%YYXTY>L{jXVnU;c#$iIVJppmufiUw9bss^aAQ~qEL)cf(Z_!|mG9Z#d
z1IZwdNr1o&^Ly$z*S$g5;q0o9#)cpm3E^`y{S{Eo4!c_bd4Qt*8>sVO*taISc8xUM
zL^Qb;W{YJ4yAPThmyS=U^$|83x|Won4GM<RjJR7sB@UwjvVmkxhJhk5z*G~p=(3}^
z7hHUN1V|_#_(zNXf9<_zR8?8pHF^MZLJAX^5*3v!K{BWaB1KS8P)ULkizFZ*QBg!m
z3L>Z=D3TEhB}g^_ib@VDnWK`ElEdN7McMD$-EVc@zTf@E{c*=P24fV3`|Pv#UVA-h
z&SyRlZ2*HOywt>DLFMJ0N;zn6lddTUQXfM#3-M*$)BH*w%Viy==$}?bG<oF22}{WJ
zHKFD4>L^_EmQ>vg2^86bz-ikeE-qeX;L>_lA4)*xvN79CkrI~yQ4V_YWDoVyYC&OK
zn>pS*e!V}QTR_fR1aC!LdU|D2db)(ye7_MJ(QKO-YPDvUw12(@RFzRQXhlzmgZw8K
zhw#44?ibywPRKhJX9RftV2evgD6^N2Rz}#6>D+&uh_r#4S0>dRWs0`wvZRcRz3A&Q
zj=~ZsJ2+&LP|y@%r84+C1vIF<S*~H(^BS79!X#i!Gbm_GN=_Epy?b{cxJFa}ykaP-
z27$2FeCW_2VIV>_C{MCLa%jkLpe3(v3azY|9S{hnFALtKq$5T~cI@&VAE%(^A(yTR
zns7lhqQWwCIR0MW(7-|pwjyZQB#Ly9Jl<<!Y8nJJj}J+ysiJ08_idDvmGd*UQ`*(k
z)y45b1|ij{n_vW=05l^9Z^H;d11oYTWza$j%+Hsmg5@S6SUhDy&Iid^WgHL)jwLHp
zCnU&i6SX6Jx{Ql9viy}*wnVZG!Ollhjpi#PJ*NH`1?JC!I*AxTr{y&Pu8dajWo*gL
z&Msrtls)|@KAr`?4^09}**%#DK_S~id;0qN?m7*jL22>nkf$aKo!v5gd)35I2K2>r
zI^snsLo<DkJmkA(gN$h3Bgzoa2jPa>4ZNWI#j|Hm?MG-xSXS}Abw!I>OF5c|vm8J|
zVDsx-RZS{7-K?yvfe@&pl8qLD7h_=$*ajN2tmX7hAZrv1K<P*f(2dy^KfXxlHn+{U
zfYoUn<vx_BjAXFjcT}+#?>eryxXl8pRhd|#Fhg@Fm6QqOwsajikHC^`^=3gaz+Qpa
zt5k?Jil?Ne%Jl`}Y>`k^hdm2c&Zruto7r44b!9dWYghN?{d+O2H}0rp%Vq<oo3nXH
z6?o9iM_D^{#L6l~#GB^vN?4;^VT+%+4=2PNN0F!&^dpsGL-Z<&3@?aCLMTI`BQB$<
zo?>8VXbVxtEz8%j+eZ++;<C#tnOnN0<m(&p5wy99$WL`JOwLLl^RcbS%XrqcYSIxh
z;t~MI+{+{w1g`C39|?ej{4?+1;NatUFz1KBl&{xual`wKJ&Dx7t?UOu61_Fl)yE@e
z9OfK}Xa;VyA;U-kC3rr0c|*wmRYNY$yQ~Sj)fepb_D4;hplt%i;RYtA&Lbtrh@nt$
z1bm`Og@w5=!??7x-!clnUoWs39UUbqWJJjxx&r->`)VW{k{^{anna~0$F_C1>I{`@
zVdpDhG2`bHGEV>sW|CiLEe1YwwSO~3Lnq{Ys$2~yu1%R$Uwdb+G6Rb3^<J3tZXJj0
z)0r)PyZQLC$x_U6iITg+zJ#Vg{=49D2i}}hL|N)qR8-3&OE-0?$!qg^>gsD9Uw3Mw
z6dnlf(S`Rl5TLmMg95Pektl5l7Q96Ptlw0q8V;F)txe0rxfi2+2%f?wFY@<Gyh~)1
z?pG3#a5T1x_c~)%MGC<}yG^)KV}MDJNf8h;;dBfMf%;PoIzd})0A<ENxAPGZAY9~f
z5*!A!siVlr3G+lXT_}t|Dv0(MDLK$o&F_}m+W;q#c>VgS$;{JRILj$>RK50d3A9dE
z>;iI_Y#;b(dc;Q^K3uUYB1g8EOC}n3*0(TUvXF?T{KY3x&fYGxS4<2{l$WjM_lKPC
zp0j=4zM^`?7NUVd22T7`x1Iq;o<TH!1Ox^HOJ<2W(p4_6YTV9Py|$)Xi;_zr8tI9+
zf64R*J2L)4_2}8N@kClyB00!=>J^vHsZ*(VE?Qth`axr)Ou3&Z%b|^wta1`;64I#0
zdYSv7utr+%cW6W<wbLeY5e6i}l0w~0n&YSR1e84=piw;r0f{c&tSF5{m|Y?nX^9LP
zn%y@bS_uzD<?QtO8m`-NIu}dvM30{{u3g;$!tv02Q$LySlne`$0s(-N?OwBwlN~7|
z_v^+Pc}Y@9058&A8kz-lsiN-{e4&q^zC{$P?#j96Ku+c%o?ahFJGZw*!))M@j8BCu
zl;J(GfKvEO>#$ei-Mg$r6qU3&JHHdieb$dIiG6Cf`D<$8RK&p%A{_+}0|d{NR3`8D
z(5_rNg#s8I6(+QvCq(w)!gS0+4Ov~Vi%te?7P74!Hgsuq?t`!x>^c#@?}KG2-PMno
z7F1G1Ke*Z8kz=KlimstsJQ-UcT1*UjJ(vSfZafBkt}6N2G^mEuShNzLuJrmwIO^Pb
zB1ns`nSqh)OILt=>iAIWDb&D-t_&d}!$iRXSm3E%7|1LqIu`ifo@>1KwGJHcg`Fh?
zY4QbdR2QHsO498^p+vWZtSll%)Y9;B$i<J1jqNZfFeSkkuJ9nE?1&Fc3~72EA8t1S
zZ8f2i;kk@PlR{Jvr1%%V8b-G_%CYiDhGEedwEI*;;qZtvA#Vs#CSST)KbA=Kgtqw7
zLnE@UId|By`6*O+AU~8sghxC5*v8;X-P8I#AiBFKCrAdd5S=J!w^+62XK(9uDVdEX
z{Rl+l;@kT>2~}7Jwq))eq|N~NcJlDNvd-vqk|ry7x#5-{9tXI??>g-*0L#&A*ZGJ(
z10qR8OZ#z&<1X4BIlrzZm^*|0ED>jdQRszWEqQjZMzh{|^egm{eb0QmXIBpzg!kU1
zD^@S*QBM0OFr*kZQdi&M3Zi2%Xkh)p8a{TFY#cn$AX;50<OwsCji`66(=BV7R-OT;
zq7Hf)86?8Qx{?<gP3+ynW-ruw`uD#AVB3IBptUS}lp^%csRrDi)$tt0&i(i>gJB6<
zf_+ph*btiyT_0d>LOj3_uq&)VvcsH(b4_`eKqHRv1?rK)`!M??otT5oMjkaGkC#y#
zQK%r8v-}G*7lq%v`Cx3!2^vny*u8a6&c2iFBR$<{9zfWLhs7($NVynXypD1g3V#Fx
z4k>L1K}GZP8y*M`VPD+bMj8U7T}NVJar^7e(?9<d3kkt*0b?!<muk>%4~YO5OQ1vq
z119_uTyC^Zm2K`j+g!DCL0{Mx&LS@`z7TD=WH~M8feo}~i8@Tly-sF<nkMn3F?SL=
zrK4GzlD=NnHfmyD(!;!M`*yd1`#MLCIMfcQpwq!Qqyhn2FM)wPJmyrppM3fqKcSCM
zgzblrqq73A8NyHzDrA^q^KPzg1g7G>_rbxblnm=t@}lamoTWw6kGxwMUR-%<sw#^a
zDl>mYQ-L>h5YktGg@J}a4VU^-A2UcT5*k$uH>xrK%Nr2GH#{}hBr;7<D<G;*P<0Y~
zJ8txhw1JPjTs;m`2l>D^RsXdPkuPifFI}RC7^LehXiR>X|A82rXg3@EtnN_67Ns!K
z(=~vjrz9s!osCB5G7llQ@zbdtYvGZOK?@t?enK^rZ@N`1xV~oLu8L6;`i`GphLeHk
zgq>606cAq{P4pBosv_i+-)+3t#6{wO9A#xatHVZ{{d>GWpu#5qO_5<oeSJM-8$^<d
zwT$nh(H==hbx<*#z?EV_+3F^7E{Jnn6S_6>yq%>pxJuErW4Hz1#{EW*e^V=Ru^}qr
zQ7hyk2~|%4`eT~LE=<LNsp&)O9WjHj?K#HQ-d$K1yrinNWF!;blIU@%M9_LLCZ_vV
zm5@Qc4n`+fp7W$!^erCZ9do{TRjOL?1`%Qm-LAGc6$I}2_X%h_4n2miI~;0TFerg?
z-kzT0`jTFy{$>0)1?k<wXpgWyh865%&I%7QR?A|S8-gY96Kn32oOM*j943hLyycCN
z?ajknAOo-IuN80K)@;<(fAIbLIe76PcncNckwFeHM4OTJeQ(4~L(yMQL5)LNs+xZD
zR?Pf>IT7;~eAYoPH}afesND?)#u?2psYGA~Tk_bkW7}1Uj3^BH0|dkQE?F|s7%M`A
zn^5+SKsqoe=nX|#=vyWuTU8>ioajh-8I`xaJHROS2plI;5GR6rCA4uN0ZgDGqQ3+l
zfx@eM{lb?pO)-mVZcI9|Lel`YWd}?1@>qw9wM`_e9&;XZmIs-w<9{G?<}O1Z$nl|K
z9tVLxbWEK>hf_5oQ_xq0dWZ%Y#4fh0BDSdoV+uly#{o-9ST8Yy*&vETL|zo~7o;9d
zA}eB>ix3eK^$L>xqL)LBI1eK9j!a({U<@mihb^m$PQO}4WI!=oBol*9kWz?4IFUo{
zf=))!#9dP2hTTOMXfbi2<-;(Fx^S)eB*=th9_%#6aTy{F4hohaKr1qJLOLhfAE2!z
z!8#;lET_y!&*YE1Cbxa?pn~TEc5Q>f-3q5r3J<m|8b6K>tMWJ^21#UFx07DSCfs66
zl>x?25vh4#N_gG{W<^8feZc0+wrM^NjT;21bt+t*j{qf+2}%|3-qj)R*^F4$5~iKJ
z>*P&<M<Sw@iE@)T>0&nnE*jk=HC0{{qMoRT6DeXMD@4wUB$KF=sg}5>sUu7f%0dqq
z^txjvvWfgXQdc)Hd?LQ&-@kb?keBjdpOK2u=3r=UM$AolZ@8y#+X%IH>klysMsS!`
zN9U2TO;-7xj(TE<G!I^XTkoUzpkM_RG?x(4AZq()MLUf4zj}-YAx<41ZGV4C${KC5
z1cxV(`rbCz14tG|dixnV1B)wNdiv=hQ!o-rn7;Q27!9b-)u{t}qZWnw?W*gpE^@Vd
ze<Z}mTLNYYb^1K*TdogWgxqNHOdp?WEvg?`Hq8=bBpGzAW<bB$t_p1r=|MNaqOqd_
z*tS@75)P7yCI%=gkSQx9c0_Q7J=BXX1Ib>O)RYtw<`ErT?A&@#4+K?F!mx2FHUdYK
zq$iq0=hynj6>yR6j~3BOvUqZM8c2djVR%%ZKYGf$>0z9~SAnl6G&DefsimBRrrE|X
z=*bp?J~4@a&dx*;m3ShL0j<7rYD;1*f+?c1NA@Nhy%YAgHP)NRBm)jeFPb<>>MoEZ
znV2d9P>=f6r+G~z07O{b5_^)c=xs~p&lBmtpQt6+3#4jCVmJUToZZBmVn9H0R1~|x
zxwqR1MRA)YvRgtZBGPS>5%;zcmrd$HL&dYU*zKqorJL~3#op=Z@9BRTN*B{Z_RA=&
zad(VgM;d(k*Fa~K2Rdle@eyly#2bY9L6lY#a?7xJFo4!SQZc}?1Y%SJ3wl3B-u6pO
zOv;ZwU8DISPk!vGI`v(D{jK65R~QnG_Jw6UGx;4aemnO(Ag%Ff$!d;JD6yeFp`)!G
zho&>4r$w{<#fMtml)VNAf+n|q1PMXObiqnW2TaflU#Wjvx?OS;x_2m8X+mp-=mCL)
z(Nd18guu*6yf4I3ZhPD7WcB6EO>@sm^BB+L-l!I6P8Zho*25}h>}XoP{%%kI`W|q3
zR%}I#N%Xz9S;8TaeFqEG-F;lV6>@A2pSz<q65-rnGGli%{QMLcJ%w*OZ`9U4@?X&u
z_o}8#$N@{|_WJ=g0-QDMW8{$5!5eiPIUXd!Aho`Yyyv9ohkz;(G*tw#21DPWsY})c
z;^gvFYa#UT;hs=)DH=m7dj7las+f$`LWNok%hpswm(&UFPxYK96L=olp`>g|*nSwB
zM^Y+F%jEl>6DP<ZMq*<S0`?qPAQeiecPHap@VYgb<_b`}HgS6zK+?abzk?Y*uC2@d
zki<P5gF$IQK1U8U`DTX>r0I1i#$i~Q?pjV+JzQr44DRBzH)INUK4FGl(`ON9keO-K
z5FK_T%6^1VJXEqktGn+`@>obcWy5Hrm1ZANu0$_w3c9FC^JNX>`|J>I^g)G)NXx30
zC}6}H890eVfi%dYHtSr@;o=%7h5;O(4ce`;`ydKjO9rH2XN!x8Wet?lSIg4XH{mQ+
zP*fjt9+9RO>k#;Fvprc8Q2~e+`9HrZo(6e1IO{leE7AN7-G;JDKWY7dcg?bEmp3^7
z>1^mD(ozOhB<Zht3M9qB`T>*={n9t4fJ%!hsq~`S%B@OVG^}t=n3n<`J#sI?OoqPU
zOkHsZ^2QXDME17WgBXcp%mf)IcB^8Ghp=~Wl`F|r628PeAau2I`)jJJI~;t~q46}^
zjexfoK^=N#iGDluO3U3)H)(n}C>@6Sk~WzMM{FCU;XxvQo0!2Ed!@Q5dXuMtg~>Xc
zm5@g45)fk$HQ&%IjP!(YjLz|bjU!U|lM#mj<dV@XV0@Ew2(C7CTRF=&;dNVt6#*+2
zlMrQQ<^n{#522(;xZ4B*pGS%%ePJxg(0wpF%q#c7gMIMF)o6qno1%sg;XoA0jv`jX
z;3n6j^Qa~xfg%d>p^p$!CC$8M5~9!+{f{5~Pws~QVuXC(L<%28r~+boR|BMN>#-Dd
zjdqZ0N8of!ZmPqVjP(K->n;@!T%HV$#6mi@Uv_+togE$(H9$**a3F}U<Z4KNDQr@2
zem|k=hk7l{wv%p1;t)_AM1j8BAsshITx=io`N;f45~`s_MWmgH|CkICFL#HLhKw{j
z4EtiDkN%!h9es3?)+8Y6eB+}zhSo&tRMBMH56uJi$Dtl^u1p#Qx~bldf3%j5Z7zQK
z(^@LysgntDrpQUXY>)g9`wMefdBA7PnR7mavAXGs5W?!VGubl(JMJnj&R0WpgmAtF
zHAv4gHgn1nDH5d!AMjaL-4*=Ud==$EV6mWw2;m$3skV0}%Kp{L{mkCxmu2J`z^iaX
zUj-dAs)DI7EVZ4NC0_|=xeOahh$tdMT+KM<{Gps3833Wl>FD&J5UX%D8oNw7Zhdu*
z&8bPjV8v&aKcoIQ;Og{Z#{2(d`A^n0I!H(N)g_{5cnL)yW0$7^_rrxN8HCvO|1WTP
zh4w!-l!V<E0!o6njyPEo1P~Qdu9GV%7q2c-50P*aWdf*{#*?5A+4g62fvZF#3e*7=
zxCX8$0W&~J`|6msH68aKSRw8;kSqHO@=+WF%xX}U+olRdO<j15`5GO{3dx%^TflRZ
zh%LiIL%ZZPFp2^|!oU%8SZ5+6HjeVd38G#BwP<R;8NH$+;f1miRpZw4%{Gg&dnmY(
z{%IqR(JN?JS?G>$wT-~fHoOPu0=au3IwfUyT?upWGY#`H@k<$t7YWHeOLS`wQIQbh
zcei>nt{1q|F4cSh9>?I=ArwmV^f|}T*@mBO$=(Yz0=OmTNJeY?-RTIy_h0cJ>mHxl
zOJ*;Hj)Bsn3*d<O9EH29DRhLQdzvsl{&gO$I#U-zdxUc8E>s>Ci=2Y`J5kQ<zRlpX
zK$s6N$uyP!Zt#%e<5Po+8+U$6;qDOmYeB02ds^E61GWKoq0~Rn1u1lbr2mZa-0mMp
zcdAcL&Nm67;RJ*}#i`x9XLiYQk&FZ^YeE)&sk=ykgq|5?eI*7Z>8_)YC4PD0?+AlG
zkBs;y!r&77lW8EnuCOD7L7!a+0<dBKqxsbTlYEDQxi#uP#0((H9m4>lSB#G#Qs$!o
zj=t8Pq+a4wYZ3Ql;q8M$WNBPwT!?%M5(3itN_acYD0t5#XX7eH#OcRYlCjL7bP-4x
zwGLJY;^V{Lj+8?l1aU~{BM7gOBK_Z*4x(3RYMnSigwrP-P~#wDfJ3A)W~Tu*^RdFh
zbQMVvP%Pju+RKcu@Y?TV>xWiTG6A^`raTqOvIO~@LFEt>PN`VC+~%{n&D&KG;GC0=
z9_ogNlMQB)o@%D2tp5yh{~>5F60o2g2Y|f$JxLVM4k0{;k{~}7KsP}|5Ch5eL6`$S
zsz>1`Noi@)5seJgEf0K41i*O=lB(u?1J&{cyJ4DXf7-D7MenFcE&Wy7cIBC9u<so=
zR-L+8b7q6Xm8r~K^=x@*t8dt6)%<FedC!csPGCW)!r~D>x8SXu(^tYY&NdWA+BzN9
zQ>))yf3KqFVNLwI?Mj(Budf+MSM+4?|KUTOEl=hckbFBWE#EEGEtL^*d7yjLbe3vb
zxz|pidv@Mqc4npv57z;iaZhjWHT13q<|RCSEC5`Eg^39i0INdqnjLV1=0^l6lw&&-
zL+V;u*nv0&14A!IX$E!mA%B>)a)h_0Xq{kU8tsV-A$VjLLaytWj#7>?8;82OIt$)a
zMe)d&A{Q$wD|-e8*d1*V8OIGH`rV0fZlzX`<WO8|Pkh=hguHQs3{v-59I=k^Cv+5U
z!tkan>S`S#TS0Mg?E1eyRMK~7$bVwO-?HwoT|vRq^c9phM;#lNVwlHJa8KWL$qQLd
zE$v3q!TvipJV{+mc@r#CZO_`@7`>v3Czw+pdO3wPjIuV-p-1bp8-;seiG^fPPEOVj
ziXv?#j$_-aoRt+Q*Hn8{nvo*5gN{`rPRoDiCZ2CqAr@A<iv%_aa4+#uPFTG>YTpBH
zVFn8CGeWZc#~ax3j<X%Z<@vW>ek<9^+5Y(PC&(G<M9EYOvrJ4a^YPoR`0C6@DjS9M
zHf3$t=@7ef#<0T<pUy@g@mC=@E3fAq^jSjC)chKSuP)393qm{KPT$=u<KONz8Yl*>
zk3GF7fNpS9J(`V`^*YpPUPCj23nC;zAQ=)QzFhyurz$TAK-nA+ya>i%pR#h000L9C
z=NG713(U;&_+8aZD_H;VS7v&#Ae%EH#(fHhC#DiA0bue;OG}$z5-oOhpkB7sF$^vn
zywzmanlFxI#-MZ+3s{r>)bjHCPlxFD@F>=OtTYobez4wQ#JHC$+jsdD^S+9X4&(QD
zpT=B4$`z56l<dr&pPzTYw2WuH3HMTtaZ*pX(1{;plVQE*eXAo71E)+d4_O55)hiVY
z`8UPs@hdC6#yF#~iHRMa^TKK-yno{YAQIi?t?_EQrMv4*lvK^E_rmbUAEIVTt^>-`
z-q);{Vl98Sv5B{2V~)PcZf#=|gigT_PR2EBl=^pFh@2c70~RLhP?3RndZR%}k1zDq
zYo7sq2({`R98{-8`H4zyDN>l)*D~TwtGebWAP}Ey%2{xPX7upnZ=JE3C)OPDMc34c
zvGMUzIPxy&l-JbNF+-+DvNs<^o>JIM>N9XfM3X{~T-k5=_=@vpE#@T_toKTN#FD;!
z`vz{RBfn9xEM|OcK{BesZ&)Y^6UhZsk*g>@F_9OU!$z{Lj^E=6>zxrbY2@!0j8#T&
z!<9?k{uYO#qoZp+eUkdt0*YrDJ8~J{V`)3PmGfzV{Q8`|Gt8&gT#}})6Qk(7`s6fE
zXLTK$t0A_Z@rxh%#Xe<7KA)!8iigB*{?v7O$M3x}40*pQ)@|8QaCYx=otp9G=MMNJ
zi0BnE)L+`ak5Bpy9o^kB<I!~#;e)sb-%=UMbBvX+T(fDILcyGb#UphN2%MiyF1;ID
z`KjVw+39z-v7uHDsY55zR-Sv#VL^#pmHlxrag+5WcI<ZJ-SsB9G31|M1$>gV0$RPS
zGI!uxSSbnn|2z%%@+B7E-=;XmezwYN))Z)5A{D!O`Lt<+)Q#|y0~6W-;`^c<m*1Vs
z>|>`Sant?PFTDf&0=L~V4yQ@yYB5n*Ll0j5qUPfA+Gy_J>yeRbj_T@6F70yJ0Tvp$
zTAA6-eZ`CvEytFNJB)UmRFsru9}Tk8q0Q<gt)Q&k=d;5oCEKTZ|3Qj{9=+J~(Z@?K
zZ`w`YFw&O2$$dfzL*pwy2^9Tto+05fMZH+Zy7^=?l|EeJN#ny31rt}vHrexDPM)K>
zD(h4t3~xnWEcYxL6QTsJk9`|qWpm2fT|c(A>~Z#%z(z^dVJ=4uVP9>%*SMnR<v!Zq
zP9<imUYZ$sr1I=p@5PT2D{58qOnjqgm!4Ea^k~0xd7niHgNT6NL%DTIlWCgK>fxVQ
z6%uDZ?0uT<zYP6FiUO>;jSt0_#ciiqzu&H^CM01SIOHhLWRSUiCQi=J^yHX|yVLw!
z=1kNwxtgdUnwMjK@+zzbz9oCpQ#1B!U0ieTzC;*qSd}?+NhnG-WJb@?Q^WVZg~sG4
z^_!D6MBRS#JFcZDzF`&9a>m=2IydSnbym^3dpckE_P!HdSL8-}-QDlNnkChG=8%t{
z&dGgr=DSi$UN|<ql-E^QGp_veZ@<1<{F~Q9F0;>B((|X2_R@<z`u^ls`%O>g*kpAD
z26$&qf2+!<<MO+}qj+Gq5F5RDr;fMvq)|1;U`TFxR@H9-6Xnfu?wV(9RZCn>jwgRR
z;xe49Az*GZ$0%xjym4a<_Y#Xk;i*lE7p?A%*bVaK$Mt$UA6(F!HI?*>TS%O*8Q0g{
z@gmz@)5S?nmn~cI@h`E-YIh`FylJqGRm@f&+v{4FkrB}OE%S$TaIvFU<6s)?^L1yL
zrqUW0tGG=`$8x?W3<dH(XSs7g@(G3a;dPcqiMHq4g4S^KH-;xTW@R57Gm;!F3_hUn
z7@yPjtm3!2c=GEyF=rL#HL-0$G5InJFRwLeWkq*+Y>)19H#@SremC9S{J53NZtP{V
zYuh5w9T?qIb529;_rkmxYd))Fd?o$jr@8G`>27%FYAy23NK)bM!(HS**K;?;*&iHF
zyxwqBKr8kA#e|{XTO`k~b5fE}7tSvzuRPuItR~CC_&Yvnwc}AE4hmRe(O0Ro+CGKD
zP&4;NW_I=kX9}gkeARO2<ftIZ?zj7uDHi%z>q}PxDb>%GSVS&a%|aLJu(Dcs<@q1W
ze9FR`Df2CkRi&u9JQegkoK!2{mN+Ys*z!RppJ!s=two3fYxwB8*!**Bu`zDn?o53(
zqJ4GkxVYQ|wD}VeuXI=}y=&UJL}Wc4&X;aj-f*t?gX(G)>gE`y0lkCPzN3LIlm_w@
z5#G}gqe2Vcg+@CWy(Q&GuW7yp1jtruBI1QIuDJoLdEDVap&jEkS>L#M%4Doggd2)o
zPN}f*3HE>1acMR2F_{=A?g~u4%Wh%9j*1a3!9pILt4~CfojH~~U;3yxsFD558&?&9
zl-dA4E$6(o)v(MV?X<!6U`(s|mCSkih*FyD!rYHhkSr7cdeo%7gEmn;&E#mivh}(>
zCWUdw-A)G=;#<8i*^Wy5IP$31iC~~dikfREDP?74y^aBEz+9gr?0EddW@gWOno)He
z3&V$IlZ~!M^@MHR8qDrcaOL`*wj-&9!YYPe&+`p~_9QHr_a(H33dMdt8dZVD^SQa1
z3Us>ik3E%#!zV^L7i{FkQ)M`3-E_qkx|g`Sgy_qDp9tDp=9RfHR~43VW!D*8T6T80
z-@9v-2jt!ja`d2uUM@jhW|XQPBQJ*n^~4%JlL{S6uHi|qdsQ(df_Fp0N2Sj^(8@m`
zcfW~0nVZUyeYncw`|e!28K9cLJk0_*t0F~s5JBHQeXu_JR@1MV;|q6i=z1L5cT4U^
z(H52HcQ!tlOcg-l$hdNJ+XgUtjrg)0S29pL!98`KTf4Ez=P54j_OafB2a-}+mi1sD
z!yek0=7QMg&!0^oy*P@xRXGat;m#-OQ}G?AP8FG{9%C8aqB(pvDXm3t>)gmy1@`v&
zAr~@omWj!MHr;Km&*Us{cC;()M10duzQSrI<t@Xm88_SBsc1g^#v&<$m1SZ=w?h*Z
zCDm(S{<)8}AlMa6^UpuN^VUL!W<|E}(m<C$NrU0DuYUP0s70gm8DRA;%xk<x-WU-K
zQSzQ|^5&Rl9<QZKax9Z+Q^VC1;_|)c%Dn?KJihW_#_jv5Foo$bSIkQ^LEdm8VyW=g
zK$pS@^&=0yV=uH%7Ec~jvgA$Yn(J8ZP$|{@+B&1?w!<-sc9m<ndvDR2@Q7xqYAm?*
z+!tPjYnt9Obq*l3sc36!dwdUzd7sbkv&+1YT9lG?nolfCzwM&D*PLq$WWXHIl2ST6
zS2?T%+<TxoR|wNSv>LlHXo^R$VBkqlPme$NOFfT5z!IC!Jo7eG@5vceLgpkNR38t4
zQnKT9Rq5u@V2@gNmfd_2bQhrPmudFlZVeR-)xCm+BQKxGyI5H4gq#91Xv7`^tp$65
z(6J%}<#7C9t}y%QBUlpyZACJepU8%yN_k?I7wh{&Rmn7(e`hDNT;gSm3VR7^>mA!o
zN=hDJ_68x3D@6iH1u}Lw^e$bWfgppE(rb`2*+Gc;eGMZcuV))5QP3{u0l8s4s)hcb
zik3ocNW*O&^XLiT)BKcR(14R%V(gkQmf?!L6V**ko6-Jt9ZU!j%u)+?g$z<+zLQV>
zEtnI4J*61-$O$oeTYdW%OGhs<dT;~4yCmw>_HP5&gfv^$8~3%N;@<|kd~M<X_~fh_
zSg1Py+akZhq9=c>qG)NJEASsd>Bg<JL<O^5K^)nW00Fhm@_<s44FRS8{#&+e@o#Hu
zs|f*B>dbqQ>6>!}^DW(5v6-gl0$G@7To1@i?8rYcqPZ}pX`0Src?G<prc#L9$|CS{
z0MRANATzM9t&XGre0snE1-T72n{!Ib?(J3XToo4nfb)xl;=(J1xmTCFfuGLQc^O(;
zTZ6XhI_7F@%zg6IDc-^yM4T9Fc}?5vae8`s&rB&TSuH{^2&V4zm}yZ_k!kSTkT30n
z9UUJ%U3d=5C0;LGu@Iu=_cqwY{5!<!lri!QA$l1ib$uYGnHYH(wn-yq$ZVe!>!{z$
zdv--E84F|Mdb0|R%5x#Lc}+%ca>WuO*^$m|X)6^HnYa>lIC)aG!<+z0VDVhQVgv)f
zEme{KeiO1In>A+LAgmuyTwE+^-=Ub5lcS6WIoDt`-=NwA61b2Tnu*Hc0c5^r*BOsR
zb@8j474mhPy+B_rq1;>p%ZgWi>;YqGe1svZr>AEoFfoy)=%3fOf0jB}vrrf25)kFx
zIVU-#vYdA=O4%dos9RvtZR}(6TD%_xxI+!0^@LPTxqbNXArrMIy<!iyHEY+FqFap_
ze6Qf~{-$yu<)$F?FdfOCErf@$07>-FyjbM^ea2?BJ4D;U9iMifL=ixeGYF6kmt*cQ
z6N-Q#sD}UMHXqbet;uUhwNLq>X!90Qb#};_M&X-oJKLC>n>(F9A1t-EII#~bFUe!^
zbiXB^D=Ro>yNTox#o+IiJv<m-f-3mv8=jEYQ0z<{)BbX*`HBs5+eO;~Vb7AmAzLBi
zec4|n6f^K}$A(7WrA;S?T8Sw!*WdWO{@#GAOTxXSNIOP9nue5`hR8`|FJ&fihe4Gy
zA{>o0PYbEU?GkNKIqegN*rso8-@s$@h3B$G(jz=3j6Vo2Zb$t91o7bh^a8<G*D1SS
zx$1f==i5ZtY!0r=zPUN}!8w>a-E@7HVfXe~<x}s`L{<uRU0i(pc4|e^ij32>&jVe?
zlkE?7rZqQO@-TjSGpM{9P>Ew9dhFuL?SR?O&15WG@Y{Gq1X2etScX}3n|CGr7Sksy
zEKV$`*X*DqoU%#?jM1E=G|gG?`-|fr#U!Z))X&c{>m_JKCr=*kzF2|BO8^C~_Sv>c
z<q)XMeU(I$+nzN4>!YkY)>gVUJzI69_P*MWZWS@{@U~3>W`FF9|9XDi7zl)%#7oT8
z3e6vdEh=XOO_B_id?FZd*ut7my*aW+R*XV5Fs)POOcBa;iH#6AwFxdSK*(z}U=$pU
zGT?WtT+?q<o(QIXb9Pb5%2oO8-th~k#w}lm?TW3)IBiDT@<Osq#;1>j4uE~TN@)Eh
zT>#cX-U*e+g30)c_k_9kZRZo>0za}{!vtcl|M{6EKf|M<v+r&{>v<TU9D?ChEFjXn
z{o~>1i0PAsPj|&?`AfPAcTJ7fHm@qWe`nfs3A1n;osO<8!x*dyZYM-a9(;AbjSa=`
zo4U>#KEKnj@0u7l)nHy!d`+Jk4g;I|7jO~GL5C;&To$s}A)W)KCnAp3?HT61>qcjO
zeEIfTAFSesdf_Am_*c;2?8&^o#hb@qb{6pDN+W}oNuKyNuOuqe@xCusiS@>4dhs3B
zW#hcLi!R0i3U*5F?(PGqLKwmX=smMP-FUVk>;3FCv;V#-PjXe7Lo<rQ?IjAt>!8s#
zh(r|*C_bWckPG&8AYUruok`{og+8LGzQ0(mFDvk<uHeL{h7=bnGUg7$gRLHQ@t$-U
zjUM*kB~A)$^i_X))OX9ZU&YeC2r7CHS{ns`m9vd-V5i`NSs-l7{#2TSTPV<F+SqDK
zeN4;Jo@Uzu0aC(1FzY}?z(w8_A~iSnf@^M{2j+ji&zJStWm@<yWOw!xJIT)sM|b6|
zuI6b&|4KQ!C^kZ<K!44eHLo*heHmrYZHjU}Vfv->n!$L?uZ)bWzhz$7G93FkYx$#4
z@}d*&V-2boxRCexw~p8qpIFcwNMY1aJ|G!)l6{zaHBPU(mMPY{ro(t%Av0B8l4UL3
z&aep|olr{jP3^%ar!;>a@AE`_CUCDz`B7DRCuPh9p7M~B`+0nYN!dMo;x<0f*f`^D
zqU&kDhn4<TY2e;Lx@3MjX`ZKM4GYW#eZFBe>tp%N3d1cJ55<46<r}^bdR5>-DoylD
z=RSrc;bz|hXY)`VlQJ!1!^!^k3lxcN9N;<a)|y}2(tUI3<^5t&a!5MQSv(4FJ+wFY
z=JE4P_>%w;kh+@n#J2)MKY#a&M#B@^#<a4wF1CbA<dkAZ_zGA=w|Y%H{l~w-mv=?G
zlo^^EialbEJ^duVplIEy`mE7YC(r==y6Mj5wU$U^0&mi^43{tMdm%*Y@OE)M$?iX)
z$AK0nkstA`>^`Eupw2dSmA<dkoz@Yv=~WHqeCGCi<pX1p0>${}Co53Vn>j!?ZgNRs
zs+~)Ps%hH#_LN6~6y<Ct^RK@tpE}iB`Ch4;&g$(h+8=gy(i@N?yF9+fAbgvY0-49J
zvi^|XsTd^Lz%{`ft(mupHIlOSEbG&BGlO4?9VegBv`lenNld%EN60CeVjrwwew=!O
zw)sj=o9l>@_s&k1q=U@jm*pF{=5{shXAoV^T4A5>EKW0^vod_3MKcaRCm^=d1s`i{
zpip?0SnTx0EQ99|{irZDSW6M!@8h>ePnUrb_=o@ciZl4j`c9ONDBFmz(qdg!r#>=T
zzgewZtBR<f>^3{Y+jR3o(#)({Ia6b}R!Ix_<2NUny!}PL#M1d(&kW;p+lCtMMpIth
z?}t$5-Cd@U{^ce`{aae)vNu99d7&!#_BVE=KBfC!g?|W=sXkND_dCWAt@sze_(-2F
z-5dBztm^T~EhmR0PhD&HROaflOX5ja0V%Oey}oY8z`yqJ=5?XB>cn}npO)#QS2i{<
z#L{czb~Se#W;nO}?zXQ^AKpB9>X*(**=UTD{i+!DC`v@^okQam#;?ubRO^E-0A4#2
zBl<E_xx69ENd<U7^duKcuWCj!N0%Kt%KXc5k9*20JnLgsk5+EEd)9odfn2QYfddCf
zOzet$!xSV|bS5=^M7345i4e6j5ekzSxCMk97M;WDI~_}NlWWCR(czi`ADlSfSlz%-
z0d<gc5{)AccnxtC6R6c?f@+ZV?&0z9ysl<3Y1S(g_0V%=)rXxLBP(Wn{9n9y5xLi{
zt!ET5j;X!<$fJG`MLs<P@7)yHtpgOyt!^KBylcbOb)12cUWH@lFI{XPXAxxI#+PVa
zrgOgWQ25tLKiNF^RT3~ep!uK-xz$Ec^jwHd<TcR!=K^fZpG8-ml&j^vjxhm}=Lf!!
zlB-VK*ag?XH=u^a(q-r51n0%c3Jb4){P?kH`W<AvJ+q*E+sT3z$^0|AJXrDu!9p3h
zA2)Q=FeLMPx7foAu2IXDm*@Nvv{OBfZxc_-kqdIN%3pk8BGwWrdkx}ufb#z6-Z)5t
zXuV96_u`-le(sqii8;tt#|wVz3$%0+Tjdf<$ExM%==ghEp%gf{oLC$OSk~7_DNHd{
zEEAPpeEOgVDg_}39#wGYd&o`S_nPTVdzY{-a&dL(Satpuapj(V<ySMg*XHvEBoHa*
zPf*0GHb9!=U!oMw4+>-+7Ri}>&Z=!58cO4%r?7@goUbn&8p@P8H9i>;%rnNil#)`#
zVPVyl?`)cTC1@u{#;FK(If|kwb8Ir^6T}0WNZLks_ij_;?em;T4%E%!h0|t@8|knk
zlimIvn5e&V^WfU~d8Svd;&bQck1u0zKIoJ1ZR0AYQj^zn(@g5>@wufBmtNM2*m1Jn
ze(u@9wTcEcSA6recCL7TWA!F+ic|+SdV}c^h8qkR%g`WN^U%+-j^B#Xhk~)<_-|QB
zr)eIJSw8tdA+-#y?0@~jH@*~f80GhO!<`RORM9rg{b}lan>17HA*W6XB{VSW2WRi0
zj|6lPX1iKO&VJ*}!}Dz_Ay(b$P_3l%+4kA3u7`Z?+W9syX;}M1dydEq-`+G>XvvOC
zSUj3svKWy|IwVmXojaocYK%SV+~<#UlrQ1;eK!)@q-fYHrN@NhfBrJI1oN1gz*aip
z<mB`kY+12RaijS@T~pxnnXaQ<&k&EM=>JMlWEo>mTj3a6&his2$t@m#94X^s%YQ0|
zA50u3P!U;+5;8J`&~dRr2(5PSqohPZB^h7Cw~QirC5WxO@?vncqy4I9404r!orIR@
zKScnh_tZ?Ce0J+r_~|e*p)4`6+xXy$YU-MG>&h_DTsZ`2BAc(TuW9-xu+j$*!64zy
zpXyOjAoSRz^LHqeuQiVL6?a~WtVG4Z7FfeUY!(tAfbYu%Jn|Z_AdC0TD0^KmHBnN+
zzbM(+4fm8IMBMW&a&DwxsgDP9Y`Nv%!ynex{gKuXE?~p-X?65!NR2*V^xO&VWo=J?
ze*i8#C@(!Vl`r3`#@Ydt9|ve~2YRLD=DHv;?Q~g9nKoNhF8O!K7`x1QeiHK_=`l^u
z!*pWhVc}wssWL$k(EG!9-?9dJik1!inB?CjP5k=LgomC{gY^ns9U#vlk4{5QCMHND
zg<!&KIkK28wEG3BaJp>+D>~W%l8ZVp1k6Ac0|73IbfbtNJmZDphb-pFl2qkb0o>J|
zZ;MXre-%vpH#-`x@aI`u74r*FZ{tZ&=PMc<s?6U?@tLWMa;DQ!adA=7)qPzP7f+?p
zY8{l69^5c6pisukIC_?4+O}RBk?rcKsdb}JDtHz5`X9JjTggBX7NN`Emz7O+yRiwq
zlkdl8A`j!w4&>dw{a**}|Mj6>V5EY=!{p?#9K9@`^%U-KLH4zewAM$`Q7REV{O<u$
zDOMV|es}lRS(fp+t*s|?m)P$IkV<*(j{L;Q`5?R8N$H_sAr2L#4XXfQ3cm!PV^xvw
z%=_|(f1HXM0vn*Wit18MI|pSIvsYGA>TfL~sSP_|H?)B@SXi>Sr^m-TamH=}_*;ML
zf819RN4`DqmsLxl7jH21AL76N(UScSQ1&kY=+oDo7BjKJxpTcXE-tXOl<;aS?2YBE
ztxOnErxJ2Ic$UOFOjq`wccw#Oj>967{_nBm?+1PScG)jt$+2E2h%f#8nJn_S8vgu;
zgQ=3*yN?7$1iV~~XmRPPEB+=TCbYTQ(Yfh()72~JxV@b^eRdt)d5hLR7=$FP--_aM
z*x~;jULLjYpO*DZvQ$4kCHD4TF6)_cEURRYbHB#rZgRtw3_+@Z=jtTMvfl89wT(g~
z``L{^<v1>gjpyXavisJSyP}s9Af1wUWT(=G0H*&+nK!N9&r=6(PGu}!*Ne;~O8k!j
z|NGBT{_1-2PA~wSuC6@~u4>Fo5vBp^cMkj~o@~3#f8rOTnc9dwm^|vk|M8lA6h@R1
zo9oeks{J}d2+GjUM2u1<s9GX_#$hPg1g0#ZaPti*hWyKT3r|P+FN`#=!6Z5z%1|R_
z0dFi2XtM58CD@!ZtID^vpe-*GlX4^{TO6UWd;dCGC}c>|8kWwtDc}hICZ4S%B_x4I
zO~^d#`*n0nKeH0omh;M;hV3g(Q%kZ$dk(v3WB4n_1X<2R(H~pXqv2=G4EC%4X}~!g
zRown8)|Kmv!t~cnLPx!gKqouqP8t{(AQKetEC(T%6*c|^*`*t`t><ii2#B*w+y#rD
zXWCKLyn^Z=EgHhqVDfJaxEl}=<#t@GJ8bkHpU=`$ZZNsPgV@>zV>V3r_pk3vQR_vh
zc>iy<n|vi;Z+|d`e`Esrw!ooiYbnKIMLU0j<-SId`zK0>-RG=baIa3s+o?L*GD+Te
z3g0GW#$^+}j!H(~DfhRsWW}77%KQ}+O44RfCkYoM|K;zi?`z@#Yoe`;SFmJ`!(AUk
zjFw$cRobWXeWijtDZ@`)_1?n4!Qp@&qcT*~i+1b2`6O94j3GFSh3cB_LH9&g2dlDI
zqM|ubce;7SPe)#N1I7R9W_~^4IRXT|T9|KNkn_L^)|HS@I(-_g-?D{>?E`n8c|ya`
z0Iw-DJ#!isx{hAv0%}kq?}75`8ZHGdk10}Iq-7h;X8-)&$!4^|j@Z$@*&WYInK6^)
z56S%cUREJ_>5!uC4VxD~`qBIlf@c-|ol4M<_SvHLfP{pE#Rx!Q`mJVJmcpZ)=-x>+
zY3gaWKmqe_)2agNWBK0XbOfcgw9Ae6;|s5&i(|L^L}o}SUq)f5vy^z1tE70Kz}L~a
zN3~0O0*SPP0~%DOnt)ME1+`BFJwrRoQ^A#zzNkV;&&<wJXWJLNAcP=+4yv3qwTh+G
zU$tNS2(!W$CAHYRK;Ycsztd(ljqtj&m3PNQkuQd_gFkpC3R@3*QrnT@sM?%vJrsCV
z`_;#DX7^)sd?^O|a+ccZllNi~|G?rbqB7_w3M5sO{wI-J73Mk_O{X>sjJXVc-iT4*
z#P^%)AF~nV)L3i!?v9(+(Y5Edrg1l4>TUuzI1Xz_#)FYq&EuPB%r(@Yc(-mv?YWqQ
z1$G`(Hj=yv2ndWq|3r#<UDY9>eYkz@L;GvebW7e!kZy$#slb))qNZ5i8~sDJ?6&BS
zxksBmURw{JKx8BTiN5pSp2AP{uDs_3Vt623&8}<Ud`xzxC(Z#3Gk^b{!NH*RnaAxF
z&-VK~HBa*m>t$FQ<^+VW6dg-ETe}dfg&?rwLt!GO)Lj0@oj^W@Na-?Z$va5?mm)f3
zMgRvAH~Yy{2Dh_mzQ541CcL?{bhQa~8i_p9)aa77m~t$Vrj+1JzLUZ+p~w&vu`!;a
z6PURU1&m)5W@<vZ(5)1p>Di}UQ(u3}S&FjpG|p!rX&j2_-j4~k1Ae?r4QjL{vjrt=
z1fW8ViK1Phx186SH75{*cPq?@_Uhl_o&U^v{;#pO{x^E>|8oKV=L!6O`~>E_FPU7~
VBa;}y0s1LL^`M4Q`hMd}{}-3hDRKY+

diff --git a/examples_skyline/accuracy_vs_fnr-skyline-order.png b/examples_skyline/accuracy_vs_fnr-skyline-order.png
deleted file mode 100644
index 27de7a95802e9783e412cc0940848099eff9e369..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 43584
zcmeFZ2UL{Vwl(?ziUAQ75k*3)AP5SgB*_L)KypSxE0QJW3@W0Mv`9*poRK6sm_TwW
zKqQN#BIgWmZMyrMbMHOpf9JjN|8KlI#_ciMvdXG2>~HV2=A3J;{X$OWHu*uCgBXU9
zOWe7ofMLYV7`8WNKPh}--)hqaFMF(SNGR=x|6KMzd=BpqSlm&!#xSyD=zpTtUbjv7
zkk3X;%|_8&-^TvFl^$k#-^SvRxy>V^2fx|rSy>yIn_XqU!p?c|H$xj63w{oczyCYC
zxs?G&+&upe4Eqg}xOH90A!4%EQ8U=KZ)c7-oaOOv#_Z`u?{FSui65T7?cg&WXJLP+
zuGrI<cURiE$6l8#Ay7|oIXss0^wNHrqO7;)6ZEeP85T%tgm}d0rWE}~-&I~AQYhqj
z98c!?dEQBwUifh5isops@Tlhc;FN9V*M^h2d*Jd1UKW-V!z&iUenyl8UQZl}A%YjO
z{V*Ju$KE}L@KQ-k%mpvM6A{tC%X<%u9A4B!J<wQP{O$i2KeO=s>O;-O$at%~y!=PG
zt8;z^vyhP1$B!RFba?e@kN4VkYT@D|B52pv*Hh!;<5TO>jtveDY8047rqyB19&po@
zd!rnF9%zWTrnt7YMreuiPOB@5Z*G<w9v!Xw{P}Zx0w$UPw=6e$QddbsgFv{FDQRc-
z;q19{Y)e0W{`|?BnP`?wSm~7$xs9m?VB|5ai87I?hw1pgH8sih7TdP{{OUv3RJOCd
zLHLoMHdN%kv(b->C)sz{5T7h>+@2b&k!SeMo8IXTZs80GE+{ydiI<o6_O};4K8gwo
zIJ3e*(zVsqF>eO9=E<&Hx(UPn^0GeX8KI>05O%H9H*ZdIn{?cz<F{!J<2G4rxy_KG
znx!qXP&QX}x_HQ6fYl}XUV)i%IFD(>;rnW(4%xA>v5XR@PMsq3mK4A*5<4>PO$4}a
z>U!1jJ1?gXSv(&bGqfFRzOK-9e}7w&tg4by(B{?_x80;#*OZ+^1?@9V(;r;<Ms4CN
za1mm4TKUGo#kLdYMMOmQ?%iv&zBoRdxsy@Qr%+(ln`b#Bo^RIsV_fKxM&6y4mKN%=
z#8T5ak(~vR;=yO9e~peNow@mJ?Az;e{%aNM1h%D_ejOqrqL=>uDo##LUeVr+m+pPI
z`}+0kgZuZ(X=+B4uZ^6pwuXln#$y@=_kqtdQlz4$Hks}zI!H#Q=yT#yP(;LJUV3Y*
zLUn)LiAx$R!ou3s{&dz(47vIZ3WkQsW<|pxQXP9Ssw<eB#LjB9$X#vis0$Y^a2gpI
zRaIA4(+WB?Zw@i+T%P)oe-#EPF)t7AveIjRo{g>UI5l;6ax%-pSX@BZ=BA7FNd2i?
zxItZCPWSb;=({&=Jl5^nSefC)-IbSr%~|z^h=labt5-*IO*+{Xm6RUB6*j!2VyiEA
z-MDLHlp-N3%PSqm#o7JA^lH_~T-yniwsiHYxpq?;@X#MvSz9;o_Pgj#+Lajfl@`_1
z*W0ACoj+gYvN=~B>XJ3pUBGM7nQ41BfL`cF={g~;>ZI=a@|4L){ad)45qyfGeqG3=
z(5*?vw=?A%jhu8RPUt_-)s=>uz2>x-Jl>k%vo@i#W9-YR|M?kt70hwLCA?ylSwZ;Y
z$B%tNIuoT&FMOreIjQS=&2I9|koEWX_n#a%c<@Tq$%Rj39B=4_oZc|FZ?>G&z1x|o
z)sU*3Mq5|7&?wNbINs*R+1MRdYT5z|G1q=ZTTxE#chW;Ae~)roH-bwmDk?gE`Eql*
zW>ITUP|%xvlg_ZpO3`Y5jy(A8(9n=>N2Znu8mT;^w!7Yk>FT?34d|k!Y;5vsYHB!P
zg6XeJ_a+*(C5|jKiSXK$^a?J2*G1RwvOb;!gV5))AZ@+1)EN$|>tso<*)dAWhC<7s
zh`#DCFDRwm+}w(y&taMemPJXD^Eh>@lC?_gtnBSOzh~>VrzkS(J$yK3?!FNo<wUsf
zwW_MNp+TBXz)nd`?ah@dS8Qenu9q$UFfsm-mqNp9mVVm2bnay*uTd+bn7DXtTbrV`
zw)Xl`mU{+T`0z^-(nHO!&fG+6K%zBWJ=dn+b;W3Q;3J$*RZi|zvCa5db8~Y)Isui1
zk%mCVslKv%>gwt=JdeIJbqcv{IwT|}O5$W9`4g1WRE#!P=WvW1931fAWpDe_zRt?Z
zg6DnjZGd8c;QaF+`6fv)jN}Z$ngVu{7J{SG)0x!T4%hqo`ixuSYs9ja=I0gk^%G^H
z1XD&D!qW_zqWnjFIoDyLC&ME!hB<)35aW21^R%VQVOTM3%+W5f2Pn^L%4fhb&*HbA
zPCqXl2Fs~WexlTIfy-u0j+~BPX?l8^M$qBC)5>%PTxD=c#9p!!W1SV-`J@!|7vXY+
z=Dtw!TaWmbZ_Z04H8DYqhAH1t&ETGX|MRo)k#O^nf`S5Vr*WpDlAZaq$zJ<@Wh<+k
z$4{ONPfcZ9P)r;f3J|g5*VL^Ix&Xs&^vA;^=JJpXziQzB;K76AM~=jJcwn$trC?NA
zJW2M^dY_M&sq`d)ITV<bWRTE7eeBpzt}>@3<E5!?RZmY(2qn#E!g?Q5QhK6vf_Ahi
znnTxjq%ksWb=ASq*;&V`m-77G=2$Two8?J$6t{$&m(di6&w0N+RnVU*70Uk7&#xVp
zGV9VdT-4BNS9SGmNF4s-Dj6CnmP1vzcvWpjF~~;qOH1v)Y66)g$f()zaMI%^P9#E9
zFod<~-$^gxrU{Y9X}QZ_cxGniWW_{gNuJA^Riv=128v8JZPL6crrjS@l}#Yo2(OQw
ztL}dRH*5MMPXbm(Yn9(=I@@T0?TvwtPyCzYSvLZ1+`037etw=-chr34$vz5xX<pMG
zN^l1jcMS{-On!ZP?Qd&8^+TDSo}S-j)uO15#G^L3$D9)**M<|&I}8wI=u~KLtuF<7
zdwav+KJ=A`1C~SD^3kR|b}SZBC<I%b<FSxSQXfAiu3mKCUWoS5o*5Y#SzUwFwBBIu
z7I)2hWXxY+`c-IX=yd6V%qd;k3Dzu92|{;477e#?yx(bF<L%A0FcJ4{y?%4Z5ok=>
z(^O61Lm6jd0b4>m9<LM}5@OsED@G&al;=+`l$t0Vu5DMOO4wZ4?y+tfFB*-qb`s{{
zQGrLH46#<fF(L&X9vz>BC=OrhXx$LbD?2wgrvb?hB~J+jSZ?b}6G>t|CyXFftoP2i
z*|*z1d-~My=a&~wJ6kJtuG^bC$L}bV*iGeGkIGc%af<IdOe+UJ)1BO0>HA=MFG)9^
z?BGG0#WDG$CHL8n`}|iN%G1-+eM0tO!5`@jv0$t9MI#7q*ADM{Bjo(SZt4e<L@|sU
zx9hqMJVWWKlW663b#)Qik^+8wFikUvauiRJjSlHG>x31_B*Dqax&G6KBPcL1j*?ks
ztX);7?AJOO8JR6i!hGX)$*8C(jWVZv2#*sE+XWR{8Tw688A}tLX?b}{j*gB$at#88
z)=Fl|?@<UXzA7y(&0M{VgY`Sc;lAcmzWS5Q@6^>QPGR#t!L7wMSrmdqMLig=(W%Jw
z!b;K0?Xvm#<p3NfMLk!aIzxn=U4|!II!@eAL0tTq?&fsKOh@$2romKCks^FpOweeY
zYTy19D66EPaJHm3b8*pTcw{6*z}>~AR3%e01zlHi@|~=T?c&n1vi1u19jj3Wc$O;Z
zYVV}Oc~aiLmunK)DQ%-tKQ9}V9v{!BXJ8<Ob6X#0zNA&WnA%T`MIacZ0JtI%f`ul`
zNyUkdjC}NLv@0NPvs}0!0ju;?baXVDWDun9!mQRPcPUz)>P`aWws+q?laG)0;^IF(
z*=I8OJ^SU$mkL?hW$ey(bKp09P7{~DR#&GvE{woLEEo4ZIp|O>8!a3LkB;7Xl5=fi
z!=%!K$X||00v-lFQ>!E*bJOzLP<24(XtaCgJ+mS)046rRS?amsfKJddXV{+ImynRq
zzB&h2*{JN`P>8UVeZQ+JPxw@<BgDgW3gK1J&)yXFnpApdEq<nVadF8k_6LXncde9F
zI{#85-#7u@qAWwDZFl+X*&C1=+CEc=q=#I(XS_B)jP4Ltk<#=goYrYH(w^)7r#<<m
z-6}BU?Th8>(N*}J<h*h7*`e2PpO7NbGVT><FKrNeEEK)9?zblvC9wkBqGjE{y}nYh
zlL7gqsF*gC(ERQRf|R{Z6B(K8dojWu$dHDww%0o=Sb2CpeDogmBBO?M=x1`4#A9#7
zt^F|{4^U8qa*9iZT(qPK@b$IfuOEtlOd)&Vz=4pC`?+0<Iy*}`)PCMxUPh3yI}T=^
zx3sjZt*^)T6j_@rzA?p;C-|Ns((&Fy>oC0h_n{mAEjZEtT;zXGF8jxGBcw|No*%E+
z*{H||^eA@gR$(zMW|5+^MNnPe=--(pTw!HmX1)ywgZ07%OWO;TRA)rv8cW6#!x4;D
z4V9k)6d40(sJ6df2k260q&tLA0-$~fIb0S4q7ZLY-Ctvx`b&iN;?XGQOyg6EiFaGH
z?0c2PsFW`X3Ti^QwZ7~9>=~<whz^9-_Hwfr)*~_US%3gail0he$eM<vOeo)&ZZpX$
zv+L?)xMp<=QbVg0M_IgXwSS8JOdpMa-6PwegE7zW?QW#_+j|W0r@9NwRhy!OfLAAF
z>sHsA6-eP)+uEpTX(!)sz)9N@rGY600lQ26`SXG6QoAalzoLSenD{syT?QnX>P-kD
z>oYss=7=f6?b8Um6e57|^ywZ%g0r);Yinz*>UV}ibs%o3Y;SKPs1~Z@c0oep{qLTD
zQ>T;`d{15FdhlhxL@RKMr7x7SEL>bl5G6YRso6qoOj^18j|6w70wo4M%h(?U=I&ih
z2$sC`^~Jy8)WsKC>Po-8BBtZMmA^dMWv$1+<DoJ5$?XHs6RG8#imfSg|8W%bfn+E{
zAh+!{Ex6`NmhZe`MourN23XX%w>V$hahNIt@(7CmCoZb}rlX@HcKfzY&Idqn#$TQv
zJ$UGl5&-h0?Ts0MsPHR#FU14sGtl6HwgAW!->Kt%NdjG;m6a72z&cPl2u<$Wxj8vG
zzYs8!{WKgx$H0(<umSKNbnO+}OIc~_5kz%|<n~1fId?pL_AHFI&*3vbv9JK)^=}SN
zPGOLgbMJp9r96K8*Bjo87wd1`z8ww#5Yai{rqaLv9;04lm4KKt@KA2Jmk~(RTvaEp
z{$7c-y}P+D*SufYdUMsXE#E}74XD+^cv3V__On+lf8M$ZoC&z#?ErdiVEcgHt|cZX
zf2)rc>EE+w&zp0%{ojD@a2f_;0q{1z@i+(-64KJ#Kt2)m9{lVb9(Ow+ERb0?3|Pcl
zKvF>b>YtIH_66MPNJvTY?KR^Csz~K_BHC>|<p3F3$*<9+Z|4Hu)C4jM_yJoB3lP~}
ztrfAs*zMg!N^siCFu136ebsXwz8Hs02yB~6OiXMfAyoSctQ(-`*7Lt4U~-<m3UChb
ze)N6;Fxn|y-?g^r?T5#X9m`Dq5N=jnvpH>Fan+zPf)l1XFMtHDiOwwhi`8vyZ7!hA
zB))&&TngF!yr5tYj5dE?XHv8qL9JtZYr~HbroN3>-SYA>51VQx*VnIKN8sK#;rVX0
zyWu^UCOyR?XW^VFfUMvtB?HLga#}P5o~pvPb1pzI7LWtCQR^)P8JgeS<OYgXR8nFB
zj1wt6Kuw0<>_CoJ2JY}{W+n>-6O{JN%9d0dot%v8-*AF>R>EjBTpLWlrv#ptiC~eH
zmGxr~(bg)qX_+6c8y*^p6PPZ%nUa!%iw9Xj#(8<N<@wQbz!M}TU+UMtc?TR6co4$p
ziW(YWaisS4_GPo5$mW-qRRYgTXA%}*(uP)4RJ6T;KLC2zD%b6mTWDO$WgFKUE!~(d
zL1d^MW(-^q0glq<PbUx*8JR)FrjiV}(Fw$lc35=i9P1NV1xjrY;_@69^x@y6fmo>A
zz3V$&vE$b8_G&e!EWkZ9J^*URLEef3C8ZpM5?1wPKi<CNcQ?IjKYzXn=+0)cm(=(m
z5smF9fQnSdk28YyR;>pAc`bN9eah$7*x1-{o53|6m{)Z)oCqe!O@BvF52sd(AWBn9
zkB*KC5DG2I?%%&(TUQszDE;R0$D=VoE3*V<N)?5Lg+ZB!R+5v8(JphkW?Z<n+`ZH$
zD{?KQEmfI~+pI?ow^+-T^{um072;W#pyPb7yWCDSyk(Y+vY`=hoR8n!+`Py^=3x>!
zVW8St*H{A=gm7N93#_^;RV*AFvWTp}W6exWM6}P2G{V}AL$Jau<<Y+D78dUhL$G10
zGiyuq8!}*EU;q`*|LASMv!Zb_hTR`BfG|Gvwe%hS^-GCwXYINGaCC%u=Z323+-0JL
zGtly~q63CqGG=_@+O=y34<5uZqWC4UwHRnO`TZTxHMr&LIDC$t6q<h}_A_>P3!oFX
zFIT)Df#f0s&kG)WCqy8VqQ-CE?j#9a0qWDS((j%H@9OG^2Q)^Vo&;;30Fnt9vdO3q
zR#3A0&i26r2U;jZHZ5vwwPb*s*-U=tOft`JobgiViuz?ODJ2yg9-anVOUq73HRn-f
z`~K!+N+5d7FmKD2I_|NuvbF+`4|@IDQY@?NPkQY-U&9Pp^L4M8s*_VO#Ij4BoUl|}
zH>L^@7kKI8lmGiiVm*C*V-US|H5Z^x;<CbP@h}+^VU9?#c5lpAl2D9gSJSsaAYqY!
z$By(8tFII6mo5dvfq|3ya^`eTv0l7*(Z28Xx!cAdlW7!Myn{)o?J@_kMJxrlYe@f9
zj~^c;Il=`UbF^KTl#v94LqyCKLIo}!NhC1YoE92*5etOnA_+=vu+p-rriw}k+&zc7
z6cF_Fm6<GqXjc`x5-k7>J88h+u4xgQgRTg&@@u%CCD2)FySwkXZY<lTC&eCdXR0IB
z^uzr4-^s^`qwffB&feThN@@&*3CIPWs*DFF0bE}@TUKc-1da>vHVa6Jp8NG6M4~ma
zA9!iIZ`tM>w%kAxK>mZT2lF3wwQ@E+JwTaf*2|AZ3<#_L9Av^2Km&;UBRX`?ysW)-
zDXjwdD4HoZy$>tP$oPkG8O4GO%ohbbBB((>P|1$(+b1uw;Gm#*pvESEL}aS(kt92(
zFXq5d5@0oTfTV{9wG{A*En>xtHINKZtPkPP$>cHpk*t^`^8+MHMk9!7xLPMIz#JUv
zvc|@#kQIT>3HZ@oQ{X6_eT;%%^@4qmr8lJB7<4w!2hfl}aAlNe65dGHc3aciyJwFf
z=-MHn<b`43*LDOfF-(rR*JYsrF??H4hf9i&BOtJ`Jk<u`n5Nc2xl7+k$T3km5c*;m
z+*Yr5YCC77NfFWPA-1{Jk**&1{P}*{$?x)TnPp>Zlar6gX}Ipl%ge*VO2cR8VFSjC
z#p{ERP?bPx;ew3X>oDY}Ts&#5N>V)A1j0|6jgY9;=QFg_K!oK-8zYxsP+0l+)BpfU
zN=qw)tU`-mjyMQ=Z9*$Oad0Pha}An0fP87Gabs#Xvq1aNrC8hxx+Xqb_jbN{zqTI@
zcPor*p3S&IlH--DS8eMq72Ls*9;S6@S-|AFTPCoW`Tfn*-|FhDi&vc?uC#saX1c)7
zuK}_<j`5OqsalduWNS;@ZIe71tdM4rA4?W&s%v^oCLJxTg;WVN&CsZ!2;%KsEw+#l
zAJ4+XB!<faMCA4CS%*zZ;Zu<Ai%J&Uzx{}3e-sF5o&d#xO!heO<_SVT@vM2;yOgQ!
z_%lnzwZDn#{KmraZG2h+WEbk*EhOdomtBEagBBWKMs^LA2M=OF0A`Xndh}>W=9&*7
zG%6~yu#itR)N=vvW!hR|Z>}0D$24Y3E~T|v(B6z+HdWAZ(y~n{+gY-F{r$U|D?HmD
z748)vk$@tVc=Vh&UxW3pZ%3=WCUjuEf?VmE$wGy(`e}m<blaocBPl5fiBa8VCDI7=
zfgA!wqNUA+CX=olJzVW}=gFf--!(=GOw0~`mXwsVvTVnGCoemUu?EzAZ$DDrk?v}|
z(P=;~ul4<Wne$2q05uLk^Cm45#z}wO{8MdvBU%!2H8~}v_?{k3h~}!H3LKzVOoW@4
zTiTw&#l!qygSh~5G*J0u%f(`X>b-TkvP`S}%^rJUu;v#Q6hWWpfnRIMlvxw-7huqe
ziVvOAMfw&9IaVeQx$dB!okl;K=Od+gUujm51}lH_M#?Ds4j(H<wp{-1V{9k)p~d3b
zbD|_C=ExH6|1J>sZm3`MbERD-Eiso|bzxD__*^xE{arxNTEH)|OCur$>{-vB7YDTq
za4kN+Q#%dnS4eeGR-57gD!~2d`(=q%D0L-6J!~vBbPbIil+dVzD<0MygQ%=Mw}}~3
zIQ4Rj+2&JJi;~#-f`SZ{o_G`j0dJ)*S{2*RWP&1e2M3BZ3Zh1fV=c<c`hE^`U#U6P
zHgQZ+TyPfKiH@n4Bm3l>n<w0^XQjEqf>k;fE`5+9w5PX@-t<G7P9i)_m_FRPRlYt^
zb+xsvmkN4fp#X#yD<nc(t(%eH{I_$6q5wC@f>coWX}J39*K6!rHZV!NyuBe+1YR-5
zv-G-cOd(8#>I&5+py8#tZLOz(3Qu!Y-xnSSlY~h}x_l_R7CzTNwwi8=T@MO5#nZ(T
zHc-4AheD8zZ^dG()XV43n-O?{+-w5z$bZFlx=e7P{wkp(BOIz>i@;^Z>NqMO5!Wq(
zIEMzh-+ia7ZF6UHR0LEwcOVvN@GSUk33$B;5;<X~$b9XUs?+9UfTof<+wV_$lBT|B
zIL~xma_?TfwyrL{HeApv3)?z3P&}!qsD|<1nZ^gb89Go%ah@)^&EPWYN$>dU=q2@>
zTR5As<}*l}{u33l=I_%7-hwbCXn<5O@8PMcA^9IJF&#!f1(E|Wy|6xqA#x$8J~<6|
z?py3Kh(-wmsN*D4Xqjm@57Kg>4tJ#rU1O&Q5s8wfW;%#bgo&&Q709zr%abVxSd}f0
zwZtLC7ns;NfN*r6t^>KC;e2qb$T=JqFe+ePQhP57A(9_J6(pSpYYUiYR@&XHaFG3k
zHzuzFb!wd;uz>W{&Uj%wREK5s`0*e=oecekuyG`TLi!r!D%Yz!v5R-OfXTr9*39)p
z+V@=ms0;Tb?M=auiFevZPQ&Q#?vBtM%w0uL*8zI6!R#eGJ8dQhD0O>#rGiaB;2s()
zSOjeXv|4|80nz{$*FgCQQd=hsfl=Fe^3yzb0eA*O%}3miCIKnvfbmgn7co#+yGE<C
zgVbfqn$n1JLQ*~p<zh`ME$5YKWk3N`r%$t`OfD=gK6EFRJGIC^QDz46<6<$tKSr(;
zxU(TX_p&;}2@nS`=!;N%SbsUPvb=1!>5_8#6;xqUL14zQ7w5IC)y>Y%%2`@w`*Nb1
z4}yNM%2=*mRfO7V&b&%y7_1k_CkS+BsAb<M=y!AW%VaS8HfGE1t*z>OB*xSD7>%)M
zPb$y^P+tYOXMy&xUAiQVdkFe7R6EqQ%^)s)Fzr4s5i0058v&J4bf>Vmq8Z1bw79<9
zZQh<BMXjf&H@-1b5y}W9GHt7CyuCJoNOgvi1gf3*P;+!h^Y+G{HY>bd7s?R|c+AbM
z9HeK<7L$dA1zW&ZsC0EzzrJ~wJHz^8TXCp>lngcWJ7}pe1@|_`6c(=L%urQ+9kcCl
z?uD2!W@}4JM3a2Jrl!@aKXYw0$WJvGGPEYtoh<KG0IEmgF|0x^`)SRsVGj4~(t747
zha2u&6FTj2w<(hJRO6vUwm$orf*Xo^2m$I;xFOjd)e?h(j&Fh1sscF0wt>5ouWTAb
zUnR(~LrvVBnsFd+ERDC_hRZEO{AOva<t9vHnA=a$vQN|5lI$(9-z8Q#s9*>&?*c2U
zBvAS|w74WtT3`HOl9d9C+h4~F2d;y2Xej`+ejghb$L%otyBQ;<bgZn=DJz`I2<7}c
zS!s2LjBf$7(9$Aie-BDPT^7*dJg%Km$Ev4Zr-nvUmif%1hLw3(fTvuv--MDguy7y@
zrcO%)J$QyH=azyB%AA)7%I79B(;!+0w!7WVN^=`t5^&P${rLKqD7WOD!u8zdGfd~t
zv%)L_{JPR(9RLe@oPfh3V9IdAnC~N^5zXLr-*!<A#jtjNZB=`VL;p)ditRUmAXJo;
zagbP$OX2$U>;B`PK7C@<QU%nUSu)Kpp#Z=i<N~|)I#k-{2DW}My1&pe4yu%OY*7rN
zTT1=TTJW6@X1z{d7vh!&S!qS-fJu-|eb2s$;{s0!5?Nj_M_E`ZtD4RKstI**8>;dH
zX~+w}xVzW_Npdv7IZH}b7K&h@fH6DR!Wi-?oM8bx>^4$d4Ii{}|EM)+{;cjpo67LS
z8yNC7Q$tr*SEgX@F~BE&^g<fIx#hu8;LC}Nhx(BbR3ixOX`yB%Q#>$Bzj4xoAO=2-
zGAO!-YS{y@MM)1Bj*P6xR-Ud4U`#?!Q7$BjxP9~vq6mG(=jc2IStXPaF{3{-2A5e6
zA6iULD1@5<E*^(bgXqHFPyyl%O3KKH0aG$^`orGJua9^x2x?l{b2#U<Ot804^trBR
zq0>Vk@fB(o<<6eUyI|=Ud+_B!Mu{%woZv!1d0bF)nf^52d)Ey+L~VdWwvqGp#6U=&
zK*1c<H4>ykp)RU|#5o)exA<`%!x(53?e%Z3+5+vFs9}yS{gp#OFPIEEi3S8SMTiI7
z55FBj8OL>fu?5ME$W8-`xDBP}v9U29Z4q07(o}zi4wS$7-8KsW3!>V43g8!iA!($?
z)JMO16@w~ppfv?Sd;s%DyPF!uTCE>=j56VIj&{=IWR6iv)pVvpr&BNJg2ThZ09F|#
z7K@9Yqo^6bB;h~nB6D5d6W77quI4cfRXwO$@0RgEfmDFv8qm@N#1?UY^YBnXL<y-n
zfT0tz`NdtmP9;lQYcOYLt;wAg%551M`8cJHMMRAxS|QCHKY5Zx0>OiF2rWL^3{+GJ
z5CwW2M<a`2&^gwG(LygR6+}!1hw=a$1BnMAC}>g2K{zQU=27W0Vy?2o+<xe4BPg$h
zRI$~829@m{3x<~QJ*UBWP~8N&E}S}q6V<SO<mmadS^0z7lVGdWSR2f?1pW_0C^<>W
z$aDZ=HL=;kQt!y@70QU6lZ@?r$1`}+FedEjy0->RR{HF$1uAml@*Z_v0i2*^wGGh}
z6&n3@SRvB=D041AA}$o`A&*Jm9)(_emJK*$9l)I}ASGaWEZp46V3?45^eA0;XVZEM
z{sk_r185+JlPS1m4%6}>JsX(-_$_}vg~yFFWZ+L)Toq90g0asBe7Gy=Dq9TFb6}ZA
zHLlC18J6P4Y?Gz;!~B$>54y;}!H~oWe=t!<X0|Q~&Q_A3OQ3?a1jsQN@JMJx1#sf^
z)uDhTxOz4qH-J?ArK%gEgzkZo8v|8wTIRusT(>PJs4xm_Kh)DR0$@kOXAuK|6w25f
zy_VH<KyX^En}m`emyV%?4@nIiRFNUd3JS3$)5TH{tX09dvj~zT%WuDlLZU1=!J$3L
zmp$SFwZnFR83_P0k;IK=N9sTuiA6gukcnYbo0%axXO2}V+Nl=^1ftzeT3Y&Y=fIvZ
zNjPTV2}nWBO&W|I32UQUGlG;n=jsckEwA3c`l-HjC(^(%(zYc<G0B&+>NpkETcn-B
zlF(|I09%N9q@KD;-|{rX7&6urfj`2vL4N03#|7U5WLK*iGA1qQ%ciZ26)WO8)vGyS
zKqbeb<aC)2b(}|)1E|zKEa--n`Km68k0i!l>P~VtC8-Ielh$W3FwPA(YBe3l2v_0l
zSI)oOg_UQmKOSICj{e?gIJU~$eEEGOn>R+T)(n%GPRJ?fnPSr)jKu7L2dR&C=jG-J
z)&SiYA~JZ`NcRE0U(=p|m$EA<>)@IP(jzeah?I-#Khhl_LB*iAYH5j=1q1m5<5FSq
zp_q?xoysEXTD?_gX#c15`2UHElN=BF&v76DAfg$=ilG*Y2B!n8=v2TYP@HdtETAdJ
zgyDt3b!zAVF>`U?3579~r=Vu61o;gFiO3Phwc$`IsAYhc0^!Drovj%yrAru|efLNI
zO3MJkY6E(`MC%JmW+eWo1|7um{gcwjFboemkDwRGyopaBf`A7v$uaTnTck#fLFu7m
zye$dri^XB=l$bFS_+Vo+LHh>gfTMTcv^4F?5&P2}@N0T{b65{Va0LYgMhTIf&0L_e
zT5^}znM6dQrxfbmOA~nD=Uk!yM))lTJaqYac~#->G7I%#+)}{(UNeGaLKYAZaxnqo
z8{*RW2kipMLVZ8)t2;UIcPE70567E=0DzIBU<m~{0hH>zyu3gVYy&kuV&nihc{`|}
zOcMPr3x1$Df~#Qor4><(3b=elNq&uwr@XytK={yYrU<Ge8cq;lg?2U<G(J3f%Q(K&
zsnZGIIt$PR(*8lcdWVz(&1+a8y|)D>N`eXt0HE<klaL{7HhVJe@tT4~F|m%=LnA=w
zuf+}P>ff^t*JbNB&_IbZ8E9V^FjKLn(O<s|2P!?04;SEVo^d-XwbB7No+?N_G4b)Y
zP#}innNeSV7YG1A$2OxVDDkO-<p;6N*4EZVD3bPxwG)H2Iq_%=$E(+`(Y#9q0n`|u
z{2D?<65tO0hM|#>M5rQx%PxZN9pvkxSS2`3AyzP7E(%V`8#ivywt<cKN3m^Mn;?pE
zVCP!|4l)5;1PTCJ(xK?GsLzP1-hdg3)CNQd?tOUF0%k+-)v*_q2+TCgN=Tsk2;(?d
zN5IxvOqMDp2KlS&eEy;CJ(wKzkr<96$B!dBFkhmzwKbF=esMK{Uoa2GMM9y%Z3_Y`
z%96;lA+vzMGAE}z4y~oa!a~9&`jGgfq+Ds^Qy&h1k$d3e7}=MqDgtQjZNRRS)zp&n
z^0>ef^2sO)Y%$2-g<Av&@(uJHz+IUv)NlvfzL+Z!Xy(WOUC?V6UTim&3Vtdmlo8fO
zqdP%PV3a^!o<8>-SC9})z>J7^3@Ak-KWo7;E=S4tG>#V3+;)(wBdd_YUEKu4w+#uZ
z#TVfWcw&$i;PwAZtZA0}{z$>0@Yl5VvSDsf@mUNbFv_Y@V=<)g`GZ)fz3@avYVll>
zIB*~nR=o%Ra{#P2U;Pa$K@w2A;Xz9CID*z>2<FU<J%%x7_LJxHfH8&jmIo=}&1Y}Z
zA302#<Kxrl#hXIsV~s~m4K5&H;Z1;S7azD-o?_(qGu{;4dG_Y9jZqCD4{eqoeOR8B
zE_Hb>bLW+ULj3#7HIPI(nY<|X6Jg1FKp*o4JrpM?6)8)4@zSMc3K2~>=z#+YDk0!8
zU%ul&ip3}ZaG|qWezeBCrOX8vlS&T^^k!*Uoyd4rEK;GD&=*?EU2}SR;V?G#EAku5
z5r_6;qSCO|ZSPEU9C-Qs&g%mEH;lZF&fXW!o)ycur`kJL{~?mctKt;<3Uv?FwtnFw
zfudnv3>zI%=@X@EE4IB6OgggAW%nZpF#KLIJ!V$@=x-`N=X374G24vB-^4IRkVyDs
zTh|$HBe0n)DlzfL0l;&?_bY;B$C#eVj|6KzaunV_T@Q8V4tQBbgc&}E#sXU0KBl_r
zx;0h?VQpba`Y+1D&_@Ajzw*pG=f?`ql340R?wVOCzHxs8Z!Cz<z-0mb!Xg-LWPiI%
zf>|8+7ZXYJ((kzvgM+vd=dQR6{+TGPI5rY2W_ytF#x8T=lxJP7IF0EWgi1>(+)$3L
zQ{DY{w+9ql0}mu5Dt>iYA3RuFHL+ZaRqkC4A?NlGYWDCMR<a14d|5ToSc}Cxo4a({
zc)xQD5j9!H0DBjeUuR@6ktq4lJs$iKVmJtz4hy}x!$XQZ+6Ahlu&B(t<ZgoG>chl%
z=fVXAwKu<jGrz$pNcZz}jQ#C^AYtZ<-nc=o|D8tb6o!xa^>9=-hN)cdX8!LKz;lxs
z(8WAOVKruZ0M3QM(8EH4Rt-<5Zy(wcXL&ylZ$B>i)PE?{l$rM!HsFLNp`KJ>Yg=zV
zyc}jo<b7xM^T^?tsJ}6^TAA1y;nx{ohV^)l-5G1_48t^sX77i?Y*jKcQa9&!=DFv*
z60!uH0prri_dpoMUzL~$q#ay~cpOX2lrbt!jA@Q~GvLrCt*yjcvm|8)Y$;%B8d%8$
zxqtS6haGc1Lg=0wMXDifas$j!AxeVIHaQ6^%|4H?fhnzj_=iA27Ik6m#)66Exq12d
z-#~|`wEa;a8&XnYZEm$YpOx|9f2W?7kI$y=p6OUIGmD<X%TJr1f1T&`TJyf3=Wvku
z2(d*<M+W8bqp#1J&s>%;n_oVxc<YvAdxWTAZg8*XWr2HF5+^x1Umx)^{;bUB;l%RJ
znb`A`Xa<_|OPZGFF4U|o_bPPRC>-NezW(O}=(b|zJDMG9Ilv0D858Y2q4Dv>bC}+$
zFHkEC;lr<oHPvf-#E>9F(!SdsL-_D?&jlv2gJgj{J+<_T3VhqA+Mfl&su%4IeI@$4
zxZi#BWdC(bP4M4lvaV^_y-6=yvvpUx3$&PrhHsh$cA^Vn+D{I9zlKKk@DSt90|V3@
zT_W9y^zPC@N_AxWu$T~C3a;Pf6<RsoEk-<koHO_1*C&1*d0bfP;p2xf(Mz{INH5mb
zWb;h3Qg=|M@+&~}$_HiZJqs89M`4<#nyf$J&LD(rt38I5vFPrq9qX%;9`#TiX82%K
z7h;i+RAK@zbi$f;4+gw!g_9n34G(WV;K<53HBqbn8U=eW2&&mkyOhMo=fpAGxLJ24
znXU1&fXny9GG|I<4Gp_T3thQgKT3G#b_jPZQgiiUe%XGKW)(IkGBax(X=v@=(cWl!
zxd-k9)_@#13jd6WW2q5qpDA{kw~>)20F%itl?xLtb-WVvU%eX1SS|)BC=jaP5-hwP
z378&qzp6RSQOLI0V9{kapFxv)0%qvv#;8j>dn&D}+4$Ap$H#A4%CM)e{FjPcHNEp)
zAbu}jyg<b@sv}1R9<vgez%eB?qmm~nyRKZR-CaK~o*F7$K#|<hIn<r__o5-H$ctlx
z^4omO)Ey>FOiWIoo9Bb|9F$*HAh>Vez6HehhVv$<+syz-79ok6fEdS~dZN$@R+y69
zugwgNT~T2HuJ<sy-appD!YRM$X<_63@`$T6Z2TLXoY{7fR4-_NxBnAhqr&G7aHJ}L
zLFWk-ad4n@?524WtUT5ykfQEq++${eG|<;~AJ_A1iZ1c_^W*tuT8-Dc4jj;{7V!28
zjubq3LWO5*XIp1&C_+>%3#1RI|9;`5qoqv;v}`rn0R97j7N|UkS8^0h@LCVunNk-U
z64kSyIvnF~(#!vDvr;qa>C^cqmSRI5@d$r_phZtbI~|C_#e&o+(F*S0^u@(Rn)Jh}
zXGKIRK0XTNJ=@n@b5q^ezBO6?h0@p=x<@04`lg>v+6|gteUNa4EbKwb!oqR~tPU1;
zM@L3jSy+;L9NPfvH44lq6{dgra*JuZKY`-h*|X56qneQalaP><#2Yd7AU9?8qrtGz
z&T#6@uYRX%7R(v$9I(txn;<4P4#(f5{Y3PykTiRkIf!5D0D;=GbVLAyD&0`PDLRBG
z@iZh{llOvT){G(N464;kedxx6Zv|v+wkuZ@khvKgyCZKnfTe&k7#JQNPL{U!V)EFr
z2+c4`<>xo~uozDoew*ZHhv^7h%V7NGmzI5dC4rQUI91fpu!~N)?)1J?$Tptqwfcb?
zAC$`PDFzo8vDIC=8<3Y*Q&qMH!y8a(<pfGu#~wTi-}<H>I9;at^>#p=>Rzi*`9iBj
zLd&g;lF%dMG4yF;xJXT-{mP(R;crVwD9FDS9CitxPGoJ4eu?7?B^BKYdG(u~o<u@|
zUcgfWB~F;o3|c=QzLgucV22Q=>WO|A%BOz*_PYov60A}=xq!$Clp)qjmuziMd6+OV
zOKEGfhAc!5$<*EEaA2}#mlnENLE;gnY)-QHcrO`9!_26_gT(H%IwE708;`l1KM3C2
z!=fY18XTT;Jk{V{{8&fV0oucAP0ce{jDCAJCInV>)BbWT)Yld-5r~H;01Dg~94nKZ
zJo(gt1Ff&m&2uLWn@Lic<t%j^{bcQW>(6uMF(NM@)Q*u&09B<>Z5mnSy3H%52(4I*
zWDA^R1mBI7fnJ3KG;kGyR|I+85w8HKtAXHd8Z=0-{&^$j!(|qOH-w<ssCaE8JhI9a
ze3Vcwp;q3DDF?Xi!?@G>NXMXIB^h)u|8cOdjDbLNkb<H!=K2#&wNseVrFA+CpYG(m
zzZ#r_zoxhAx))|=o8j-8YDY08ia-AKbc~OWPdljd)i&Uzgr<|RPcN|;r@zLe7aT(%
zRMm5qL5&_6F1)CSM3-Xv-<`<?@fG4+QN=(e00O;kBWqg;LyR!R*$h6dOmL?yHs7SE
zIT;5QF=UQv*CP@Qj_ZH-F6W<$wvj%4aqBQVz2M_RCQRhoVg$*%xbQ5NOus0=rSy+*
zKo*=t8t}v}Nq}t)KiylRSI|wTi<STx8?guBdd1C~LDpxhTlwUa(FFnF>VQdscvRiV
zI54u^fv$*5b}}rLcIz;?a<Y4Jf@Mr9oP7&SoW|e{GzN0Hy_h5dx#tq90BLERz?817
z)6t~<x?VFsH*ZO^fS^OnLu%!x&tl;*0Kb!BEuriL^%d#1z)Ok>S+Qg%bm0ejDTA&E
zirwG7Zxt90-~rxv4ModOp(_J#ekYVxm!Nc-3bk3V2DOn;2o4Z(u$bDGndVna2!h5{
zwbx-<8XBqBpO9Vv@vdU6!5o#7p$-o<_>y9-m_{_HMy@z3nO`h;5v7_ZxU~djGV}~b
z+Usue_4M{GfwaM&SXNecY3mBOb^OkrJ&Q~j(vkd2v7-#ofAICo7ZxbqLa|&$R8%w=
zPmWg^&CCyTJ$p~Q$}&UQOl`nPwR;A`I^bF^NjxDXZv~-L&o_Iq7>7G1v^wwvnPNaO
z2@VTO1(uxTh|GCVtTTjytkdeCH<)~qQ!mmb+p^ncC>Lt$ZP3Y?1FZOyLJf&*Bx@}3
zh*p3X3n|d3g#*ksP-sNmA)&5Xuuk~xCX=DnA`R41s9!}uBdW&WCWs?6B5p3qA&2q&
znkM;nX^$#M2u$J9l6Ql0a+!2%=_x4>v?MFQp`*de1l<zgkHUi|6UxySBEk8VlT}w&
zrwEoNU4HN{fN^iQF&x}~TN^7AA@|Hua*92FJU7YcFbkft7cC14Vx;18*G&Y3V;S=j
zUl+KaY21gSoW*_<>`s|Ew($uGsPI2OJIk`N21y5^#~qy8_L>2t1=X%%%O&986L_9$
zc6V#*0uPcUi2KFI{C45;d;Jsh;$Kdq{H~x#IB9Kft@_4YE!gdut!X&^1b54f2miUG
z$L3Wip}vF32F{*_xm>6rz@R|oD*>D~lejGC6>0@>i?$8y7N7>R+V?>gMAZ^1b~a$|
zFJ8RB#e+l+9#3_LeaU9y<Kwpa;Sn3+^P7>Z!omuYFM})+s6Q3fdIi~2`OEjFEwAnr
z=Sdp})-DSS?Bv?7Sjx9M?o{_bVR`xLxd!WJzgw|yJb0fQT}ae>jhh-ajWbbY>eipi
zH+}m-&@D4m><3{u%%sHY?Q}<aaYiO@zPi3VK_etQ!{~fSk8wz*{qy~w*0wk~uRZ3w
z?Rn}==fd19(LyALfv^?5%Sgju#DXdA>{;3*Q#{vU{Oj|lV|X8ZADAzUi4m>X9OmHI
zu$7RIsI94Kf%QfO&0N83ss!jdlFD~PekfQ_wouw_S8u3Hz%Nrv6p$<`n>}^TaLQtX
zrx+OYR(;{)<GU)%w|b3oMn!e%jq?fM4=9}!g!(y(+ps8Ap@9Td?@`$kA}b!Wt;hvb
z@P%X7J;=rzba}+{ulq2%cPyToUKWUv$mybWSNHJ^x_F}cwxW9`342xzy{8*gxFZKY
z<m8|(FNK-DvflL{U|j?Ssn{3f{qEs8&|C=dGL>|;^VAaC>0J-Cq$^0m7{mQ4eVy7L
zlGyZ@yGBNDKxfdWrw5SWzmJVLQ!{c?3@2j3qrNDB8LCMyANNQ8C+I2zBWKPdo^a&N
z!}&Dr>=FD{@47wlivwqyee?wLtmLAALrJkQotb%3;)_KN{5u#`FmR@VN6SP+9GLze
z`vUMYL<>2qfQf72cd^QV%wNJgJmEp%Y|1M9!rCo5%vt*9F3Jq#kJ-=a+qU!f^vOt5
zf&(yNV<yS{T_ac5ZkURR`E7Iv{_y`+X~I>zvB_H->q@e9xGz0Z!@IjX#pr=!(a`pa
z>b*U)Pv9EBUzmX^){WHy_R~sfDj9-mM`z>29dG)ItLK@=^nVO*SDvtN7SQLpyYq?4
zIEVTuyJn#r7-SGm);swOidFZZl2zNLM}(spC+DQmYbp11+h?w#n(=vRO4Imf0?$EB
zPdE`E!8|B3qU?gBWh3G-@9X)jOmx_9{@F>t9K(LmiYL5jX&r;hcE(}fON)knj+4m_
z#kdc$a&;;4X*&$=jl5<pj_TWHrE}uYk}yHxzWOAoj+oJLOn67tdG9*ai3YEkCCJQh
z>aE79DLj{D*Ol>LRW;4Mwp9;KJ{ufAOSlOVE2w_MS300>;|k7W`V?j!o(@Nw{e5S^
zn*rYTDE7GfN?O5gbML~P+(<Y#a{TT`FG3qBvKl7USqw^V`D>3x(tK_T$(mbs{pnZM
z7|zvot{F;T#k2NLUcNjStmU$oeSl5SW*@m(Ps`;od#w3=*yOcl<Tyk!Wc1dg0WLmN
z+eyN{&oRk6jacx@P5b77DmwBf04m7S(7c^48nv0oC?J5^iiZG~93|vje31)V4U*C_
z(r!|XaURoiJ-l^T*vIGMz!_;C(T%mW3)ilxgPrh4wyrno>av~b>xCy0_4@5N-`eU8
za*Msi-r60b%`*dmYIIce_+K)E-@)aEy7Q3J8e~!w(;3INR{B>g^gQ+{Yh{PN9^`XB
zd-l`o2NP=tZ79mpn_w~|wcEYkD^0qTUbA-$Wb$zc(k;*x{SGSc60I|CD*_a%dmV_v
z8_t=t(mSa~!$XxFI!opt7bxJR+YrW$98u2>or(q6WC&$l<5+tF$-ofIW+TmqL=?tW
zL{7ksG@#1>_4I)xZ#ekLYaM(ziwn^^nlq&f{z(AP_;3F0sASE`pvq=A&y=a2=6NJ{
z|Nc7^kimf52sbYp8Q1@j+3}%-YUR5c(lIjBGzVe#N2z0Ok`$4LlNcN(yDPp&GbN{W
z?)9uM=}*et9WRpEptb;7HVAfzvlH)rco`ruynFM<HW+vR*aqYLA00=Of9#%d;BP<k
zAiZB8B`=SQjvk8ZogRM)3iB2C$E8NI(AKtuQo)+bv<z)+2U;+Kzd%KNw|zHN{$)wG
zor$DmFZrCA!~=u1aA16i@1d0VFShCNUpNnAs&$b%ty3jalKrnTO9LIRJY<=97}d%f
zPZ?_|2B#-zxs?}LVwXR|x32OSB**vIDo8lKxWcm#E?5s{y=42NM@OoNDi7JaNi>P5
zFT_jvk>rD5DE&56vncfKg}tQ3{(DLJtB3}Six#AuZiR<2YZ=oflM~Sxrk0M?3WZD8
z*5b~dWAgbr$H`WQu7{0#X@2PK&8t-3dzN_DDn~qPX=}~rFGCY2m$J6qa8$$j;StH;
zV1M8jnC&pbPT^)@Mv}Dpx!mJj=M_$<qd(!z!wHG2j~%z8*fCL|!CE0q_*1W)!GzJd
z%W*Omd*IxKRumoqF}?5J0pm%0$+m*6?*BnYLxV1>r_8EJY8p?3slFr&ZfWM*M`Kv~
z=);fx>#5UtdE4y!?UDSc<tH$j7j^jpG3=ImXlC`a?UEqS(vX07S68UG55u5TK%`TZ
zFoC(e2=8>UFxt-!y5#DWz|hQvMB<nhD{EU74vjBV+P>&h%0E7pw_XUNU?=w!0HH6-
z+S<>wPYL5PHq&1Cm~h4dyel5}92AL~lZzZ1_JuyHq=V;fHYmMUG;U;sn<3qJzdbzK
zLN8t_%r6X1`9Aeb&qv9p7^B2(|L;o;yr=l>*<nA7qc8m%@wS~?tN!v!Bd=j0nw~rR
z>1dQA9N4h1Nb^h@%}=m)Vm7~?Yw5TKXVywqD?c7>93D^0qLp&ep^b$vVypZzgIJ7t
zd5y)hQC8SsWPRkcSvC<?`4#<eTC<^`wGhu2_6*z~sBa3ik1zpyGe7Iy22gM(2ste_
zm#y@5S5gu8`FJI`oKCX9<U~cEHb&N^S<*@?iivrlEl;4jUt^pP?s^GuR6y0rx)FA5
zXo<LHod-@nwBo>q5xI|zwDflSCk5q|sw@pP$)8-jIv<Z3VsMi|(a{r^cs4hQspuLm
z@r0!!dl|rB*f#`OYe3M*0*hTyJ!;cIX9JUwBrdaTSwlAHpab2B5LlZYnI(;h-@jkn
zyXTf<4%F5ava{9MsgJ}Ofgef%%J2)>?Z&8Mye<MjrOD3r7E0tONyClv*u3>neQ^JR
z#ld};g#Qug->bYDqeZfy=?S{!(x7D-;cTEcRIs7I6%IBwDcG$67Y{wbVC3s2X^-Ur
zG&dQ_zqq-epsv0&cj6MGeMQ4$U=>&}FkzO97w<w#Qe!m~>cEU>{Z=<$BRC==9V#Sg
z;Iv2k!a)BPY>Dyf3I}XnGcFQ=(VRY=1|K+OF^oJkFv!&*$jQH+xYmr*y`PePv8LDj
z>PEcf5GZ=H*Z+>clvPv^{sLy+px2lz9~W`^8)}b(n*}e!50ZA`KCkwJXtZaqLvI|i
zOFjd`A|$F__nj?Na6|58u-(A{q!As04wg7DBq2Tx5n2(Dlb{y=0F|0rZ2?*=1_imo
z1p&x$4uVtNch~|TGWy=d6UbXecH*MpSSwbUw!5er?h;{_{jAaY`s1%?37P31#5{~Z
zWegS5e2_8?K?rtY;u>hXa+QvkqfGAi7+omL;6b(c>#IFpB(t_X^~W(xk&pPF2Vne#
z60Mgnahsd83{!Tps(V0l{dx8VX8SDoBN0~aaSnpxA+{$Zm<w1#MI-O)SzsO<5WbVI
zrXPZlL|Y#H^;PIUZ+w4+538iMpgxS*zC;I*rn?OqFox~M_(}tO=L6Y`XNT|cdBK&h
zK3&(Cd?6O^LIK~g__fDFHaY8Sl-jM@6^LE4mbBN12G|HD*~}+i&kNCdlDg2|mcdqE
z{`DK8zrJ$iYQji^R&^L}#rhw=h6uq>{hAUh|MDPZYcBZrqZj&bdi?csNDArowy+U_
znT&ew-?LXw9*p`Qzk00LA*WghvT#U0=>R3bczc>uBihSPU?i{@@u<SvF0SSJHzRVY
zT$fx)(A|l0?|xXJ-_CR83V{AwxA;C;Yi+p~W|hwj>_;VuD$14T=FdCRn|!v=hba4b
z7z<i{)Y7f^mX+N}u>vRm^Dw#<9Im=q4AYQwcE1y_TIHs*UZpM411dE5`5b=I&-rbY
zFZz|a6$X@8KpQxX<3D`ljjj^&_**WuA?GpcNd{v+5M)E>)ZlJJyV#_{3`N`k=&Cu1
z>VEiz>>%x!kCKv-AwUdp3<ajEq=CCE2rT^Y)zJu8(x8IeMs_mze+g{-j_)79&$t-<
zMRgtmM@GUgY&nCrWkOzSC}px+hg?*T1@lZQ2t2rW=y~m+On~}wB?+`d{CmL#PB42!
zX0VVLTEkSK?IkBS7nNs0M?f9&(CTFQHZ(XmAwB(~L@S&%I5Kj|)7bI?iYk;GpM{Aq
zF48??K2N>8Qet8%*9svZLB|=pHM+OJz;@X7!gjn>40RpAvEgntM}X+EK`g-?|63rL
zu4`=6k9Ig(qP5=Sp0gyf6Bzq^e|4=^j}#R);K%1RboPH0N^1UND3N~B%bt)VrKg9x
zf5*|$zuK-O36<RVi5JDd?Ue5@_W<oj174@y-F3hMi^|L3z6)hUpMx1t5q5M!_z!kr
zNP+97&!jahy>cZUcdRoTcdQ!95`E7U`HWwIh5<Y+G#k2g1w;zeE+VfWbn7%XClE<1
z>;fH3@C8g1LH4tsdiXHSZ{^*Y0G;MHbVS51`Qlu8N8R7K!IX5+)K5Hqa*Y4P$+46B
zJbrw)?+z|MqvgoF@bH^lUyWdb`5oq4aoXsM5e?@-4!xvt=Y-GkQ{xWxT5wdEs*yv%
zK_%L^<Q24IeH%1yd9HuH){K=thVL3k1qVlE?h)0+x#x`dFtKwV#_#n<GhUo6R=R;&
zN)x~n8N!3jhI)q^lzzAF^kE?m-^XgrLM<P&jbNi_oZf_yV#l|OVjo@viJgEymEcX6
zlhx6^1d7q!Lp5}?iT^JC<R@B!f8tOFy1xgzo{m%6PZeR%JXEv06Bcb!etCBbj1FI;
zm+E1l@AE_psWR>1!_AXXkW~1ph|G7ft2WclF_Rd$K0`xveFMs8*|SM&6RJF)iKF$Y
z&2&}qppDjPF;7o_%gWQ}O5yNb@#z;YTv5nTqh-j=DfBqf_XMq6EK4`{=$pZ^e-fGd
zzgT$}-iQFt3g$vlNB5HzA|eD#`VfIH%DdRP|1*5y;O-STt4$w4iZgsQAb`XE+ftw^
zpfL1~g2P_TcI%;b)Wee^+9DFq?HCXiUp)_Kv>}+4jgG<fH$f>)5q0Wa051G=f;kJy
z<#5~H7F)mCQ3S$Pk1$lY<;bA>z+9eF=!mErsMVf^uT+13v}Ys@B}RC&3-v{*R8D%H
z-i_>{>y}=xQ^OfmGcKRCEGN<2y?W6HUPlX2YR7)d{<Obmceq}u-QQ!jd!jq(mx)n^
zth{a7Qz#E!ZHOo>(0r99Q~^8u*uw5=AOT6<y^9B}W73n=ll05Ma~72LP<CTbl9zu6
zamE<hjdL_IzzC%Pt+cSooaP8%6d5qLFUtjc?0XOAtlZ7xP)kA`1;~B~4rOH7frhyS
z8+ELi1@5yF^=({$ZsWD@#`&qq$;m%T!KKIs+oFL!t^yXMn<4u#Z67poxM{9g#h<_X
z25qQR2Hhq|!T|PIX1oARb7)gK31p^)sCBwfK~u)kl5^v0-FsOYULZ!RG4J1B4n0@<
z7r1Eym(}Pf^65mm%-7HqAnV;CgNlidpPw?^Q7^S`8ZI)j!ENpe!bfoO%6(8Ry8v4n
z?4zJpQdfVA)E!XFLa#wTg`HbD*e!v>mc6UngJpgB_s|P8rL}@k1UoBbf$!y#PI*t8
z9*GB7h)6vC6JTkAQ~*8QMq@cx<%A`GzrU6j_Wy|H{rf+&Ao}~VMHpq^<?Lr)r(ZMU
z#8~CM?^6m+#Yzf49U?+U9WF}ngpU%4e?KNtpKJi2!Gi%-@T%Lr1jGVkjt+XCS^_eH
ziF|9~<vm;}H3x@*iRi&_>(-mo`vwwL?~4n;bgZNRP*Ew`nGGAQ@p*_pWzff8g@?9T
zsUMh=6aVc_%>zr9`r4mi=VDrRj|i)|a_=LaK3&iAp77msD3oE?zoNJQ`M#SyChd$?
z%xaosU%s>w6dwX(4Tc&1W>*pv#U30i@$Aen1SQJY)6|Gii;~oQdnGBUSuH6^J7g%h
zyK>1HMf!XAN_XjZj3>GkZ~L?861ZN~-@S0T+4kb)W={XA&;4zC;oPeS!yD#cN5R)1
z_PrjhuV<)k;oDu`G_U?1nN?q+3H~Q;`v*Mn_fYyDpLOA@UUo4t2{$+Y0F>1phYkuO
zAfh~JcR@&t^@62fW1Y2Nh}zU18F(mwh;XmxbR9Ak6n>uE=e8SHlX21JX0*b)76CaX
zl$TtNp;#0n`42hh&#yknl~7k#co_dMn@?C@&ywhjCIXz82U*vnJ^J{U{|N&9(@W0$
z2x1cS`JiUHhDd>e9}Cd<q6&gy(3>~OP#G<Q#s>Oi*t-;qsx+NsSNJzV419O?_`ifh
zQBbde^zaySeGPj$K&P7k!@Ud@2sO&#1yJvXGN=h??@*&P{B{gt?ZEYjPnLVd{b^HR
zODbe}<|tdlp{<_)vSs}!1;_(k2ogAGMaLJI_djE&%3p;Y8QP)zBpd2U#`Vk<7al$_
zeymhOUjArVBN_Lz=u0|7QyWjURRw(RpL~|&GkYGkTj&G<7nPNya8QvwC*cG;K^RBq
zz`a5Mfbv34#}o)jNV_xYJ>iq~Nq9~eqECS9Smbw(K5tiOk~vr{hF~HXQ=yGMH9h?<
z6tuvL6;>AkTu49!`gO{n&7!Ya4*|cM2rm8KlBj<ghF~6e9*HsekH{t(cC=f-AGQb7
z)+u3{EHFtq^I6zv^SgK7i>;ro-<kbRBOCzh(Bkl)bD%g1dX;usObdDwQLhtn?ty>b
zt)>Y2aB=O%?XV|As|VG(f8C`9x9Og1D18922S5KvhY<OI)kb_Rw3O&&s}(Br`m{`x
z?kj8DXr2gTC$<rKD`1}qmFZLf%HZ_O>tKl3S=rf+CMBoAd$h0lrzUju!wy<gy(Ow_
zD(TIAWzJe77hwx4fX3}m^;rC_8-Vr!rBChPEzFhOZQXL(UK<6z3%PEP-6_&OrqhwZ
zNvloVBa<1jRom_!d8jZ58yT8J6u>rLbg-QaY9PBAGJ3T~J#6cfD>t;2fzMQ3lSa<H
zr;8bP?h)!MhH~hlXs`u!S))MnL@7mCrqMak&Om7EKrP@n)UuBnh}%b$R8$DiirTrk
zws5*wFnX}Db&|SqZe>Lk8mwVkfo#%@wotg%ZP>pN;I$|;->KdV-4{4!+vll0>N;jC
zeYzmK<REINhO4?~DhQ_2;q0#uUVnX%!ZU-!X9Yz?y?l*{4Dp!kl=O6bmfI5Bbk?-k
zn5}4UTqn2Lp0Tu9kS?|7i%^peWHl9dNH(jp_qCj-^4muqL8Tk{M$x_wAZMq++=JYl
zA9ZuTHMA~)fpNe^6S%_r_wQ*&)iP8fMclQa?MeywI8Cuqv}KVi+J4Rz`dcL>BwjP3
zy$p~W81i5H?s}qzMqScH{+XLuCO;41eV{Mx(;Jv>uyi;L8fkTbBIP&gfcuvdDsj(P
z8j^xSLr+5k5+~YkbqHXKI5aZD!F3Aj)j=|t%Z=WuzkKMk8f*i=sb7CmHi&Q)o(MJS
z3Y>$E9j@QKE<e#8rGvrq*6z%+JPZ?xQ}-G4<?r_7p^7xrBhN^zLyH9}g*p}T_gfCO
zPX0+1BBgS(c5SL?j|JG66G0~^81u2JG{IR61^?lhupNb!2rB}?$^RI^vHv2<dYnc(
z5I`^-tpGbSNIFnDGK2@(=y*=tR}NY<881LDIzTnaJ9p$@+g(%{ct;?rbX^n}m<M(5
zG}@dEXv(!6=xC9^rK_0a5BqX}H+KmPA`_@97Pc&<N!Gx?d$IeAO65CnB$!|Dd7{Q>
zFLXG9$-p=!6^ogMG9m-OLteD|&e~`_^wL%j7y0@g5`u2te77x#=BPxwe(@;m#I|*~
zP|#hL8t9+0duw_P7z!9AP@QHuqCsB~uoCEO*O-rp#)(#tum_w2KWHId{uFG=XI{Q0
zj@A^^*a@ECH#7s+16Vtl%__F+P&3jF^q{5*8Dy(1!gH8_Z-I|=Lr@NE3Uupqki`gh
zp`;1LOJkQza0;WnSGdjlN=Y*hVg&lF!{v>CR8e5yJ*po6J4p%><$rxD=;wc<_y07v
z{m;+--xM8ST>s~1|1-J%OBDRyIV6TZVY6t|OyE#_2wR#2A09q>dy!c-)DrbKKp#Uq
z2+Oje#D!n~3KbrsQ1+7o3tuw`XewD*l+e|PIyWE;r*(D`5y)=G3@U=?-v3+H|JB}?
z$78v+Yd_IEE@jA+La7vGN@S=Ek)jaEkPsP?%(H4$W-BU`DMAP<k(mZVQsyCZk<29X
z^!Sd;T5G?<e)s<N{`U9J_xpYCKkt(JdEEDXP3L)>$8ntR&d^hAlb_5od6aWW$GpLw
zdj$@$-T?vR#3s5`7;-2E5>O5?FU&&-VvT?|XK->KBZ4lir_iquR}*3>XJi!1eFcOy
z0`?PoPe?OPL$h9!ZQVmW6TpW&)~I724#SBZ;Sg~*G`_E>c!&V%Wu;^iLPlO_so^v|
zGY>nEc-Wx!5WpK@3$bJyKmxA@bqBN}b%3~_uV_EK2ppVjmki9z2{`n6!AG%c_yJ>f
zqT&V4M|8bNqr>$wfdxQMLCiXE+FCVS`|^4jaaMw>0C@aE!dFd-E&y5igA33q8az^o
z=qk(>%1?C!D8WPrhY|v)YR!kI;6ZQ(dLSI-F~ny90c-<q&85ke66Gc%E4$>$g;Sgq
zNQon8niQPyv*z;(etzKXZ_m#qmb5Td<K^eCoM}ac*#=&9d*F*l-b;k*<aq*P5IY>W
z0mXt%04L`<trnPQ5lj`~HNj)c0(VyQz$zh@P#{_nO>f+I4oN#X9#1Y=alkX3knIrI
z%nn()C*c%6zmKtpe@$@9B#!wB_^7EtZ9}RMlF*JO%}Ub7F-z0>`1%SVD)&X6(S^(8
z;BDh8MI8+&LFvQ~hcq*o3POl>{N%Se7G-rmyug<$p&*ALv4UF^7oH~zPV>GyPd?mw
zM}I3QWP%6-F7MMye-GoVLjWUu^{79COAsXbL4d@Js7wHMu=-%>#xyiKz#G^PkHj1x
zF&T13OUo367*@D$mEfdY|L%Y<xJ#^!+SYzApd5>dj(#g@+kf8*?&GXr`j%jJ`9kW<
zis!AEAyWH;(Amep08Z4Mn)d<)u^32kQ5O<AcKo;>yxwbp&KziW>|y5;5~`a<u;>d9
zL@rU$&)I&UurceojNT>z8_WZ`h%X#s3184fSP(W^%}~F6`zAWDe!~V~N$RvgOkQ4V
z7E&uecu_Ossfh^)9UGhZbKU%A^3;rc+9uO;bK-@;6c$}W!>9Lt`|TpU_#2x!@KA87
z9h`SWpp(4xBruT>hQE~`^w-T<FqkBdjnJAIwK<{hm64Vtf+3GRiuq0c7U9FKfBfnD
zIg8BJ>pr6>)whbGt-vUZJ(wZeI|Qc3V5>ju-$o$;XGPjq0>>m@$X#QeKmRzhNYHxg
z3qXN{xS2uTyJ{i6efw(%NWuNadMa7r*>C`0ck1MY4I#%o<Zh9<L5#C{x)5S9OU*S^
z`a-(a5dv2c>v@Q@?%?lO!2nMylc&r99Ez|2J#a_5VB{gZ{vBNuY~#Z#Q8Hs8LrRvG
z0daru!#HT^#q5=4Vqy|6E%9YG_y~H?f%WnS!lC>d%f?VvWx<=p7rR@}=jqWdzmECu
zhvZNKt%d`d0W@{>y2dC@lt2WgQ$nIG1MEF=FwUrkgJ2i&mp=ojK)|@>sY1#9J*vLg
zMOg}^m{#^es{eYX6*fLsib_gY@G?nX0bg*f#Uqgs2t7Z?i!xCZ)OrRm@V}#9*N*WL
zmhOP!xP;({VelQ(hg|Lh+~<WcsSCFPWE3ul$XzBTCDrj`Y)lY4M>pGJ#Nr1Xe;W4K
z@vYB5$}We3haY*UnZ+D64?E))%r;xeOykOJF-GMNgOiz`oeX=sEC8W+%-9&LI|$td
zIxiSW+W4<pv4TMNn2XqW-{N&f%FD^+gw0xgwI-7e%d0M4nF9jL-LMcNq!;2QUS1%F
zvP*rAty!jS555UB*<fh6%jdDgA^cSXF))WZJCKww2N53e88f!nv112O;ceuQe2#V%
zX7!2cPRK`yWk2ce@rOYOF~wg0P6dKFvJ>ZHV+)>luN%*Y3Ef^sAK7zB6KOD(o=EEN
zNqa@T_5#Ash04YG0Bc>eg}~=f6(+&%pwwkH*nm8F@&x;roxl1-me0ULZZx+zpL{O!
zbh)*03)9pKm{e8J$=~9XerIMg+}hO5mQNVJ*qu<?%^hrHq^B2^s;sQk04XXwU_YbJ
z*cjnTB8nyyXd*U*ZnPIkZ^bX?AESUg0K0>$`KNKUIx@W@Ou43R@S7)|LLianggI;P
zgf}V-6t6ef!Mub}pRf>qTPmaAFs#A8T@Mi^aUcuq`ecZGSj?vPH{z-<G5_OT#aKie
zc~PL|iE*&r`TBFKhq2nk!}5u{>wwz~z~l0IC~RoXZ`gW3npnl=Wpv+meJ%p-X#>37
zIN+$#eE3{_SEBakU|wEcRIGXjLo@RI#amV$0sNhM5C+sp!j+JUE}f^S$7!MH4@qy%
zU}FW$d*FEhsq2jhcdBEsPcI+o`9X~RiFqFEvv?FRcWMYR`<x%yI*_Nw;p=)~b419|
zreVQ2;V|hICNm4w9UZfAZf3xYguvV|GbR3O6U6=mC;wS2z*m+XDNxW~0RfNLzqPx~
zS*WY2K>?`-jxiyd!at7q(~(yQb0CcO^=)6>uwjSmX9B!P)7-}=U)W}aLywf?w6F&f
z1_0@#(BU?hoB-2msf>xcJ`!M4KSKUYtcDbbnGl(z3N;Yll8k;o7;zZNT8M`y06Clt
zB%UMCO6JV})C+=B0(sHO5)=1vW=)K^FJ2%<d)O+%t?h<e6HsaI1xE3wg;`ypyL-=o
z_THl0mlm&US6BUDAqWKABa;_rQy0(Tuo{P4*9?xGS6zot#!QCez;5tMNpt`|{~ISO
zT)`eAwjS6-1|FHERKv+-93G-Nlfs0-3kF3Z-d9?a`O<PiV8GJz3Z{ObPgM~Y0?gW_
z%$9h+_u#gANaqMrlE9LQQ;tme@IeuCh}1`yj5~qXkssUh908uGA?hNgdhm|mMIb?}
zqNwO<pcp7Q^KA><i+LgD084}bLoi?%YM4q<M9edRQ-QdC5NLyV8Iz_B7^%2dv{7z#
z4IF%Sl@e3Liw`{Gyp-(kEW{WOMkSft3Ph-j!-E(f6VwO33spD^_!TN{(U4jrA`!D>
z<6>_HxR|SBcx>@Sz3z)s?mEK*#07@zDW<T)B*zJ{<HgAx%)Mkw1s7I&7*pcuAGeEq
zxw7x+O}WZTFLt9?0W@u5k_A(X*zYQb*QlI7zbC1FFLH0N)f`~i9DLBEYCGX$&-N+1
z_xGJ;F+a-qns66h_YrwH^yt_=4ol1pz7$yhZiYvL(w+o;Gm&khJ4k2WSxTg|$G^vR
zL5wESk<PBJ2Z4c|s`5J}^byx!GAEAe>5zdibMXUl<K=0+(#J$XL2QG_W&(Ez6TAx2
z=_9`Un8cGt;SGe3LnZ>Ky90zaU2CD2ICZrtE%vnDDid>paN>X<<P^lQ+)4FlpTte9
zouHSfI1JV&aeo!UQ37g^Gf|I|m$&KM2FmxU0m_#TR?BBO|7CT`I+S#VAqF#nPw}SI
zoVQW(*x`bg!nSOd6OlgOA@@wJ<a)-R08mI%BW?v^@nQ>z3sJ}OqXWx5l8(Z_khcdV
z6P6eZ%)mYQZN#Bg7_S>OC_+=O|Bl#V4_G>Z={U)tFBh^&cqWTE+2_17hUz%yMjG3o
zpxxgmz_L6)rSHdNB(L7Lp}C@9r?v+WnsdYR*l6M@&^Oa04qI&Uj%pweXeMc-p2J;b
zSo~daeV`%#PW3hHOk_Cv1g_CpcJPBkIbpNL_NH#ng}jBH4@{18l2`WTrxZAhq)KiJ
zgtH}}RrVteD!56)aD--E`#YP2Q=MbiENP&D%wkF=fpQJ*Q<Q7npcnwEY(y6+VVCJ^
zsD$YtR8@zAh>-Pnr6oysS7r%V5w>XV0Rw?Ez-{g(t)y5leK^ZchxTjbqxRC4MVIeM
zd9H<jBe4F^0w(xvNJ^vNeaqCE{1iKsXhT$B8~FC^F{D>cIrn!Zfzmbmvz%W-=MDxK
zvpW%Cw{s4C)SK#oevbSUE7X^E!O57IPl9oZ9W;V3UO~)tHXYfCgw`$nt(aaRI2O;d
zFiU#rYAp8JE&<uQkT@hFa>#okWQD^S)3U3V)rB#?;!xE9rUA<VuS4`5uifsXb{W4u
z^m}CLdXW4Kk=YUBg`O@9IZPAI$7Q~RbL5|jc%0b8;G7utnScIW?D?cT7ACRCT1x$#
zzTJ~gaYEOaiaJQ+oy&R<rX|5*F-L;KV=Ag~!4oJiAq^VBr-lgc$;yY^Zh^rv9$Ow@
z>0s_YVs=g@ywiAZ&fb?t5f$b(EPW4rA$RlfRV2?)0HP1e<E6f%Gl|;PG;~(&KF+&o
zlT5t@k%ff?QU>i<7`CsEDF5`S9zC*%%AR<UL0S>WX30`GUCdA1K9Lg+AWUr8nHh;0
zm1kQ`9IY4_c9vOjck@WicnxQDZEU;tZi5r7Y{T`SDj4aFdwi8}+JynfB0f7ANO&@!
z<Psb93Wg!mm$0<Oo;v`qzdA@o!n=q~)FlH0VzNZ|?8NzYY*Br}odC5cHq*lug1DEF
zco=x36K26F1e3Gaoprvu<Il1jxaa{{5s_sfLYPgPSoVxLA7p4VY;JX1a&;7ygHh58
z82iZ<u(RxPZjOaAl2#Zr@cHQkZgU+!a|)I-1e_!evwHKLOysQP?t^BEgn2l2;?vX9
zdBtc*?<~U1CjGyV-I7mY|5B_xGUiMCr(n*Zg?&7XoHH=sJxe?qP_d2&B!Fs1N`u7+
z8kS&htR??Q(pDZG;)gAU1dIs1pi$CBl0odTainzp%s?bu+p$QsN1~#PS;=!cOa*Eb
zEKbuS%_K5GK6M4>1Wv#jl<3jXhqtz46J=b?Dclmg3>9So0XLW938W5Ks73NZH^lG6
zFrKfAl&7#u@+uJb5crDqyIcji888VU!T7TszAl?84qpvoScdQi9*1ElhC#8waRsAF
za0|@qiFm061#JE#+{O72O^QCaOT?1TZCii{*ArVtgvaD$m6|D2PKQ$=qMB0(vof)G
z6TW=8i2OUPx$#{6IIwjSU@<V^d_Ivu%n5<~gZrZp$kw6k?(WY+(YUEApE@QBCxnRk
zA55~DrJrVDYL@`tjD8ZWe%iEp^))x$7MiChf#I~j|I{>7B<C6{zOxpq6H)+TRfc7K
zk_3jR5M@BJCb;+@={pMYM2Up*pi`IMA)8Wmd!gO|+5%tQ#%n<KiAFWZbG3EXWnem{
zD9rF}Z~<yU_DJrstC>UxbTGdS=^|-G6KE>o{<|VroS>G9EOq~}fj2v`Lm{q<SFffN
zP-o2)YS5|4w5@`;g8}%#5heq}w`wxXZP&~9lXpvE288q^u_5-FlNsP#k+2Rmor&Cz
zh0~Gl)NJlOqM}(PV+&-_Ajl&w(t&L2Si;c9C~Z-LQeQpLYg<4RmO&qt`Z^0!2!DFu
z1KL-mfM_AcLBMumq(K&zhMucBk;nB}NTNN~QWGvXSZ15gokiRp)m<J;RwR(FM&e?K
zgCqfI7)-@#;C38~H#q<&KiiocTvu+IA+|^gNI9sf2*DMNEOA(<sloaj%u%AeMJ0BC
z2xK8-tw;75Y=PCPMG79kg02?KTqab_Kp|iX#Ag#-vaT0oE1U&>MqF1UW_x2$45>na
zvE&%dYli5?>H(L)8!M|JbX5-a@X15;9anf66EyC7;p8KI89A0a#hugB&1P|WH9cWs
zwu#N35d?1lVt{>jE9}}1BE)Ulo>7-xIEU<fdup2|>W)E8#23(gJg3A@u<VAD;Y*zb
z_1@oLpp$_hfi!}G!5&$n@h54nErWGnRHN`VP=$0yNNc0`lN1jp`Bbwb*t~&l{zr_B
z8*<-}oq=$p2F(Wvc)(Zb=>o4s%rzM}6p11OpTdV3Mk$?%nKRJik{$|UJ0mM0c800T
z;G_%X)Rz2>?fxJ~_?;lZDR!o%wr=#mIY~k7oqF~Nnm8s-5I<*BO0*HGTboIG#Chqi
zvMj}UAo+LaLF3HHRxC@t@~V|9Rw!H->V-g!h@>%ZgW!OzzcuhF5-?NZ-4ACu!qFua
z=u&jyx)0|%BqU$cQuAT|H8I$vN@Oqy#)sMTa%>Wabql;@ub^QXIr6v<x34A*&J>8J
z^`HAXxP`p1AL1{VjwVVP9G|AxBKgKP!D5X#NAB2h6bGZ-d5d)(cgd`lyn~{W>mQ`D
zr$wWp{FS4croS_+B95s;O)uq1rig}XWDSz}hdm3a{XiX(0VshK;oa)g{Xk>9V{xH_
z@QqPJehRRrK4%Cp1E~@>XIsY;^bt9r4q8d5BY$5iNa-!k5`@zh!TsJHJDxN*R)&jf
z5PA^2^+*a0DJ#jth&mL;PuvQR4_Vq(WHy;U!U=|?XB%b?ar?enc%G87o*YfL;6veB
z_I`SFL#(2LmqFf*b~O}8c@VHGkcolq0wC@<s+_&BMhfC4i6ek6;x<SA9B7&fODJY0
znnwk;kmgtSp=L*~zZ2Jm766~91PTG)!>*+m2+Sx3lxvN-S)A`p<|1&`R)n56abnR5
zF)U7U(r51;H&O6@%v>hz8r`T6-8jR7;+)>ku}^+_JD`2*q4P#|6}-+I0ZCZoArP=J
zj=}m{2|S84cQ`Gacu!r?2_8e--VZCqIyhS14-HK~74;A{)Ms4dIa<{Nk9JkuSlNG4
z=GW{|UYNecVz3kg8&dUBalB~tl$MnABT2F^NI=Ec40{AYr*Y!pfZD5&ruUt#1aoQ0
z9TE`;c=JRaWP(Xa4!~6$E>Q%#AlYiZlWj(7GIBm7`h?L*O#exJ8kiRuLX5KM7!0;h
zAkRR5O9cWHs9hhW<EKO%O^s=s*-oDPZyY5D_D5)QB02qq7BI63l0vjeu}vPk05&c$
zxP2t)nuE}TP+N%keBe`TkHmxzgJ=T^q!EbfAkI~|LwX4jLkhW{KmAk5L5P5UoM05_
z>rsnYi@}IRZj)u66dipGZ**y+gdA+4lc|uL{e-iE1@_KS|I=-f-JqM^34}2u{2|7S
zV$;LRvP?+;z%Nj>9gDBOZ}YW#Xl&3i-iUg3V=GhMKW&UJ50c;*K<*h`-K)OvWzvKI
zxJAzy4=}s8JpIw6eWax85fz25qp4%3FVC^(S4fPDbrc`Q+)Zm@8RTTK6oBqLAH6bA
z@u)`@>DIK)m?CtY+(Q5~JkPAvrvz7c#QEy><o<0>3^CNUd9&uxu1P@`b}8C|8w)N*
zC~2=jr~%<|sfp;Ru%$qKtIyWg#@OU(^M^lJZ4|nA6KV;_r~ukID+kvYlAJ(R&=TpP
zwUOZKMp&}{l>rE0ze5RQ)Q0(#7E0;RU3K>fP$K32+}Hnm?5qD5s>B?mF%w#^5RYvj
zD9Mv9@E9gj4JY{?uq#Oe$0P8wbKqE_M*=E_x?9~cf8ycfa9a{tMmRS4tRo!1Gm{RW
zk*<?VoJW*yJ|C~A@vz58dTb*4JvdI1=0~V#nZR}n<(u$wHlN~X>`V%Qq@5?=6=&F%
z{SYIvDN5B30B`i$whg<OXJ`@hDfeTju-QWbLNAaA1jcXUs4PJ0LizFlY&#DYlFkGE
zh+J~C16Tl<v4%;{Wrvhhqm=IDKT~KMH1Yvs!Z}K-4oPsxLZ=boBZ1K7q5_~2$U|5F
zS``9e{4QV&Xu26*y4<K0Ts;ftPC^<dba^9Pc%R`^->E=2777Xp9i94^D<XfgNU=>(
zI7ra)FA`%J85trF41CI8IDZ!CFlk%E^I-)gm2e9=NHZHWfHs2(HmPKl__HU$=FtQk
zFf9<axce5VBa7)U2PK;c0xQx&4~gVm-c^*!CHc%5$ktKOvt=eW;A3M@2q0)qeo{_D
zNyV{^pyGd^{llmKw`F+gTEl<P2Uf{UJLix(t(bA<$-cg`3SM1Sx)EJ3xSPx)lGdG~
zSrTc~2K|k4n6gAI6q1;kfv9guniAFL<$2)H#y|9lr<DJWbNJV_&=Eh_DJ!SQ=+k9z
zju8r}(@UhoDxoFP;ky3@<x}StQmwxejb-$}WSdy@!n^S#w#wO3RO=R7au<7Z24>y9
zq&|+5<$QU6tx`~2hFsdKo8e}qqHdd@G%lAb>|LAQ{_>W<ClT=qX6t9-hX<wwUUuqp
zns&dJ<!Ncsi4o7rXp`f4*ljM5J|@qu#oP2`gT<Yw&pR@tcYK-(FgsSuZPDJh;K~tT
zEqZ7wT=}h!+ZWf!JgE^525=ue6n}|Q7C3T*jrGt~U*v;_un<|1HTfaf@z4GKj)Qtl
z8*=2)WJ_1x+1c4n*AZvE+lebEQLW2+g9h94#l^(LE{%2;gE1ARS?K0!jv7X*aH!HQ
zx1T|gGBQP~$`TBzF5SGmXZBJk3}3k42B92R4B}ErduB#Pabv+`hE-!LNI;_vnFYd<
z$j?fYQQw(Hokn|D0G4pSEOT0kgO_E*sWG>6VZ`122!O-U3xyLGM#D#3cpg0}+0@%}
zJ^Ow4bvimLO!i*Pd+7|NZj>&E6&SATQJ&e+j4NnexX^`$um&TW*v06RlasqrW(JL=
zSWzVP_4F*~%iV`+ZqsyifEzuN_O&-z139_5F}b<ooSb6f;s(fVWA{gJ1FHDz8^23O
z8UkKEc;LV<Xl)$;+({OFxt;^LX%!wC8hU|<&C_$beX4*7)nW56ck0XoSPFUwY>85?
zX>DagancKrb`kv3JNuJT{c%(mqi@_L9O`s*qqqwp6w*>0+6Hu)fH`{tnh_TWRl1F}
z8O5gbHAF^3L&L__*3T>#rYofdx;Zvih-q&(*ry(QFA~<*W<nYMjc{%~mT3Fjg!{XF
zNk6beD>^~+1=35Z-T(a3XZkZPMbWZ!r%v5&bSbETvC4v3ip{duAHR*!wYAa72Qap8
z+Q`^0FJt)f0Xziz!|6<(osp2>zteLDziVFRQ9~E?L{4JXL`$5ag6<b}&n73+?I=;A
z%W++PKHdKPT`gK#YlkS(eXFZfV$zyCwfQc6ee0<d_v2CddFj=Z*vm{CpXoO}f4!d4
zefi%%T3vExDXjAT-~x#ExYA0WIWMz%;-g~+Ma4!g>Pf`m5mt)#X^%Tj-}5XfyTz8V
z^oNcW9KMZ!PY&IGZES3ckO2#PojcTe?f5>uf+J{^vyDl|=1dsKXILMX-_LGRJb{mk
zG&y#JD(Um3Z@Pa+<Ud?41{`ympgHiM1hNJAvTI%!`^YnujNeRb7D~TdEPDUmXFzJ)
zksbu-iNT_=0<N`R9SxuU$iR2LLPZ%a-JaO8OQ8(Cgg^|%@K<|MJ5Ue=a`(iS;Vtv;
z)LWakL_KLYdP%oxBmJ>)6Nhcxw6Xdd{(ePORaHH0+_-UL32GZ%r!1XRvp?JtG+0=*
z^D?Wr3B9CB$P+Zw(u7SXJXXzWM*eB(@Eh1Fm7*NPGEm?$-$hsIf66>kkSAJ^%EmGr
zQOCTDg+<ocS--i&a7ZYup`k(4;m4QHb4Xca&CCX#Hcn1V6vs<WB<yk5K;s9SMRWI2
z^a<0<fVkAG`{Tsa)S(7fW)_yhRc<o+5-0eW&g^vGxWQhz-**pN)$UyvR6f_7HHrU<
zF6aU)SFXGS1OsG+7rg@mK7^W*brn_<WQAu9rj_qi+4-@`Yd}gM^sO0R4`q+HoW~ux
zue7(f2%$|G%7d4HxV%8kFAKfR4A5le1(K`JpeKB`g`fNjgzxVS1sI+)#_=qwNHVq+
z?0!VmQ{6YKul&t^T#ijjS94W#YK*7SR}Z<sPqV-1EB(WQ%P+(s{y~%Oqc-)4i_J){
ze@rD)oW*)c;jL}rYgK*Hm<FbH^MU#&{b``7UXU?Dp52vdGCr7A`%5tomYwiOb^NZ5
zGbV|4qFWFJsf;!EoGyr|QX9IqYq^Pu(2UF1ts`-}N*!r!W9)4$(?4!2tW_x=eAugF
z5x>J*8IKuQ$uud)&bBX1u~*bG>a-z$h1{^W_P(&(0|NKFZd)_qr?3^j-v3ocH0s{?
zxUZDv_Riw2yrGI^by<VO=Wm3j9iZ9ZS7g;rxmJQd$gRrWG9+qA2@%2IdUfS3`{KK%
zs_Fz=&xX3B1iP(iW54XYNVh|X1vfNt+2-z(YT23?@%0ivq>J<x>Wd~<FKjn1g7`|M
zC9YY2ckpAUo*kmljxYVJVB#P;<P@?C5z#XX)gf*nvvbW82TuL@Tm5IbEfvS_!y(k~
z^<JB(h*KHws@Uy6J9-BGzHU`rZo;{E`RQMl^_QxNuWxen+>D!^xSCksckT)8YhUaY
zBXL?5R~f`8AAv5BUwr?!U+lUN9#KaB{g|@-(I>K=lmjK4$DY~8_!v7mt$lJh$ZIR}
zp7c}3fBedy4hpX(^P(HHxA%w1y&HR0yMK53YC1{@cCC9K4gb6dcWAh4-P--fmMcba
zpX8Nuzp`sfb*r{R#?@P!f5C-Hwyi@6{u$wAm6Wizy*1u%-n@ODRvpV=eBWAmK9-K|
z-Mib3(eu~0lED+Gx5|vyS}z(DxlHtS;J)OL!C=Or;CBvom}`PT1{UnM*U_I*cz7;s
z_-=Abl&3EyjFEa^EJOU<*SNUo(^;=1+zhrvE%Qj*TUO&;=GNzR=dO>)*Box^tOQu@
zQB=<DJg;Td$?xYGB5XE4;8k($bo$UJc^b!+Yt-KN&YUg{VAAzT`*>+Q*xy$s@5iyP
z6Vsz*sS>viSHCKi#|?Hfb62faJn*%7a(Z=D)w3hFma`mPvrF6FHa0`m9rHyk;wX(@
z)bcGUi!+h6%?VO8LPC2hx8D$Z;J%x~=c*q4sm0ViWJUc9mce2wiEe3*>1zupiRPG;
zp*{&!))1t6+)|4=Gcn2d{Ya#!-#fZCX}@x*`vJr3ZC2wm$2vr&>n>_%4&`iRX7=g{
zkdkU>$dAD}oW>L%Ad{I{=x}MeY)EJ&SCz`U3X?4x<ZWnepBZaZYC3GQHJ|=W{%?FH
zruwDXWiv^|v;bk@;?+2Fs!s{Yt@=&=P_^fY_pkkXSN4C5Q9k9qP4+_u56h5<v(wsI
zk>-|lU(ee&i~33*_e=9y`C4YkI$kP(v%5Lyfgy*P=^e@j;SZdQ9$rowQ*z~lqhq|^
zKR(*>^6}*WGqcdP&G|{6vM*jHXPQVgC1kg=mB@9CZ<oHd-oEFhL>4CX=OWd}W%xs-
zpIN!iy~ouw$2<?LVliz{5~+UQ^QnopUvIK=@ab30i*cId1{B+5-j%ieabW&_Uv?iO
zJB_pzMq~RvvE?3}m~uxoZB|o?cSs+<b?rJMg<igGWiQa<B7$M*tlZG=>;IYWp1DY;
zUdpG^H|KpWi3*Im2$kN834K4gFVUYS^xy)!v}3)62NiV;j#S4_W}UPg)6#=zM_1Eb
zC7o!O>Ej|vt<<xXZ~1XiNLI#rz5K3d@3n?;<Mxzp^~`0nmelW-ofkT0&*jb5r~2FH
z59nAm83oJo1^a*xl~WUYriXk-*;-b0h7|&;v*Nq+O<mGXbk|vU_<K4umFqHsCg=le
z?^1&R2}V+F=17ITPFGOmykYFg7=QJlt4#5{QJpKU?;5vP5mTQtyyu+&TRMQg4s$IX
z+YlCe!!=scZls-^sGD8zkNyE}<1r8T0>sW`<m!Z_cz83tSLz#yN>7|D<7809Jp)_v
zLTO+J=!{RYMoU<`Zm#3=^z?lA@ZmHf!3$sepBby?wiFxQ4)1+`&+FZui1Q807ufHg
ztCu|g^iscwbO(u1%M0hr4MrjzzlVyVU9gM{sAD<D)T>vY4fvdLLZ9xoic2(4B-^4=
zjbQnsCxy3a5}SE^)RZtfNhvvdADv#k8vIIc{CDC_hUyb%F-l3*Z{_4T8TW-uFKmfa
zaJqb65KpZ2*Ap+iUB_R9XrQDWYCIm)@6WyQj`|Qb9QxGwq406{Rk7+B%8UsiIUB0*
zI^3X^a2;F#d7%>}3C9(@H}_iS7zg4x*5f&J$#ZbxIS6^mUutnoDgZ52HGjW&)a4r&
zhZiY~*BBk05N#7wsN83Ehr01YCCflbgLWT{Vt0M2o2?57C4LZ)Af!Ls6Q=qV_u;!S
za(tXV&L};T>Fub=Twj55eO1R%N57O8+jMpb<dldVV`|!szbG#=Un{ks>N1tv;fHo)
zcF!%qK4o~mI#Ef)xxbxi(-CjlI}+t`I-^7@xM^#@-P+O->hMSjL_9Oo-*}{$xwyDS
z?cJyB52BdbmxgH0eO0%SpRg@!)MA$um5U0_hZxGQOzyMs?krGOUOWPR&jY@7QN4JV
zx(%9S^mp<*D#*~E?wpOpwiP1bY<C2ng$yZcqo)75#yToBtJ8a$0J2Nd1Cke4LxE6-
z5$BEzfB#nx)3^e0&;+(eLd4O>g)@(FV_Olr`+%b6s3lbh;G=hN(AVGJpYSesxh#To
z;fcCLDN6BO(LQ|MXo8XXd2zLh*raYEuW@t7GC>2q-jZ{{%W@JD_Mk*5I!S&0vcem1
z;OPP5yLayzz$?>gB!8*|Wex)@ofjxyS~a>18gTf)zp)6s)J{|&1MCrP$|iFtUbY#D
z#SvuH9^=fvrdoJFX*RcZ1EY+L%hz>!o~Trnpy1Mp+{yTi>B?uYs*K{|MA?rWU51o>
zeSNtXTH)pq6BEOQ<{NBKfO?UIidhwx=SKcEuyDcsOdq<`ocYhZyj;_9%J}z-7cXuH
zfst*~kss1aqYThG|8p#D-BZj=I5-(=+t#aFHJmptF4vt*o6e4$%O<;kU2`NW3yb%P
zb&S6GmS$!?C{-(@!cWx~AaUvA_VKvr`VBjcMq+b>f~Z%;Zml5dBlFg*gJi7f=#Zx!
zEu%71OcmPrI9<?8_9S(Rgo^M&DTd;0*Mb|iBW*rh30aXp%5i~>MdQ#<Jslk`E0x-Y
z$g*hiur+7mx60@pBjwDG6x<IrU3~LLKI(h*_SBt2c{}dhz1s!6#yc~A$in#CzWWLY
z_{InOz2qaEgqnL^$&d86zt=U77AQW0Db_(9>eyvb2u-2#j5ShB!mxZX1qzlo0%e8N
zA>5m?Yrm>0HD!2sxR=`06cHXC{sOd<TD4P$No?FMf!NrcYA&Nc606OJn0p_~&w{Wg
zp_BIUV`L)$-)nqosuc9%PEZd5&<W7w?b|E2R@A3fQ%lE7E9XTc9QwVosBZ_j`1Yks
zjJEM~`rIwp+7s@oYveN*DwoSM5qFh#90hi<$qA6SS<#Z#UvD;6&l_O(;(e*i>)ju=
zZr$pM*Dr5oW;Tvso;=e(2HxJf;`e>?gZpbRtNPY6rXz<3SB?dgTYruGes!yAS*Bv#
zC}Q~Qt+p%ICF;d21oTC>Fcp7M-J;y3Eo!gNHhqh>o8M%!pPh35sXq=bUuSo`tm#`)
z)3<7=+Hm_fZ%W$<Cc_p`QLR8_EXItzgiYhcrDAW*TssRHdb!BPR6UMRr8w@1<7I{~
zX{^ITF@WK3ZlZE7uml&lP33mUOU<4ko52f*(az|`O~sSFszIp(A&Kl~arjA|zi{D#
z3zqleh727Yo$C5};gq3o66%&YeQ}o_tW;5}?Hh>`)chSp=joc@0(ubA{J@|o#`|Ld
z>z2P(YQ;zGxz<Sq`KXqj0*3}W$q{`vK@H3VEJhAic1e^ENiL|nINPX^wb!Hb@fCOb
z9Ju=J+5%1Dg;u*3i{kxT6`v!>|09Vg%xaG@vO77R!WWhyk+YN9d;PcS{i9sJn#9Fp
z*if!5DY$y|sz3NjMXc>V?yNnxT!m#;e5<8RreA{HSJC>1(=<Z0n2AoP3ycN^=NX#&
zWP`dW3WXqO18P|U&wi=o9e7CHxa)&%Sorl8u?wBnas89qA3*wEDp+|$aFPlJwbvmK
z4zWA$eNWeJd&+&)mFAY(=C6j>ZGpx%!6pQ!QKWUnVXN_PJ;GJH*}m#Dj@$nhmznwd
z3aS@L<qxpqTD{qt;1cppFQz)COAd7ozN}2SD64WbLZd-=f&FRO<#j#s`?a^;E2G!7
zKK!ydq#!4pmgw_J)irdM@FD|4LvFVpw>XN(F`yW{;z4!0Zpf>H=f#DU-B<OQdsT)$
zLqvcX`#IB6&jmcF0MEC5nwW+u*VGTV%3c@?u9RmxEbt<Kc_nYM>T?`{h<JwR{mBv>
z1*cv}cdogy#7u9lS@in16RjW%<HU%IH8G>gxmO1%`&&jz74|NDl<m)8sQ?ni&Q|M6
zc4sLFzOwMsUQIeBz3PVfQ<FZNoE3-Tdyl_s*mKUGx-lvb$KIw*n@Zomf1fo4`b5%2
z*`=G(C@H?Rbcw&#!%F#mTJ+U*Y7P^1-dZY{HWrr77qrpMf}lsAQA|eGzr$^)(Ji^g
z=uJAkML><fRXbC?`VZ9|XOff_qR){4?!CO^W=p5q$DMd28y9l4!wb)$^(d{%C+}DC
zSC|ipXQY~T_l`VMk8?Ydgu6O~keDo36tlfi6Dx(89mmOs`x0!<AJ{M35|i%xIJ{XS
z<NCOL-CotgXX>_@er>3tu@OWNJjjxA$O5vyKQiOw?9^MF(Bsu+e!8c5TdYy*U|m6C
zcWhxyLsN8wmbyJxRpF0yg}<QW<35i=+5iZko)>ykN@6vs|8bnkCksya;L>HYK6>+)
z@;Vt(Jr5O(`i%lLI|KoP@Kswe>HMmWGXqP{s^dX488NA;{Gh=wS-Nzwc0tn|fD$vO
z+sKp~f5CVxzk1FQk7&G|O@NVTS8@7t@aKsrIzSX*(V<b^&nF$=)OS*kWyGaN9Dp6W
zRd3ak>0-V4;%U6Zvwdk_(iPKqOqma9WyG7!Aas27ZR~b-$E%Sf9n5EClydTVhll+E
z@u}y~i^c{HdTEvTog3=Ya-0=^<|ZL=n{H}LQ$gx0AsLT5T4w?!&Xi|WCraonJ6$>H
znt{x3@7ci9^+>wo)rJ#{PhU-Tr4drQ7ig2{KejUD?9R9dBt&m#z05n{Icf5Yb7#{Q
za<OV+QQb@PSHBxY4$YZrw1vy8&M;3O8|>So<ZNQe|7)dSzuL*4CwZxvG`qbOPe~w(
zd;7|@4heUnl7jy`yRbvApR(~(D22eyTWFtZWyl(&;wGB@TZ{oUP%g^#cYmvvlQ<m~
zDhQgu36Y$ImfnqwF)f9%tE1|q2!bKN6HxOWmu?RYlD}ez{PMStWijy@{MJRGCjc%6
zHc^M__FBn^c7yLmDKlTRGyU$_cnBa{xUr|S*JrB5Uq6T9=a{YR``#OJK6<<^m4Qv@
zfP7xTnktpbBgdDsyz#oV;*T_W?M9XmF5M2i0^^7sG=3M;-rPLHz_N7lVXt5M&pXfg
za0&2a($@LZZdj`ZUgKkikbwsP9hWZuVsndT<%Zn?xPkM`5j&c{(LT#6Te(h-Stx2b
zz#J!yx5l0^19o!p+x$}+nHO4U?iZWft*uq<^}PIP?>=trtlxi$usuN^N@sGh#Qf(n
zHNWap3G#BBx(-i%m|B}E!fpNX$6Xv0A<ktiAqQ8jpj3G-|Kmz&uFX+5n5|VP6u<X;
zY_FNE$bXd(Pny3WufM%s^E?D*F9%MYzPR%(e)l^4@QnwnF(Mb6!(PMR4!^R%uSoSK
zM1<$}%_w%M{{DgOA}9V^pK0U8&F4aCl|HR`zV*h+UHErO+%9R_Et*CfDXLqdZmfAO
zu*xH0P1Wn=xN1XL|CKlekCnSZ`u2{wNFpELpI6$YIxW^c#uFLf(7Pj2M|y>CO2jrc
z->uA!Ruk#|o(ik0UO)ZYx;f2rM9*HQC`h^B^W?W6I=EJjOS;QR<@&RaHLXzdsISFy
z-H7h(+w*Hj5Sk?6mo9e87>p)_053=FK0U!K7+%Q{snZS7`acE3sI&QxRZ2+_pcQru
z0hKP_1p%~vc>3m?^5bDwB+-9;K62&UU}6?du+M0H){5E$iF-c-4keQ4x3*xCo%#!F
z9dWF5-m9|bz5;D>gh<=m$2C>bXT~SxK*@UHipH6)RG}(k4fW6Y&h`ZH1!L3z4%!3;
z{EQZavE?^ET!|9QITb2yjgmV(@E|>^QPKPi|Eo`t9HZHlr$)w%$6uONgyk#_Cayhx
z{P>HDZ*IPrvaql?C?q8GH3xYB4`24TK!U2WLG2y3^Vm#mN<_+;xR2r4-xZlCbphs+
zA0;jEda2a6#!L_Fiy1VEKuqW)VLleEAnf{@v%nkT(GD)}<_s>El_`RJAqF~~j}NYk
zfxK#^m#DOJ1c9sMO!UZM>$R>2;(;X!MIp4b;`;DskZINUQR=JJTj%z%i0Rj>BoFkB
z^kc_DwCU+d5)ue47z!^|oN#w9oI7i@*HQV-owe2x%3&qdHD?|Ad-Q_(4&J7XT9*7T
z;aAjxpiN+%b3&p+fYs~4Ywp7ubBd1{`^1K4+1ZC*nKx|){nQHsMsz@z{zP3NhnGcR
zL_~7bOSuSpcML8C^(rq5wR&D15fyG@vF?=WHTaUmq$J*)oSc#j(hc1}4;$`bWEI-O
zX_UkW^!fOY>lqirN;xV^Z*Ai9rc&b^e!P(#o$f-+u$-b^EMoUMg#2=<u!9}Qmw&v|
zaQ2&N>eM=(k2(%dHXL8J9V_wLP5Eto%<C#+;1j_DcL>!d-jt|WEQk;=HiQz`DJ=IW
z8+;EBJfD1O=aKl}-wy6i7b=gAlFN#HoMRFajI=MPs-E1Z+j;??zq~y%pm*4RD_LQC
z^6c!_RntQk;+y^Bt+|cecMLqEoXoBYF*AMqv~T-zQ*=cwsvaFU+}T@wOq~9Pj;Mzf
zU)}-U1GE`K)q5H&l<GHn5iD>b*45jVRW;){wjjzgeo{f{YP6O)dGZOnH!1h@pA=!e
zS1!Bn&(JEHd>8z<q!-FAuehv8b|jSv^8qE{gESi`Y8llQ;qvxztZfa4-)<LtppWv(
zGt*bLam}b6Nvpq17OW@!O^J^z!V^fJHkEpOG6VX2kHpItj1!5=lUhDdrq?$-yGqxk
z%5JEfe9v{yVi3xAqktoqU*nHeWewgdT~EHQD@}fz;C5-{OAr}UH#IS{v$I>FRfdw&
zC#HfQw>441+IQjAzV?|4`-j=bC{eq~=R#;*1OHn7AgNS0G%&(?#23YYlA73)r76rv
zRzT1zrQLMo7Zk#YfkX6723m?>LyNfEMw?$Hev0nrs^V$-m&?8-%ZwD+$lmht@gaUr
zPQI)!Hx<zXFKmzVw}Z5(8Q!^b=hB;-D-1?jnwu|~n3%K*gU@u0<P)PLIdtpRaN7yy
z-g%TMJudC(+vZrur7C*)%WWoIPblY5hYUbrq!@Pn+t6rMg`fppDdpYQ?_q^=c7oGe
z-O-_U!!+~*$}O!yd%qP>REAJfO@H;O1iFLLb`CL{-6&g@>eS-7;e~t!%xZDniA0TJ
zStcENT3Xr{h<>daZSC#lrz$4{2me^%5AxjNQVR<^BodAducOpg(EA7d>v{?n^atKl
zHr5?;r0PkYY@cpk77e<?sa(awKYIMw0Q@&}p?jp6W~e}Fz%P)MeNMlk3Y<xTlar)p
zNP_|hpumb4KMVbH4WLD4NI^@U>j!QdZ2q^T3RJ9Zb7DbdMW9SFlH4a7{W7B7lKK2!
zsX<aG|K)P!KVK?+zI%+$hf!8Ym_c`JMiIuBS03@|P^v_K&Uj*5jFDeAvTD`35iZWO
zrVsTiD0^^AMzytLmk5k(@`RmYdYiK0{->CLxC~lpVI1*kS1gsyR(Y6mlj~Yot7`-b
zl4E15Hc~QhYJb@E*zbKeCnKNld&9M<*{H!#>KH+cG<xu9?MF(T2yuh>zfqH=MD_pU
zQ^|0Vnk<D9A6O}GV=vDpq2(YYs_DQ%ckk1=6h4+<&^4kGC&u%(q7)^eWpWoKf2`rj
z-k2*xDyB;do8)*;>JaGRx=Fe8bYDxeD|^z9?_Aml$}e+W&-T2%j%%sdX1r|20|PiQ
zMfDrKq!Fxt_m4pi<-b>6{^#-fuf3oz%m1k;tztn&RXXE+4_8hum;bi?c^qu}@&Y{H
zU*<)>yL!|BURRF)+w-sTI75SfII>if_CuI79lBJMW(lHf_#ycBMd<^I<d^0moWVYt
zSqC_S_T+uP&iD(>fe2V1p8k!DYS<&fh50|8pA>1GvK}s^VkV}0^S+NgNbz*41>{Ph
zc<)>~K17|;f4E)y{f8tc=i_8;ojN8Gl(+TvIUYO`Am)@_wXv93fZ29c%hxvRlbAm0
zYI<gT#ycf4aBaOqekNC#I~q_`y?cr5+d20>$$h&{w~k59P~di>b_LBb?{Ypiaueiq
zDF5#a<v2ICK4u_Tupy>{rn&sr5vZoxte{&XI49Bej)qpc#L=i@Ls@g0^GF#qegwTk
z{8wNCv=4r~B@mDh&2CppE>g-%*lrjS2_xkojcXu6JbC;Ndy*(9OIu)&8Gv%QI%%7j
z_$$RV5MUv|>o=e0UcGv?m4)TewKj<<eb*CS6v~=Q3NlC4TDJsiUsrg3A(L+-tOF`w
zxAkm0*`2+4Q#=WRa^5(cVF{W9PuvQ6cI7NSgk69PJv||rBSA~{$!>bL>319@Q0nXT
z35Y6WbOfUtrCQFfi6TNmPg9eQ)l)mvxe_Oxqf^B*J|fjW5$p487XhG`_$>U3Kl$nC
zUckl^R1oCAT?(lj+qU_WHq{X@;ZmixgYQ<<SV*1dWGrX4Qcet-JKNG)*#E6SjbrNR
ztI)Srf#e+lOdISdlmhtANJ`hpjIA08c#QxF`!Jjy&i+A@Q58M?_tX-+ZEXOif9=`N
zHD7Qxr0B3lP8abInn^WSC_Ul!x~&+q2YjrcrqD(400Lvcrrp&H3jFsMsfW;ONF>7`
z2V5ARv#iBk5o$bBv9Ge->&sjdB!?9&NYY5zOi~)G`rgemqXF)0mX75YI)zgmPvv#*
zxAX!#TiJ0)8B`!L;sjdfH@NNBWh4Jm3TU&{C*ejJjIp*;U+k;lJ-2Gxg3b^S+E9R*
zw>~MPAjPS7ATfxVBDtn&CnO1(MeyY^0Mrby+V#Vt3H6&Ym75QEtH8NNQMrK2q%VsB
z=!0+_X!s0hGs{d8%6Fop?<_?Bhc1G`Zvv#mkofALQcr}8%n{frjN+z`@WIz{ty-B)
z-$3H6f)p`?&A}8MP7s-Vz|a%n<RsDg^j-Jqy9%juC5H;=0;H%8c@xQ&`;8qp7>LQg
zjC4qazJv+%_q8W!6^+HFxTAl0B{}_OQ^`r)pMn4Wo0mRM@D5K;QjQ1!liua6FciT7
zQ-zBQqfG9jOx=8``bs?}lUq`+0P;g6pd?6H@_d6BYnjb(8Nt=Sytp`z%7;^SIoCbh
z^dGOvt*K;qeI5kE{v7jPR&8l^c6Kh!uQU|6S7wM3E<`1SIt0x2!E?E1OMg&m{+(2M
zp%X)O<IyX<Bc{O=*Odk6UZOmczm7le09-*{I27~h(!uCK6TGMuTT&d+yRD0iWS77A
zuqO&aMPSNyh5-=lW;<5%&%p~%eq3_$e(>pUJHU4C62gOSz6|}a;fSSoPxxDK&S0RU
zlfuio1X;`rl=zm`Uir>(+YD(vjqq`MTk~H7r(>jMW9}i=3qXa!*BnyXZ_sQV>7Iz3
zYweLk$wbq?hUEiED}N?x<eP%xzm3@ubay$Dmfa(WPge@^k@EK=^A#iHD}2oXl734}
zN=cMSz!*+_lPR1{Dir%k#>43OZ}hV{{ZXk&r)P6#jW*hNx%3|A@J7DEuRHdru_ze*
zU8bw<N}|^+*!YcI5+jbvx0B4+c(#2i_>*!{>VFx<IA<A|`AN$~yCXB-f|j;PSkaw%
zvH5X@#1lJ=*I#pu^g!Hx<3Bw|l-{AFaS3~>fja~}+h^zIxDZZZPbf*5^Xr)QBf+Mp
zClWkakec#_ZBv)FR}_Z!s!6;a?;H6tzT70_4lsXlFp~r)8>3}`SVku&TfDU@;rY-r
zHVWqvaR6^q#ZdZa#+fd=j3OH{%y-O7Bs3)Pet7(m`;9twO-4w1)#4H<-T-3xH%IBS
zfUO()qBPVJ(!5gF)cANF$@#CFB7gSS<ABi6bu`*@a{2DH3+!RK*0Jk6v<}Rkuoi&3
zjUx(D00l#TAcbHB9Or|%6tZP?9g41D5p*WIZzjkY{hV<*^7B2H)lf!b2^8IW@aytE
z>As3ah`t!r=LS)i>1*?i?hB2-ju;7qPQR3zdpX^&QOJV*ElJvzQXErAyNty<4tInz
zK(ueF0@@4*K*byoE0~2Q(d?F7VE5w_gihjYA18$p1#8>?F=6h15)Ay`i4Ok<(a8V5
kf&c#o{`YU-=HjN$7uH>0oNcNGADf~eds-&t*u|Uw3pPyB=l}o!

diff --git a/examples_skyline/accuracy_vs_fnr-skyline_formula.png b/examples_skyline/accuracy_vs_fnr-skyline_formula.png
deleted file mode 100644
index 3738ce64a7a6acf22a568d500d7250fbe815d2d7..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 43356
zcmeFZbyQVr_cpu`0SmBDX@d|@Q0YcdbkkiDD$+>jCQU3rHzi0Z-AXqo3Ift4NQ$&H
zTN=K3bIy65^FGfnzVDCk9q(_9_l$84N8D@eweFbnx~_T6`-!sRRmweddoT>6#9h0r
zieWpNFpMl_7diaKrNyxoe(ZF(gj3%I|9I@W{TTkd+xD7{1BUHAi2fyM>GfKNUkW*1
z(RNg`F>!RcX>W|(x#?(YW#edNe(SiivAu)2jWsX#1@7}_j+;3;+KTY-{PXX*ZR|~X
zV&_G=G3+>oyL?gIHR4yVyS9dL-_~5C%>FC;FDnccEIg3i!SpUiUZMVxiriJPm&OWn
z%ksX-<F}h%F`nV7OZX*t#YkZMNW{FAMLaG{=c`b6EFqrt_}D95&TC)BH#a^MSaK!p
zy<0!HQN+!oy;QcOYxuh=leyD?;`RBeUGVf3WOKGIB=A2{o|k;ktDiY8MG8NT?2kde
z-Mb401@j@>X$C(kcI@DTA2&!y=-|gYAB+-yXiNPcahWn{$l2=&O-;%xtE(2@^X~BH
zMdsx2J$&#$+pF(6Ew5v%f*_8Nm7P62Ihoqh(lWu_INtO|x@aWa^5^Ji;@V7kc~zdn
zi3_~E#@Su!s;WOaM6fo)sA+BtqkBui<SOMb+R#oZ*fTJ#+Q4<@jQq;V3V)H`&!4w*
z&0A&I6(e5Mo1`}1A(2urqUQ4vsXi8)m36@-)J`QUE9)>Vtxo)?z$fuo9UYyCzEbz#
zj}Fov;#ivv3~F#Vzh#Js^RLsAl9Em{efo7FXB{MdkG!@?xt?#`P`$Y^<1I({J=33g
z-l!_+&i6cv-lE*;qDgLChVQ;>rrysb%$km~va*h+q-GET0s_jmHr?x^CHo8gs2D}Y
z`~>@KlWjgfrFr_~i3;4g{k_rE^y~SGGi7TT_@(Ko3p4Sd1|p+T13cyHs#D+dU4$Rp
z?8-8J`t&J*B|b4RRViA+qAM%DJJ&K%a${U6IAgLcS*@Y+#M{_dbSoU8*ShmpS2hg|
z4Xd}ew;!{l!NZ3S30}hch6+cc)Yg7aIZdYLN9FeMTXrP5FHO%cEU4MqzB6fz(jWY7
z=r&A)Uih`Z9o1szDLt=*c)3tc5s}rjQT0WG^_52N4h=1RRu~q2mtH3k`=5^;Hjf&5
zyxH7X<2rs^s=2vYR!%N;k@zBxLoH!!#(S$c%6rodP5#CzQN(E?Xu5RCfQ*cc<Miok
zOMUK{ijg83<C8Fo`#x7!PbR#!CPYU^;}^%?*pzQ*HcG6eaToqP`0&xAiArioN9$JW
zd`q}b!!WP+me-U=xhZCQgcJ*2u9n=qY46~mU~Zm_Q&8ZSkdPRuxJx?1ytVARd(WN=
z<KyEAX=z_!Osa32n1mAu2M^KE7^5c%jE=VP+FUWsHg9D!dVW4Ya<c$;?V4?ddw-dy
z41t52TcGL~ukC=;Av!t>=jk5vr_Y`#sHnWuDaGe@6}x0|aB=aPo0})NEsU5<cBEGw
zGb&r3mut;3mRlI~We#s?Q4I_XG%DL%wN%sAHq*_uFuri%Ld18N_wzHQOMSxL9k3o=
zQA@5j6^c3k(q8zL-AHHjyl}pjt6p@b1RjWuho{l1D{HtUGm<6Sw2^7l-qp3>iJu>T
zVPWCOSaWQGWoO1nruRnb`Knib^n&4R3gNGOnebl@A2||!@7}%aj?OYq_l5DqXt?dl
zTs3oKW@VKhy#h=(T)-aX`f>4BSL^z0B{e*Hk<+9}N1D!%UNtOGGguCHp4)j!a^iVJ
z9LH{yt&GsCDJ$O~r(!T;@>;$L&n7G^%y#ZvQ_0rmT0%}vo5%8?FZy#~eQl3@-B~sc
zj+(;tQOT_?J-g2|Be1DQe3`w@!R+w{i2v4(&#WE_khCr-Ev<*|mAA8dA13CO_5S<R
z2t2I7l`B`+goV4lO*w~luEhpO*mFluj*55A&|>Nj#G5cq8fNCqn3&^kuC8Y5%X1Sw
zh1o5wt;GED4FziP#TU1Xj38RTo?w6I>&qoBu1`V5(6pDQB<9Fj-6WlClcLp$R32A%
zwFG&U+qdJy#Knh)hvWM<f9Fqpdmn2h@EYDRX)>c&5%vIqrT(>`Qhu-V>qnIIY}C}$
zWE2#sFy`~We|Id;4dvPnUc$9x=obftg(V+2B@^NB{SdpNx~%L|77iXBCCiSq=1)%!
zvEmF247y>t_%jHfG&D5ScN-4Ya1(ZF6kj?G+u-iqyL7_#ad|fV(;i<dE2Wjh<{rSf
zw3oOoWJgWHqdkAVFH*?vQo3%gYNkPHL;$VPB5Zsdv)A$)0qdTmygT1z3D$+fM@A=S
zVN56PQ%f|(;R3mnI<DK4YUmd^sw|Au^IP`_nWpbO%B`-a7xClAkDr5sZ*EpTN`_dX
zn{DD3-}!OS)@*fgyyz!Et}XA*b@9#RA>DU3J_s8IkdTs_Ezb_lFD|Nkc$82vOQe~%
zCe+l`$sJ;oS2i(8@Mjb?Nv6V$u8WiUJcab6J`w5Cry?)^e8-L*hxhM~i4=92`10~p
zce!`@Lem4o@9^_uW##a=PVPdWw6wH-NN0wADpzKSDh$M)UAvkf^hWp|RM)((&ctx^
ztS*bu^CM@pW9){iYQBG8i})VJXHE#=)=yDOR5TxNiJxds4TamW2Ky3G$j~o#*08cl
z)66u`kB`puT>ITH<LTj{r`TipJue;Z%!b2{gqi-FoaBOuP=Me=$hlo}FkL5GW^izj
zPRKSU&#LQ($Mk!n>Y#{-w5v>Rmp3;z8|GeH7vWMh(_3l+*@*$(>(Zp;`<fs~<5hA^
zjE%>8UHY9i6?g93IhNfh(fZ*&#o5v$u;y&b$_R&<m@-HysMMG75gi%&qN7W&p0r?L
zwZ-8Mjb_DhM(%UV=;y_fc9GCYJ%p^}Sk35SHr-Pwf?rI6fla96Dc7qyh5}&mOquAU
z@S8VpY8x6dX5%u^VvdQaC|r-;S|3dlE|>W53zqz9?~GR>N{qf#j4kjMhmRg*H+t@H
zou>UbPRqb8K3w2Vio$CF^X9i#!o}R?-MsszdyBcm#PncMx4*seOz(Fa#5o-f4vtlb
zV2cnLEPs7_FXA@;)2-9|bEfz2tIVG2kY#Hd8#^a6%eiq+{ruXW(r_(8d>*T!w`nV|
z8;V%Co@)``o~ktoE3kc}{&nQghYuenIy0luypioaqSrKOD*<l;Pml#KIkN6B{}TsK
zRo{Y#pcNDxJgnEy(4eWT{K8@7R%V;siE^8k74JqY<|L$M8hDsE*rf}$J}V2OmZJ@k
zgt+fE<@yjFJ37*@8$WMGxjZ;HzRYuNqD79UJ;%J&5$;cO<cJ8VJ}gU#jYTzMQ`1S<
zzxdx>rj3OgZ{MDb6mgP;JRY2wm<aPUX;1hDxeMiNLwW%#WgVR`NH4kWOV*=9g8d%Z
zDAl{Ux#`c|WC>(bXffQJKkB(Up7=B%z{Q_OuRz(CnmNSR*B3SbOEB{-<?E2Tf7%K;
zt}olcRmsM6A*Oe{n_@273o+2~nmmWt6}f#|lq#pby_YtsdR{oDt2ooXIdGp^T~{~d
z<Ht+!@$nq&?A0xJHpPgf#Aq*F_>PX3r>-vk3{=v@6AU#zefo5opI_ti=g+I1<(u&+
zxqfsTUHfZOqD)@|p}IZOP%_uDQ&H6UR};$Q$?M<7;^J5o!ufDsUS36WRrEv%A}P2)
zHb_^&zo&as-@Q8rc+G6?$7ezugbZ_txbQxzGkv9v?($iXt_=V)K=M(90M!CZh|i+!
z^ym<5z4rB$1*d+`rJDYJL)Zh3U9rM$^S7sZ3R_iAzZSB4E3xwPkiPp=u0?BtpMytl
zk<-)X&x0~EGdCfhbMo`&ejEz2Y`I6a_jdmGc1WqEo2!#BVy5-90#>j3@RMi#7=&Bg
zG=iFAuX1s5$;Dm`2nh&iywZ7qjxL3UOLHvDG)e^!Ck_1ZsDbN;d1DG@_h09U<E>X!
zR8%IKV}0?&iTG+J_g`-(MW>Pzqgq`0Jrgz4b;lu3W-fl4S=fjok-B2_s4*Gilakz8
z&qh#WW37Kn6V_a>zW;TnU*A-0O6DFtqGFtwR>fj2-T;tEK915cq#$}vSA(qEPhh2H
zl>XLr3EhEFlIPEpHw{2mqczI$Sg{YC`|{;lXr-8<`;ebt(2Ey0%;~9orMDI&w;VcL
zA<sKcbsMJ-&|Ls5X?**(*@t^%x%OZ4GBa%+4x!!ZLqcldv4$yM7kYr0!8;#mA^7n>
zhZ_8k;4lC8A^&Ic&Ohd@?K*(GW}^V{1jsHNCr^It9magFZ;S|*H}`LCy10a@uy%EI
z%@c_lz#NL@NVp*l6Z>FeE!GS67Zg9I%eNbnp=1)%1`v?-eQGqa{AMM&(^R(xFbp~|
z*LRSxEy}K8*4fTey1f8;kLLQ*@|uhRjY&y(%4wSaJ=bdkKWeY7{R&7{JAg8}egS31
zT}@_9va+&0`j>*$)m@$)IqUfSPX74AqXt4V3jqwmZ-EVsp?CyrBuT<^6+v8n6;2;r
zRl_yfru_$ERBWui=P8upiHIvz1u$nqdcdjb>V{vsbSYxtmveEH`?N4X=;SNUj&O2t
zH2)hvb7)}ZGkXD%?>up)->|H#4B=0ICh?R94|e8Sb*a98|GxI?SI35}^@T<ndU`Hk
zk)NxoCKehcLs`xmlxTadF4`r#HAD#0@mmD3j05bz58R~y&XE8i%Y33Od39wpTHou@
zXq3Am@Yl}t{62g~V}HJlp+c0H4x@;p(2QlS<GAV-85vFq34{7@{#MWe7TdK9&6lQo
zkN&<znslh)mua-;?=E5YE<omjqw|1e<E?v(k~LB_=_EV~>JpWKp0>QZS$X)_u_XA~
zaR~R#KqVHx@;0Jeg(lU}aZ%V()ONr}#*Z$(qC(1Pq77$Lwvw>$9%ji5a1EWLSIPSN
zdRwX%k0-HRix3C+j!RVZ`l+k_C=sE2Y;0muYmzOz6=+-L6aYyMrGK~t)PrR2jY-3!
zL=n;u+88DwptigB?&akvT_kXDa$Z1Oer|D*4~N5<3|0BD2?>1@78DcfxqS8N>tDaD
z5S6mGx0k<hBgSoOV+j%06~I~w@aGl4UC+-~?PCyXa{TqJ`B!goCxfW-SI5bYddKOW
zE|10Ls|f`K-{hmjy5z$yd_HFM6%wF=l$4L#Xrv1wQaU+iFL&?Sb?&M^!+Cb4C_b3q
z5dcU|9W9AU%z*^Lt>@=Q`rK#u;U)R51~Bo1Omgm&%=2@t2@1Sb$FgC?w*oj_Xp*K5
zgM|f8syuEv*L;;ZY<70mq0C^Z#~~n0#3`jt8_1Iz&{*#ED-;9H<<q7|Tk85e78HOQ
z)`jv2GKxC$(bCeQXB8v1sU;YedjZK*j1aonoo{1<=p`cF4nIEaUif^}@R-qaxAnPd
zg-e${lC=xmsK6G=x7Nx>Mx(t2Acynb>CQP1K+HCyy0x{nWNo@QAt~ug$?V6yRo3~h
z0ltkk65$DjJ9BONB~H}<=rwFADg|x{0+J1EwXc8yztU*w*}doCzp>jrMuo>??(<Pj
zZtf0X;aogCJbsL#ny~7F8_HqL9tOk!sMT`t<3r&4YCxF8#KiclzMVtZ0XUsqLpibI
zI+9ve*48Ek%UAqBDN1@H?OWU0sx~(_*PVCgHOu(MQoj}e1%7}$Zg{-Kk55k~x^sa;
z=&~|1Gxyq7Qp)0RFIlF#vac4q%ycBF#NK|cVsD={@--|LmMTE(ad;yQ3yYL&a^Rif
zbDs_Y>C^-&$1_0$^c`AWUS9VK{xd&6&uvkPS@e&4IiD!~mD6eHtsfvZ{}}cHo05`}
z`07|}cbR7~gsL1o(E<j@V5#4`98n=8;s7@fu|U*Jz@{$^?N`{0Y5<b0%Qu`$7aBwx
zgiGC*I=;O(LL3>XO`K=Wv_U{oIXhVfG@9n<(L~T3%;taAG?adqx^qG*)<6h0+`aqv
zfr3XE0jUaZpr)jB@`?sT4ggDwgoLC)5_WbN6_eOlMC)IiNYYYRQnFvCI_x_oHMN$t
zbq0%hOB}8ND>Z9A+3PywH$OMmJXq<gp{yJO@Ltri0KlNr^6bq~JdAd#Nqsnu5al*P
z3qn(EXQvi$!_wKK6XhPBo=&j9Lf-=`|I*mlIA4SnbpAO`X=!IR<^sGO8}{zo7j#|*
z1Qj$4Fq5t-{ER}j6@|P5IfhFg4Li>U&m7;idw26)Qi>GgUZlr)%?u62!rt&da5eoT
zym7%|i(K&Q*IYof&WMT@x2%*?b^k~7#-8!w#fzr;^(P_303+Il)8n&)m1urcRaN=S
zTUc<ah#Z!<IZSou*6CEv4OK^eSpOQ%?765iJUYrHDQTElzF9OZa_ZD6{oh|A#N1Q?
zU`L)jb?R*9ULW_!X-ZAJj+t3fpX(51Xdwcvt4lMPAisJ-lo{>mSlLNNHeSu_?YN>o
z*;%U0*&pbki^z?={o%uh9s8#Az7&zToPt*x1GWar2(gwcGpV3JTP~Da1CL)N6JLVd
z){>%;y7+;DIVmSc9h5~7{L~Z`{f9+BR#=?M@3#PMiPFxG&(DHizI+3+$TdPqNy%Z*
zY$%vLlqENpW09DW>@icv=+g5BHnWoC`s{rc4naXJW{=qqWH!ZVY4UJm*q<TiWzhd{
za;Df$gWQ$tFsgV~C+jl&4nJSR4hm;Hh%dUEVI*zv;z;r#f`n!Pf&w9Ks5)R0h<PxJ
z-?8(TD5#m*fJ&#q!a%}|q1Q@$+`Q5d=*k+Jn$8Wnm!zfJ%e}qD=f9jn3w&#%f6KhD
z#Em<&;(L)(Dx$3*8kVll*CF92Sj%9%N74Ai=A{nG0(w9UmmqYad{(;g`}>}K``X~m
z`AxpQ+Kab_snAT*eq$QtCIjqJ1O8~>I{0v5lT_s;a8*m<L^~b=QG1b-r8p#8U62*K
zbIgY#cwr@^8<rbDFag=c(O(jRE~M4KjFNd-l%S=3sd61RCX^&M)RCS55eg)95R4p<
zDBDmCi(}YU2&gAj7Kj_5L!+#@I$y`5@1IYuyrn&E3}mST7<=08Lu^SvM~mRiask8O
z5TcM+9Xkb7P8L==3ShaA5OAuX{KU;0bf#)$I&?q|dHeONJgB(li(}3BA5Yl%9LJP_
z8TZ)UKf+^>1~JYY$ck=}V=~M;3l2r2D6ds>vYk84e*fwU>7*=pUBAA{YlmUVJo*mx
z{6#QsP7C#d#l?r~U!9i*rrQ4T;oe^TJ<6xPky2t(;##onXUew>LiL@)0H)EA?QkT6
z@k@OCco)j)*(UWOlXI}4DQPcU269KoXQmwNA5Vt4#CG!ULyHSiPA2T%iHX$Aj1H&h
zGP<TPZ-G1A>gnG}Cg`aKg`^<?7dei<4Uky9cH+bd5H&PF#4B={nT|LF>lW7VSjctj
zX4pAtNOXX=vrePK(%y8ENVUJGSltH-GnyYzN@D<OT6}!4Tlf8~hjkGk-64ichyzh5
znka;|IW3A{Ng_KIX+clW=iX#Gc<_h)oHOpJzkf$dyc{bI$^mKy22lXUbW@#nRD6y>
z-{2PIy{CcC{kK)GQ>Q+DZ7ScXp1<?7kdQX8c412c$Qp#eKvo<8B2{f|?K`1Q9Zqv`
z1^QBK`KHNDRJz+Qj;SyT+kfgcZrg*`uT){)wB{FZ+7LIFFU{cG)nvMhipsgrT~$%B
zLB~aM&z?Q@)4??$?+{q!K)$o>C^~rPQ1Gi)r*TgoKYr&}Z`f7>6&y-BzAwGT16pu<
z1S?!ZmCuej`N?@M7q@?FBQC|gSu6@yb=jI6VuQ5WooyQMd+K{4B<lEt1dFyL6>bO6
zLVuLlZh&Bx`rW#?o-uxhRB-bo7Bc}4G`qOi)>G)f@3Cy_G})m5_(=`0MuJjwMlhF_
zEa#S?(dEkz&~(8*0Hvy?UW3$!e|uoLjkM&JwX>&_Cz+%9Wr%~~<E@DSo5W1-!s6#5
z&kVE#+kMNg`zh#qD~&mgR-JG1Ap^anwx&kq=FK<7a^^4AH#UgGPQ&8jeK=KERfi8A
zJPGIk$a|XzUcWw-Ys23_i+`tC#l46K21wC0*7<cN*@eXs<m?btKtGx0maaO+D2`VL
zalD6)*6(0x#z^}^udA7<BK~*D&zwGUMjh$|D-`i}v%9RIXvFW{-(=Q2Nk@U#Z$A)2
z@@$zR-fqBC#;CggZq?nTLW^BZQsPG0U5*gFMMF&+qtXGmY8y0yXm<F}q1acij**d*
zPh{0`qm&IYgsxxv=E<{X6#&k?U_sHS2y|?2d}M2)Qs*jhlFQR$Wq!E!%a?+I6>NeE
z-^(-R<V0O9MM`|0pP!GS91=VMYbDR)Q4IznfE*z~KI|>5e*~A?HB8c7A?`M=EYA<u
z@!$Hi3ra;B_eMyQM8}-UV>;W#f88awcu{$xYl`4|bw$f}=tn2)jO0(BF4u?gj;2*Z
zymgw&?SyzPEB$DninDWG+`Kp6bm1ryA|gNxn%rkfwZ>N|DCvHdI5va41=<g5@C*)K
z`*}wBCf*6ONhI+v2qX0yUMbO_6wd)10w#kH2a6>clo2EeFZer;UT~S#)Aw2#M$wH3
z(NDBG8b85t<?`h*2m>7;%=5#R8f|o1H4PKE6Aqg_BPr>$_-qB~D)63M`(at}<$?Q9
zt7!(@GJ0VPbk;<aO~=N}h*RISK*<|Tr<VFW=G1}jmE$Rm<H{(wXP_1Eo`Npv1eI^l
ztJ5@xgLq7hgL&jsFMa%+zSbZ*qpq$VBxe*F2U_m+{2u#fNVP8oELjA|3n@al7KjtX
zF4JhlgMeJA2mh|H>Mzp=2pkAAbn@g$$E9AER46y$L2g9kGS{s664VhQ4DjQu;)_kv
zMMM6giy&a7LZ$<Dvo73O>NKf&7BoqihznI?Pq{LhpfskTs`{2yE@Uh<vrH8z15&~O
zRuzjaL6X(YvkG&b={x(nKEbHkp989Tu`+%swl-=|_(E*+^RuH|0s@*q8<YW!o-Jp>
z#+)?9ofbwLJ0U)v0bL!hFaDXyEanjl>&85FtvMOl@X*mzlx9{&UL!<N@Z6>=>k25R
z1QtXH5gJ^0Pj6GA!o<V`mDpk^imAs>Q9&>V-fhv57KY{q!n-HrN?5%}w~U9izX4TP
zEltf3c-Zth?qYc;G=$WfKt;LNvBAk3q+Po6w;t%`n4N$}@}uRI0c6zzxZbIfQV^Aa
zfReXDU7`^`Js$40wbpM5)dYZ<!lQYp*rksjkB2l6tiJiu&HT_`UI08EpCw6>v<Dwa
zG!Voun!%f1^&z+T;w`CCGGLdAN{Xmp1hlIpW~HyMuVJRqiBuI}I0(|X7M)N5D1wIb
znZE@zL2HC4#T_)e`8HZIGLPWVOoy))nnguL$to#z0_M&Iz^?DTLCh`kj5#0M+0KO5
z@4X0LcTM|Mu3hR5Qc_ga2_^3X<N|S3mcT&eOTO>e4YyL~VtuGn2gf-uNxF6eTuFw9
zvIN1F5Esa<Gy!B#10Yf`)U1G}DiVSS1p5Z!zGLU4Ajg(L0zh*4+3R`osLFxH8qh=X
zyLZa4#{i)absFm-I{-38;9uBb1T7*03;d;qYl$HQW&rj{-kS^2gjyRjWmQ%CzG#|?
zF4!gRy%03Uornx1ut9iDWlg<`MuyZR4D_eVx4ck(DqM$sX|cJ!0`fA_eHTN^*FVpT
zFuM+rY(NOtXp9oq@6Zo}de;QNOtgU!0QA~eMY{zp6Uc(Z@BN#a@Q^L6p?W{&OY-}@
zu0uX=kPOwce{%5P!8Xt!5a&UNqdnVH31khFj?fp{crDNq$C{;K$#GBC*44E>K5*(V
z+#lXf7qNO4bI9uW!G}lN;JQ&sYeHL&xf)VKAUfnj>75^D)0zcSKTz6$&wPEs_$fIx
zlYC}s<35Wkq1^gC`a2rl0g|d~toicg9iR8$C8Lj(e($?#%enQ;+yeb-pB!T6WN56b
z0}76+A+5zOHrWu*bdwX`E&)O8%gvfveF_OoXUYXqtCmVRtOsTW993J)g9pzf_cE&9
zy7d-{%j(+NN!P6lufP!4yWlXdhXz{WV@^a0C@)3k1DA3M#V~$IV&1m<{!7mY@e8O5
zG=Jn6#2JCEGz`Ce`_`gY6H0#E<?Hs_ij&OqV|LAx9vj3?A_{M-%Yc3sLHseC^}s0A
zq-UI<I*x!RkE_H4Q^rS<wo_nq=m0FhisLo;`T{f-CYW78EvLZ*#YlWZwcwrZH~@C3
z-dmeT`Ix0D+nU_<LD&Kq2@tbsZ+JkFuXvLbrTs74FPmBVOiHfkhd-mgMSoQghnk+X
z^dwaFl1na9)A3z{;I+NXU%WVO^qj>U@-C|Hfx#jjh~(tJhx-k45X!Gyy*ioOX&^e2
zr>3gf7JHQ%WehyjvLlBex~aj`p*7QMSIvlWDNFE{v=C$sTniK|-=fkPq_RKjpr|AP
zY3}Oz^XCD?2iC0vbG8_3dIQ)F%8{jd(|KeS^$b)s)}dlzwdEoJKqj0JnKN*o;)gg=
zg(9Yqvrte0rwZ9;GD2M1kdCZg@M)avn2M+i!#5BWK<0{vvbVldyGD13TXsglk3GGf
zOT7qcA{LASGY1FGrj!_BT}srfj}+DF)N}A-0#4<Su?A5ey;d%ygItSt*-m|z$PgaG
z3{)jUVq^K%h7IiD@ovjZq?_q9MuP(eCW_t0#s;JudEo6uQ+eOpfU1l}B~a1}2#vC{
zu$aM{(uw19_UzpYL`t)i=ZH^=olxAT%gHe%og7@L9y%$x#i?!i27*o#>X0!3t>FL@
zNt~1-<!3RMCh&?yYbxu-#8KjMkHmm8CH2LN{Ww(sGk_Fm{-_;We0z735C>KY6p?|R
zzd~hE%Zx%90v|{#>H5WaElyht(OYdX7w@CWHEvj8t5r!dn1zx-NEu4Xgt+BpS5E-V
zrs9$?S-W@dMnx_sD7i!}jDD%BQv<O<CV=TI4n>gQQ1A-L%C54a=Teq1kRu5<@a>jC
zG57bJmE4fVFr?8QX%$V)5&L_7kdhbHK>kRIi@RKP4Ae{ufbZOs!eHgd?QoJn(G(RM
z5t-sOZD@lBw{Tj+QgF_!;^|p)oNqhda1S1XS%0wNt3#K`S=^feyF-?snW2{qT-u=_
z;PK-oSgVm+NuX!}u+fL|xH2ex5^KMBafBhU>T@tbTG1Z!PXJB7g&~TMi!&RjxQlAC
z?9m=?qrFyf!2WcB1?n`O<>G1umkz({><w|ClOkU01&{$jR7ix~l|8byvckc_av7An
zP!?1SB{Q9#+cCg&Gk7tlwkq|pfp5_RjI!-Tn6yc4bzQab*Xp{fyuM%4I;MJHYpg;M
z@-QP~IwXxG5S>RO`XM%gwC0Xh91Z*hWb4uRp^?VuOz^d+!47E%v{n56?&f_YS$o4|
zgBR%x657E)G|^j}Z(T4*4pxgvB!Phy2Nt{*!rJ}){l@xG2`w5=c#ZZd@mH4`(j%eJ
z>IgB&5?oRZhb+DB06)VI2Wz454X->YxwWFiUxLTDyT67J0xUBF1C&-=`7N%MTL1j=
z5>gsJtVrG@8n8;(x$G7~4w<u(?59puwyZzg%5QqmxLYufR^X5Xd(WBm@`oY$I?N!3
z(HcER_8=sjf;dtdGYS@sw`_G;X@7)k>!i3%0K^KKYmw~+rboE>*u;8{y-r`ByoB#v
zX^RI&knL|2bC)it>C~GJ#bar5pclL$PIf}IRT;KsubWZ?iE<+=OI!Eu`%b${Gv;8&
zBsnK$n(o0UY2lAQj{W8QyYW(Z=)oaA`I1JnU(2FRLbDQDlq7WTVgTWQiQqJ4;$7aV
zF@SYFpD@28r5ScHof?b)`Hwt!@ZdB%`&B3-JG*dDs@}YLGt&GgsDBQ9UHi<e#+>*w
zGZsaGmVl#UFRT#9F5^%jQRG}eFT~w{JG#(H#R9EyTP$An`@e<9|4n{Nx@3R^P{+Ci
zJG3QDC%XYJ22$dE3TjQjMhFbSHY#`{izT^(KeTCAa7f4mq~$hn3L+&gDCjWMkqf@M
z0e3*i5fmwf<mBYlO6JXh1PhWF>l0F>Qm2BVqB6jhmkiw60+=A1{HqwKrWbsD-(@9Y
z_u~P}IQS-zNN_b7Qyz6KDh7+{R9$^N(zYCgA%Y-ntu6wU!60F_0(wdT%!%B8x%zBa
z8x9;CQV)^XX<U1_*L!QDJInYPn^II7fcM(UN@8Ulm^WG3+2sgctK;mj)eGO~M#y=2
zNzBMGR=TT}14)iKu!oAO1NL=f9k6Wpf9Eh>n?5n*pd_?`y8(<I$Zv$4OnM`%Y;2t8
z&#Mp)A2{&l;(hXVtGRb)&YtyKiuWHBPwp2Yr^D#peEX)2WOztFh~lWKsquq@AygQ4
zL^nhzmT~a8Xo5ZfDR$D|o+L(bdzgL#uPt;KwE$<-L`V8-h#84c%~yz)(C5<12x2K+
z{_zYCk`@Yn0PPT91#aTtFpLd&F?=M!{QmS|=~YE}%;QFvdW&5;QO%+F43@NjM6dsE
zE2?M*z#BCH-(azfgPk8qbOmw^2sAB9+#S(vRJM=~hP&EQfcONGtrJv>EI?!iKBWv&
z7eGh75wVTMRyslJ2te2gZEflsgVf%(YPoPx(w&`Sym4SX1y(uH^yWTt5rK4Q1|C*!
z2XHt*Rp$f_WfZXQ^r&#c6+J%J-J(x`q6<2B!!TrZ&z0d&ZU+?QkhLzeY&AtON-P;7
zUZ-;ydaOe`9BJYpebokWAe&0AaorKb4Mkn1gM57_sW$$7S*Sw+BYG(gCXxheqksq8
zNkL(`wYkwS4B`wVkK(Ukf;LfrCfRYwdjt^0>F1XN`t~)f+zWq`w7`dR-y+Xdn-CIG
z<*r&i8--yLr2cm$q`^i8YGo`KS(1~IuE9n@)y}#Iu*2N>_U<BK5p*IHtzj2AT`fd|
zWDL*X3Z)*fZ=n(;6krm-wZ?Jbf-2HqYJ<60aXe+qZ&8g5P^TqSgpig8b&?)EVYuFc
z)Yn%1I5)SVot+)Gk&?tp+`<|lOiO^Fx%x-o0~6Tq{~tZzW@rA1O&O*$c40pFn$})2
zEQ#!2#f*Dv6FaaNrd!l}O~-)ZAB<AsI(?d%0>eV~$K0(tF8aM2j3`NmN$AY>ef*f_
zM~eB#9*DV%A89(_n`iZ&>ukOB^}KIRV@+fc`-B`4ehv+ZARb5s9(UMT#4m58sXP}H
z-YQ1c5^g(AsE3rTu+6QX5IJ#CRx#vbB`YLROZyA#`*&c<V1JKcTG?D<2@Wod+NZ6l
z2_kX1Paw&mgK=Nwlhe}Hxdv%5AI|N&CacjzFKTt6C-Xv=&Hc_;jw`RNBEh7Hr{HU^
z`BL!?4-Vol#Q6FKh&T@1mVz1C3sm)LcdnfB<;z(mrIoB*Kc^To9zVXFYpLZ~%`56G
zblba&JAhpb{ivq0UZKP)GVQo*Vxr7Dx!>rLE@TS-h3ALaIZYTFRez++U-x2P9Eg*~
zFcz52<zRrm4YMA)7wzs61Z?YUFia7i#<o_%V6I(^lb!;LIS-lNcF#iQvISExjAhd?
z2!TTN3>SboXK1a=&doP}J)t)YgWL)+KjI*s<+R)FtdAR&#(@fhn}f1UVPAJYz_9f_
zT7%fY$v0Pb+=U@=0H+cxsDJMmn819cv%a%Z2h3;shnKiJr(5sCd>;7*?jjL%*oO53
z<$~DPT28}|xCavwp_!V3<d{_DCnmFsdIG_#COZ){uYk@OKN27MSq^#&689Jq!Mco}
z++D$bozZ?IQfG(M1lkV5i(4PjtkKWnqR*71C-u>-O0-I11zEM;q#?*dZzk$2bM*Y&
zYUz2{Qy#UCbfE*H=gj{y|Mrxg-P)=DxRimEvW;Z-?%Y>5EuD-o<!2@hj^gCHKj6D|
zp)KR{=&n+ivfOr9rK6HCmOlH>_c!xl5M{_Jw+gL3#DsnrQ3xJ{80||XoMZe<-`t$3
zJL_|XwC|WgQZ)Fi%x=J^ft4lQ`!OGAmgR5CVW+X;xAP!=_B|W6B;@$zjZGO&T}Q{w
zM#PCzuHQqg?{#->tl%53o3L3bSngc2e>B><=jq;{oV)b9G27^wawYVe+S)yf8%v99
zZIZAw@7BQNQ<B2u`*hn*sDBnu%Cd{$hE+i+|H}Kb;Jzd)2F&7qzRTbK;_KHFyu(Ao
z_+o2j9_D(6kA6F`0;N9|boXm9F^{%4()>iWe@rcvqVnZ>*cFuWmjZUO2FPm`co$e@
zar8*))tm|xzPsP&dah+?46JiEcVD@zn(v}|2M*M<JM4H<sp3WPG3`vvtNF|)Pac@H
z9Ro_wC$(Gr;I6OOM1=FyL5q>6^f+8b{!B=1Xds~=d5^_TSOS>TMu>jRlgF@rk3K=&
zDHHkcWEC%`Suqq`Lme#2<@FI9D+N(8r(xdhV#raC3taNzI4ir{;2@!1bCX{<Y$#gA
zNa)Fa3~Te=y?dug8{K~$43!~l-OiU|Vj4D_oanvPn6{8D*%k<yH>t8<55<j_l_yx)
zHtjpVrKQaqCtF8iUM7(n*$fjaotJS=CARR0LU8{htPtL0x5HbeTtnIdEo<__iHn2q
z40f<u-~O_)9GbQ_5x<$j{*V+?u0{$zig?9!VZ__1QHQB8DMpY7&x8xB&Cl5}F3fQ-
zyZB1`GN2`u!$*Cm>sIF*>0x&AD4o&R@Z?Kwg%VwLsUEjHg}Dx5J|aG_>z*k_+k^(G
zdo4<9YrBy<RlB*u!US{D6HYEJ$-I@e<TDT|e|S&I+a_l##Qb#XVGCZjAQO0<<CeX+
zRNlFfeG*<6717i$Ub+_)#sM;<x06=GI>by<SBdnrG^u`XL$6x{nUfx$1#t?pPxZ8v
z)md2CI|Ja#fgKRO{+9?Rp&?C#a76ruqA8|5=xWTqHzY(;tfA&1Wgp}_gt7oTnnN3b
zskj$Zra9QyWPt5Ww8TFHj@F~hdc)7iSNcin!C)sbIT;z9>THlZUhem^xZ*=Tao%R;
zsS=iETapz>_L?u*H6m~?eGahNUU0VP7CWZ{MgghvE!5OPL3aGw4CFW0bIln^Ajn4h
zI)#i|ujLU7&EaCNls{AjkCaCf(s1YZRE#DFT|Sqgre<~;Q~p5f7aY0q>XmkW!_&*@
z*OZjB5)~t1lTh$OO%UXqCU7)!Mn=Pz0e0)S^-!w)Hgo23Yj$nb-4yV{5w(k*cRzi6
zEhOKifaNtL2ATse)10R`g-(l}c(d+0Iy*ZbNKsi?7rNEImFgOCXcv7>Pmdm$xJ5#x
zFI@_H{rdV~56Ib2UQy~aaKCdWm_l6#wsv%sh5(g183gKyLfA{3R9|Uu*OG*eX$eqW
zlT}b?2SvC&j!@|dGG}l^#2FmG_~;-Q{%7rjm!MlpFnnn^{&A?|j~fa16p!(r4SpJa
zjY7A(ZC=$q5>H9j>fQq0v42O;8MfA7G64}PSGE(%d`X|S3|7wp#kfySnbM(aHwRr=
zbd762Y?FRAT>YuDT&LkH8_pj*+TTYwNM4wk!ONGj6i4!7KNV*709>!R(4>Oi3+!j;
zk-^r|5S|adX+V}r2!+w7>`Xb%QNt=G$7wjy5|JbW``_#VI1TurSVPGyp$9hLR`8G`
ze-+g3EQf1@8{7bUg`c~54}d0pD%qK&Uk4-f!Vk4S2HF|(@=u3O-M2s`iSaBkuukiy
z?MfoQd2`rBJ>85Ui496pbez7@7OgTXc}F<2-A=b2$#Cx$`Z-{*1Eb5T<?gD97gq6u
zLYwqY+iUrKDTLVSpQ$_CzI@5B5_jiQnY3ZOOlnraS$vyai|!IXtbR%rRf4+S8KuI$
zgyM&L7G3QQj<#TC-wyr_bhJOmDETgr3hA)PcSXpNVii|l)lfDID~8l)9)nL~wg>2}
zd<?)~U0Yk*qfb7-1bVoE0q4uf$RWkrt4gl*KlQ@NA9+z=xv3w>$3WOig+@30*c;y*
z7)zH>(zc8TJIyye-14ZD%RJQCUSFU|xT?!;Ikvu&io)u~m<y&{jJ}}2_6@@FNO%HV
zjm96XozT$H=@y^vmoJVpLO+UwDpP{Ib6~ud|NbcYldFtaTLw9O3@dV}(}_B#A^ryh
z6B^EpQC-czlc?bmN{+QZMGuz<4+kY#3$Ot|&HV+;MKG{Ou$P07U)J)Br)+F`9*gO2
zn!#e6`M+Iv;{opm6tJQ*3P<Qd?Z5{Gb%{m=4H7A0Y(I(Al*Ky|eARQ>BCyj$)*<r>
z6oObl6+0pQsNnqtEXiO|6id>l{Jq~FBt19R_4y%w^OnnClCVzOg;|hqP+=BY_bHn|
z@<Pnjp%D61fcw}RUBQ%j{~8l(=mP<PZCuX)sit7IsCtUUm^95az53T+=UXgy=~UB=
zfV*BGEd{@L5mP)}B=<)z1~j<Xx#(cZQ)4b>Y5zb3J8-HUb7xpHWP?~(7YE)hB?(gs
zNV$MX+|Xn0A!u!^II!Yb1F}Ys>Zbvr3dP`<0x29lVOLX<3Vd3w1srRc(8L1umL_2V
z&s9f3&7BzkeJ|gt(`&AI@pdsqeBX>>AX=rsqL>87Wbk<z<TV6)R}YW`b<tHuy3})9
z=j4W@oXp>mtb%8=6C}6TKYi_;(2a>aIz@<QIxmXyX@ZCs6c8Fv_X_@FLKQZi@U%&r
z?;F;ero)RHcKIO!W-g@u%nn=>)FQ*$1K>$G-(0;%xXQ)_ctN1+@HBJA&zCg}ThHk_
zVXT;tlN1D6<n%!fRj9{-$!Bmn3zk7M3^MEDwd3bvJNUm~`kI;vFtMkhzF5B2JHrRn
zWMJ;KAo2CalMq87O~9}Od5T@Xxp^~O_%_>?Svv`ue^1K!{Z!zHLs_B07}ROGtygKl
z{SO>CfGikb1#{{cLW}Iyty`!_&aGbr-w~rbK|Ybv+NoPF62*FIsHB5a@QiKwvndx0
zo20w#6A}lHVpc<7y-=l%HvlupbjgAuXv>=5U<99@3Y6#SA|UCQfk=SdTGM4~)=*MR
z5Rs#ph;Y!iq?y)g_oCp_{s-n@TsLKIM|)SW&s7@Pc)+^{;xLk1VG5#_-2Owy$d&j7
zs6z1B{rEI-hB3*ZpaaS+=)TvsDVVV_t}4h|L7wGutS<jIEHX4S5!zWn_8Es#7^+x7
zvT9f>`L9w<ZaZb$6+O#b>r{(!Qab4SNUk6r;zgjHv_0-()aBDS(2UI=4nf^OA@|NH
z=qEu{6sXj4dWZ=N>(tcL5aN)k3}v16ygPcxJc}nrtuMz0z2>5!fgT>&4<A%-+?Y!H
z37IGea=dZESK*PH3(5uFuDNB8h*RRpTo$m`i89Wt(a%x3y9gVWr)2it!WwFF0f_->
zU8~DO0gWpq0|Pgv`pf$ZBLV|s2L=ZAj3y)`OhBzB)lvV+lP9wB@<A+6%r^r9f5dA?
zTMqn#Ux<p>ygc72s^Q@g_wzQ#;3b|1b}zY-$2b{`@ai*jTf}kkWVw{?86R?vGimGZ
z{GL|bwP?9Px(JRoy^tm7=LP@4Qc2@85YRz5$pJSh2nk#;4B*lhu|&O6M5rw!gO{~k
zlUF&+!Skfs2a?s3S|4*YCo@uhi^g0G7ta0>Nyb|!dLl*XzBu`+Jd__-S0^(YMH|5#
zirSUI@YkMxT@ZE5ynOjGT*9*m#iNMO@YSuZ;<bXL+V!oWnTsXTyDpy%9+Ww$B40uO
zDd%u&*0q}7lXqe*`(^X$yL<EGuQQc*<X*eJQ8dzReKVn$Kh-*QQXxG=C#Wc7*u*l6
zEu_aHB*W$Lu8%7l=g$j26}ozt_EqP?-0b-X%M9U#mkR^og1(axmJgw%&-3iFYP2ya
zVPcGju&&c;S|Hj-qClu0S+3K)9e4Y^5R<}2Ayq*Mn>+a`2=Y#kzNe&cf^;MY1uo~w
zT*%Il>YdDAL&2!-_t!Aq$vDLdMyL6sxXrY#+xqpc6nr=V`n$Bmlp69eC^w0e|6W?^
ze6afvA2ildYwpx56j^!3;@x4Dbqw;vP5v$p24qh|{tPJabXf@?Z%F*nhOM1qU%q;7
zPX6Q}IK4k~gnd@_xy!7x6Tf3YX*8Wkx65~$+~cxd<#TFn_fBi7e%3J0BEduXO~hg3
zxx*Sf79F$=QE=_>vSnRb$?u9Q5lc>~opZh79a$0{`A+C==dF_af~l!*DyKHunwmF@
zw=`D!XS6n-c+g%5@;+%eKH0jvxxmFv#rrQWz%2y<snFnH4%G9?c<k60T3S~>8Mo*y
z8J!%P8-Kgk`Js?K!F0nJndzrBs!XbQ^upNaW4`B@$0GlD!xRXgDWK{yrR`e27+Z2e
zCiYCX;JLmJuiM+327h~5jA`%pc+welWBp^WdDhVb+}iJ?0OP2DG0*teBV<%a*DGvk
zH6|g@jZ>bdGj1-sJ4e-*mhp%xe@RWc`tc!EuWOW<lqTDklZZh14Y}(=4ZS#O-XFXH
z(vr9y%|O#>o5#(kpZ%tU0X6w+KuxU8GZy)t)$#fJjF8ka)Gj~dYQ3FkTLQwSi05jK
zb;+CzGUg(uC-dekg;FTlwfQE`&CK5ymbuP^)4Q}uR*AZ17tzi1^eIs?-U&FX(IKA6
zc|l0^h|XbJhUB!DI%yeCxlGRQtDJQ_Ks*5=E%?{$%_jk5p!!xo+aY@TNior#GcO_|
z(?Pmc3T-}@k||q0+om9%=$viy{g^)vN<%~d!T6$mNBj8t&Ud&aR?TJ)Tb69Ravjnh
zRS5q@J62mO8@W|{(bza1BBZ4oS<?fzF)F&Ag4mnq`beZIEwUV9W(tO=&~B7msZnCV
zLDPK}ivHAEWUh~Wr(-_&iPDh|F=BjaX#GQN8?`f}K<_6GxdQONe~o7~VFPO1PKEB`
z{TC~xZ6+O#%?xB0eA&U&a+@><43&VnkTwrh)G?@usexHRT>r=bZCIFpP4%dykEogJ
z*d0%*Ax4jbsj(mLhxZ)z2hGtOoLahN9)-xc2r43u(C5A7Nv%chs!XJDr=|5@IdtHF
z|KaPI`5Uv}zOU}95?#*e=FYnD;-a{vJxLL`KZ=G!t}nug1)M;^2#dx28x55jBnA~G
z3txPn`u26O!x~Gpdr2miRHHDc4aQTX9HS%%KoJ#}!^a|5v1K8+FCTZ74+VcACJw52
zy7v1^XaT)AA)RyM!wytdT^(2QPI2jTt8N`8k&3)oa$jla<X%_hz`afF0Iy<^p=+`;
zQwx-J<T3(7e;=$?sYoOE;QSXthL3b0Wz#6?&V@F}T(<=iCFqnRKAoeI_-qn5KOdLE
zS$%4dmhuz(_8Aw-S|D;kZ_jQd8(<YX|2X(!$M(S&=k9JFeDUg!gD>R&=v!j@<GhSR
z|GbRyqjl)pw=9y9)v>+P<4-ZH-~!x?*kB#n(i*?+QFD$VV0`?68Rk?SY1lGVJoQv}
zKWlmaGKa;*pkKwb4{VL$RulR%kJNQ1x;1qt%qV<PhZ}gtfA>1aXDa<`yg&`B#Eh5<
zJyGn{Abg7UHLtqvV0UuPTXovkD+){_z{uK)AMbLQ!fregEsr}IkyO|k-+i|mlvK8_
z*GJZjO=L+(GVYL&_&Z|-y3^|frUQBoiE_E5xuD-D1PRtJWp~A;o#D8knx)MUprf<>
zvE<}Xn?{E4WS^0Y@)`<0YUL4SuRvcn(3YOe=dE}NybcOw*U%7V+TDmv)JksJtvFv-
zdYvU>!twXrnMJ&P?w;Fb6gNmQH_DhX!@=o?wRuqjJ9sB{n!?jPUGc&^$c}Qxz5#Lf
zU;b1(wc?fbo-*d;&?RF&z7crRZ^ol*8a(|Pr8_PX=k*Upi(zy$ul;WBrS#=(^5HS^
zyMtl!AaRvF=g%X-&iP@K%NMxyDF)<vN=EL>N!tp8+!7jA;uuLygZXI^`I*7)cXQpg
zzM70Q>~-(J&@Q+Muidx^GbcvJFPOP9(6YbaKHNk^u=Cr9G?`TF$Ou2P9^DQXSPlgb
z(M`f%cY4^G@9F_R)xZ%6b&Su#+F;>B$|GB_%J^JgRwijmDs-!-NTvJ8n09%W6v1)%
zoxY)W<eO~My=_^>58lS|I`uk9dfZa?cD;RDDB;fK%R(P{(brQ7hH~B3kk+Gu`%p3G
zP1f<WYG@1>$j<)i^6*@VH^R;AmF9yFVusZq5CJsUXX=`vsiBdQSw3=v9>eq=W0X?R
z%k07BCzl&*{;j|+7I-zX;sE|^`!<T&OTit0%HO@PxhVBc43uhX*XEyp2tsP!^QNxs
z75=<f^OUr-Ht@s4IVfy8>6$=^IG}I~m4yZg=yV7dch5t_5Lv0f4Mx22*=H%qKs#l%
zZ@GQ198TV0fhOWEdCb@t9H>&*cdN3I2Tu0rxfPq-{ocqI;cBRALH{}GOhdwqTr`jo
zOF-Xn!b>^}vXBrjC;N2XyvYH>4P`*9r+)RSbs;YaXDqrdKzErA_tE`t&7pj*30<xZ
z-z{=eph+mc5wHP5T*$Yvv9{C+B{ljY`3kQ0?+(BENFqfzb@|XMyD~sfs8Img_JG>8
zK<x%GV(3pUSgfn9RRuR7%Q)yH&?Em_DmO_1hud8pPgOd5Pu;-F&i&aDLUF%|KW+#N
z9T+p}R|CvB>I|Xj<HwJBtJ*M5?Ci*0iS()>=cyEc(lBy`i*le4tu7ssGBPp>CJ<C-
zh?No0#N{u?$jR*-NqScJNJwohp=U$DL7q-9!4~tOhMpr;@Rd+82*u3L+oPxnh?(0H
z`p|~Oq@AG*g)|cIIxr<2ndMH?+|?jO>O(ybun6@}gY6m2G!vU^Gg<*m;)|f1q}ueC
zLCvlepd<2GorWS2RO(y7Bi#-SG)kdVu4G3-=I_8J>Fk<HYn+OOB4%)S_(Y2U*eXUl
zk374A+?mjhO<%-^&03(%WXG!pTh54zZ=kLqj03|GAQBl?)yx06`?4O(ooV+>c82P%
z54ne~Ny5-b{OI0YwmG1t)BWwE#kN?P|Ld?F*o{4=pxR>C-y-cF0oq5zIA2ah1v(3c
znyr@{uYh=Zmiy#RWm9ch25exj5GiF$!@CPYSjACTEN!%!<P_L49BAYtQdv6yn*_f8
zuaBa@yN~NWDds`qGKeWBq8l`|*+C5#!~O{V9x|S1e}6iuTK15G0luBRt@D=BnUllr
z@FhOH?3l-;KmQs9oq(aQyD>u*!)M~}9|z40?&BL65^~s4q4n1_5K{j5$jdMeew*p;
zj*^vjp6#3aLSPYUh`?r9kH~@l=H8V!4{8#+S-S{fk^!y@7vkpaJM-q}i^8%@vT<&1
zsv5yVFK2slV9=PW!wz9(_c$z*xOzY^uCpPX-3?h>l(M~z>eUBYq=cVR>N!;=(Vu-H
zA<fRvv%doU()IxQWm$hVo*5FqIXzr5tEaWaJe@it3U7vC+&`YUe-ixU-ZA>wt4FIJ
z1v!wM#Oyxk8<l^_eBVjE1Yf@Wn}FpiFT>R;1IZrHr7`U2+$ZrL^lP5u^zIwB%qWZM
zeLA%C*yh(#yv;9Sr6+nytnJI+ekBfZ0fzgZ+0_!c8o@@1U<Qb?$?qEtw^p^lRi6yK
zg+*WiqA$1v7r5?mh~a^A6q1wG5~1H;Axk$M?Gf<yK`e{#gl-6A#zu|=__rNwo0A`&
z!g6YQAw=qV7m(9wxIOF_=q*BoB@-m3qD7q&*Y{u>Oo1pSOF&KbN<rw@FzVw{hB45y
z1ifbscT(F^e*Cxvi1PPTE5sK>IS`$o7qqq-y8vG>a6*2tU<aUSWCMdH8Vz;zIPeJ|
zN9ms3yY0pS;Kzehi$iB=1Vu(p-L<gefRMm&h^O*A3C2ghb8PQ}hYyijQ>YJ@LUA}$
zLNiqZh_XdrNg?V{fCeQ_2w2bv%Dn*)2L!V1Lo=YiM79)W22P@IkS}@M%e}Ew-nF{Y
zsCM|kpmyOi_v9T|$pJMX?iWEpaS(=}BTWiiN66|2f~?-^b2|Q7w~$}xT$OyAXt1A(
zE1zxR`!=PDFH$$)rhOC?J@YN^ySP3rO!6ry!wNWp0Vr$>;-01ME~wp69KYBETs9rJ
z2afQFnmGg9g%FYz!I<CtXJDb$qT(~4rByXgw!Chh9LtDdHkF6{>EaR+;CP1i_&Xr;
zH)NhY@(lU9pe;tw(OfD}UHL`^d{F5J_)7zxJwDe-g^gy<<Z%IfTK5`COd?asnlaft
zI&l=zWA)$BO_HByzCN(5U?5wM_6<jW(OO+~yu>#_gFe`0rK%ay_V(fJ_t##&^uIIx
zkz~OBf`DmK&PU?Or!7xajjcvb?rCENq%pe(;7#ON4Gp)VgP0%g!;^5HkOFm4<)hm>
zg)aT^z+b;WkT9mX_79wNRAuKjPU3zmwE@Tq^O33RXiWf;@^qhn(3K-s$ggr*-VU)6
zFMPa%=qrq_RLLgh9C+0nvwdIe57cQ;{SRyz`CX*)**Ua-Txp8+WtBo@%gak0yvmHO
z|Ja<JrP|&`2B1Xw&g!M-v^7(?k|HcS%SjfhO**VPB_$1L2*WZD+CAY{=5D9r<NM>U
z!@^X3Cc97XL6Nv``=qy$r}b>04843w`2555YweF6mt3h~d@!uUzh36Xqa`#EQ0o0V
zM)o*V$M|E<c?#I?%S<7npEL&0XJ*CExI9jUc#C0WzQa^@G-!Y-mfrQ8$k@JknTgqO
z`o$!BCLkMhA+y2syVcbi1ev`Gya@ZJB(@(>;UDnQ^SGFto}|uEbRSIO0bI$1uB`C2
z-T2wQ{d&v$najPh6j8jwJRVaUO6r(u*8&ErsQzZPb0l>;B>su=rT8_E-K7U0fMLfs
z-hI4AR3M~k9eep|20OYvmM^q#?zG;%>9w?TJMvM$!&u;w@7v#gkn4E`q(^L41$MqV
zI6DBu@R>G5Jc}}fZU@x11wAU@6`BCE{G3#<6zQ#p2MV4*a`U17&wv#72J-xX+tDpa
zwt^IWI~v91nEx<X;fu~-W{2TXprPd1Ga4qQ>7G2@bZBraOX>&h3H3&T-x};B?@q|=
z#4<ml(xIP>KST1L4SINR&mc7fc`A<*{{`t-56IdHqo8yYf|7a9_O~EF({TDI4GqVc
zGhgGCfLccwN<upeYMZp|(Zb`4TFMR6&q5yuSaO{V&t=FuI}22QuCx7b5DX~tKiBb4
zq(LX<Nr&#n44&|!w}QqgXhk`RDmYMM+&*Xq0IE2YeI~x=XF|xzfupjJMGNfpdnhU4
zJeN8#eF4ms4A#tY5}1vkfuSe|Kuz5Wm^!l|8T@+0y%#HCQrkY%<lj)mzb@%NH{jnA
z6U_g=gOUI6GM@#Ao9aPOLUYZ5ka@s|>mxPq#LM!;y%6wj=F5$bLj*JumF$0i7ANp}
zbYjPigOXp9Okvsi%oyKZpLM&wy@O*84EoX%q4I<EbDf5`S0Oi7`5X^W(`V+&<yV#z
z1M$l@WyE2LRxDlMjrYMi^L((YqCPSYm`&PvVHi9bO#-vB;=Wv~xn(D2Vr*A`plNIt
z?og#Kgbj!Z&_}<Q3ao$!JMf<Xt^e@dFw9KKxj4r?|K}+Cl`Fz|%>~<=jSdn{e4^a&
zFyZF4l@<RxmmG(-kC`+({<m=N2UG|G-QB~7Us?LPK;;X=_+B8cI;#j<3G*Q*gmIAV
zEXX6<Db(HUy6j#APk<TT%ob+aS~G-~h;VE8y8edIk97MsBoM=$^VGpXEo&RUy^sH2
z&soLFFJ}%;+4%(rvF@p;j9mtg*3iSKXMc}hbWF}^dmjneE<}53r{Z=fND6s5Ck>nm
zk@K=`oRg%9vXaEM_dxmV$&Y%X9tXD{3V`k5{R7=c_7D8yqX61;bX5P)J+re4eD-q}
z>;-RA#flYBUyIKOko|YUL+1@>SwuagU}T4Weuw4-)Yb_nas{FK>Zi-4aKIl^5}a0w
zMOEDS#r>bbBE40lN<rG5S4Pgg*~1M|hBvDH8PTzkj*dCd+4Kgwa4bQIje@O$VFvdw
z&%1zj;M6<O{B+G!h%`JC`4;V*aI~T&l<1(d+3L$d*cSsA5$J*QAKE~L1vV>(I-Aj9
zTyQKz>9bX1V`DhjP>v839i0i7Nx(5V=`%VT3HH8g(Y}bp^9+F{C1OgU1MTn)w38*u
zM$29E{C!^X$0jXI|IqXLZUf*77j{VK^IE%uYk?ijic=CNs{3ugQBCMXCTPfSD+2j$
zZ31DC@3-tUgxLCgjaruN_LUy0{=D`_9MU@uf+g@(RD}ZhF89{wJ;)NeDC{`)Vqp@H
zOEQR2l;A7HFtG<53`b}GTLuMv#ktnKB5-g67ayNGOv{2}7hG+2eCOD|;2P+*oQmik
z%vfAp6iKFZ@-_P5Da7W-T-_vUr&mH<9cKsJ39h`c!vUol6l#b+p@|~Xx_C+e9q)j;
zCEx&$_j13*;lQ#q%~U40pD)SNnq2g(&40sTWfnaJc4#K;?Jwj{n`5D(Z0+c9h-aTa
z22U6~f8XGKuEUMFswm+Zrmut!g@E%V5RL>}ZaR@-bZeu0D-nPy{|%?ISiJdfXvjcr
zZm{obK`*{sG&J|2b4k&W2I!m@@TK-jx37}B_&?cp_oIpn$dcm9tUaH#M;VVzmQWk%
zx61fru%nZue%3%YF}QX+;FJi9U-{5<&nV%c4WQgzI%Jf$M<;w^-H0E$-|qC4=;+ZY
zd-n*l5l&el?<=%A%u5B^9yN+|xskRY@!Uy)o}?&>0j>teGJyr1m<5LOvLq!9jbJA4
zbvvm4X1=Lgpq)Nb@8!h|=gc6}KbX}ibrX5)MqjN@8g79z{SBPAmhp(2nhg|QiYk$r
zO6|@ZJNn|(bI@|-2`1a4xuzy2MON3(9ML%w@@y6PR+Yiikfv!_+P;fVKtn@gb(qJy
zW~04yml31#(y>W8!5xblu23Z8S?Ph6>O>e+;3bNH16}}+=zg|}nl7*BS#2xZ!9hVu
zB_(>$N|M`f;sLS+!$xwn{v)@hRUx<!`~el;+5qc6;%+z9n|>7mEflEi3r=SiT))SH
zA4-9+@St-v|4rRPcm&cvUgQ&P(ouBc<9VPzmFM;uf#tLcPAjpPDu)9Fj6hSNR0&rC
z1L`^861PyJRaIBEsUOt8hoHP}JDv?mcZgkb<;l%skKRKEsnOS0=c7)yn-)08HwZ$j
z#}L>e%q#ZXB>&rtIK)iWz<E3HEo^9^;$a;4;Q&@#3mhw#0)^YA%Jvoig-{b{7%Z;x
zXPndv_k@H`h(pH;0;$`MYK_s2I`8o#4d*WoefZ}Bs*(k(A*i$4Vfeqx#RHF^R?Bhz
zeDeylSSx{`lcjw|x7@2VyB|95fNPz_p~@>@HR#(50u}^SaknHqX3IRpXbW0q1tn@F
zgAtF|g1$5soMtNV?eOzPICO~!BK;(sqSFCyNSCCG71%#_sTc;861t0C#6VRMbvSo;
zL7QkG1Tgv}PVmdaC#Zq()5D=gL+cJB&>=H4uLo7%C^&;i*RUKot{)wLiiF%wACV8P
z6naqz9MIjLkTmbO4fRJt96XD`Vnn@(nwwi8FuPZWcEzaNC?Vwz;BAs(ENIiA1sel2
zd+DNG1qYf=+^xl8tk3uTKkdDFIMsXGHol}e)z)MRg`|WC$y`X9s7#5JIZ~N257iFI
zP*R36Q!;KznN?US5;A0nSe1E}d3eu{eee6;&po{N{k*^9cO36~98Z7jVf9_#^&LLL
zb)DCFo);Z3+2hS_(|ttPL|ldSM<10KRPFgFjf18RVRic{Z%0wHM`di0sG8~T0zYk<
z4Y_sL^reD<E`CLWf-ybtwP%;-dFBYgMPxbK_2;bQ-@S;1Rg7$he?d|H<9+$RW)=VQ
zwSP#o{^ybW&m;K{lMu|lzkzE1dG7LGzIKgQaQALCefQbSCuJ11kEDtT$0qk-TNbUj
zs3>{tU`S?1ZFhK_`Rh{Q(1OZ%Cfu}%yXMa+S-7VV<s78xzd5;bMipFH$}0OEaM0>u
z*+)urYW1w<GmFUTE1Ctb1nl0b;I<Gs4k&x-nwq>SI>6*efdC;MTte8T9R{CGXJB0i
z6OGuD!;KnFe8;dO2gZS5b^&5Vr{45ZY@AXb9_>4Yn5%ZZs&q7bU5JwnA-vfl_R(qO
zhRg~3EO9y@1yTHVL*Lbo5km8EoB9+9I7$PHdKDp0!3|fAY#fIUc@qx{1d02mBlp=H
zB;MZeN!qx1bMnK>1r4W*;A9&IZ(}}amLLj6#o%heC1SG<xh|Y%vE8;OpNE`;`24}n
zid;Iv@rv)$8F*Tz;isNxeI7mpFk+!c!0}pKZ#=ZQC?NEJ`({1n3VqE6Hnt>eH!?8T
zCTD?9>isefddg332y#Bk&F8h#Xtj@zu~~k}`58!4fC{JVBl&V7XeLPr-iU|<w6XAy
zM4d!DsNqtCqEmerY}`KqeE8XH^YH=*lCT_7hmJBO6OUUBbq3_uWRHdwPp``*JU$}O
zC0GS6?`Qv)Gm-9d*$eJkJNfu30U&OKfgW+cMu2^<#K3AU@JhC8002U{xuJlK8#cso
z$T)Q(NY76k*mEw7PV_g(oVa1Sf+B=CY--GjG2H}>bKdM?OJ0LneQJ0*;>B3ltvh_X
ztLzIc`uAtk_lzLzVJ2pCxF0{ZOIfeAF-}Oxf*>ardL=62az7uRm9a1qpC}P>JMRl`
zR|$9^TR{tNSnSORTHqdY*8o_TwVFYvJM`HwfCYKVL8zP<fBEGcamm}wP!xh3#_EGu
zJu~XxVxUO-VY_98AzJ?3(9jEyWSr|hqq5D&%-q>K54Sxw@EXM-{r5#piG{e#fsE-3
zj~Z@K(HaL><okj<Km*lJ7|-1g=F)Y1J9Y>|;q+))!eKB9UKIMw*m2=AwHHysX|xuo
zmH^iz9m=KP0ad`s85J<zqG6jBN2$sesZ~8Q#8tTKA^1|UzzU2C=*Jg#Z3w-pC#>8?
zoL5A(DIGk>it%_5f1g6Tah{n9rFytr8U$Rz!~umHCcp^@2Z`Gm4Q5;-3`M)4q`E-7
z0z;s8viecZLPG#g#qc)5_07NySRBmRGnvGolB()Hb&7^5Pldap394-^<cZkG0hNJy
zbKd!&;9xd94KYMiz9?LE%v%7fuW%)PXm^a@UXqnL0Qwp$k(gWcH}3-zh2@~N^>D3Q
z65T1-T3!L6iE5FS|D!nqjw(a<B4>)TAm;J|5p!RYYPh8F#Pl>9oUs$^+yi}yhWsUB
zWF;DGqQw}B7^8}cio!o)ZD4`eoJs_KmSi&f;dXUeV<qK-gMuWQ@Qa4!fQfc9DDMu3
z75GRtvCq#!hCp7IX%p0meyBjy9|Ys0Dp7;q5VC2e{Er~-`65EkdHw~#78U4|FDy8?
z;mSAx!w3_wsB=a`aW-M75@UhFQ3pOw(oJ1k#&Jb<99PEz6C=E8YUJGL_E>lmRCFXJ
zCNd<|9`;x&%_kyKtqr5!^URq6Ydzao7w9B3;D}aZNy*8gIxjC-euG@teX>m77aq;5
ztslQo^kDxuF*(TspAg#@u)JCrxHfVfUX8eiSWxb$@XGI;ao5`^pPMm&%uh112_8zg
zpQN%nhMH}Uue{<$#Mn6_3HBtAKS_*7tZgVyT)PJMvv896=N1kQUg@aE8xcGuSK;0d
z&zVyoHp1gQ7IHjPOgsIh`YVyw+c%7xp>9jKuyxa$v4o&-f@x~grLU`-43HoJZfuxC
zvq}SZ0SJK4SakXf7S|V-z9f>o3G2B|wskx)pdnsdShj=V8qKdV)pyyQzi2m6WD;8q
zDE_*)&2x&m4tz|@4l6OZTWg-*JHAAIHgPUH+0rFT&CRXA4Xz<l)8NU!-^~DX5dw!&
zhe~}hO?M($$AqYO(JkEnCTAP$sZ=$SYEv#a_V-S|u}V%(_TATRrncD3!LK4*D!05!
z3JfJ;Xo@qG6}&)oX3}o|{%ex<{k_dEw0Bjc!Ez-H%0LUqW{GUkpHnUsh=hcKktHFo
zLavwlQx?Q5P2`?7f>-V4QP&|Z7)WoeL!EU6;Fk_gq;=g@IC_W6y6O-w3m$@f5{ntU
zt{U+Bj)jUS1=5O);6+Ma&sn!gW3mxcIDYIWJXNeakD5HsXI-Qp^zBanZcQtigpED1
zDasv{!7>6Pmnax+9%*SDp0P?6$HmmS16D&C{k}m{hN>auuA7>ecoMa1oATs7*XD-3
zUoVjRC$dZ!*+B9eToECYq(9rOb@=#kO?CAeV||DzbpcNkZ#`mUqAM691$t5>ad?6c
z_R&_i=@g!)sGMP~dvQi=4l9Qy(iu47$H4XU6ej!AbbXkOLCG;nY*s*O!me2d7bY>c
z83$q>fs%E-ko)8jvJn9WCQMm!5tw*H08G@58YlsRw2G=%5=5220_t$3c2lDlh^-w)
zZ^fQwVyV5S7ezhIO}9(F_r8=OZtHz5`DaKw1tdJ9?SQqgg$rk{rTl#IVyNu^afE`Q
zXbN=chXSERjV7eqK*A`7rPQHSn|6T|VT!@I!xa{zfpJ8D3NH$nnAG#&O%j)zQ>rrY
zxGduCT?gMYFmM%3Bk|15;n0Us^N4gHz`~bE=KgrLy0^V)tH%?@s5UV;89pZNe#Esq
z7Sfk4LVSmMq}izti+l>~d-=d@Ba;F~C)bXQxz0^?5H1I?B?z-ft%8B===Wg#vtu%t
z&!m+@B9`B04Hw~c5sXpvl6Cq89G*a+s>d(YAO<00XbN^0y}fb}Jre^2R2B)MKH3GI
ziAFpoJ>e~rNQsm+@V~?d1b>4Ss~QY|+H4zB(oTSMi9ml8WtD2;6ajTDoWyp5B0`WM
zL~38AS~M(L3-XDPq|D^Edt_CCOC<3^!JDeCaii1(G*Z1*eX;o{Lny4IC`(K^@Z@8~
zFMi&l0*QDv%wV3wql%zD#2b(>*i@_|Hjp3zHY|D;G6b%I+Z~>8vU7j7a^d8M%ewtN
zE(q(0@Pn{Ph&f8i*?b*@n!!9W@R22!{TtV<ivdd`v;|{+SKRf39}Uv^0F6Ot^mB{@
zKtdR-7*vm4NJvZ!tTn%nZaAQ;N$aigC}SqJ$1a2WXW=kEX1inci!H0Z?f_Wvb#$!h
zGHD1P={%N9+>=>6+h*?ZQwB6(zE9G2SVIgUZdTnQ|Bu)%20p8#Z32Qha6{TpnQoIg
z4_pyJS@Mv+o)gT?%v}ON!Ou9}TKbUKJ&|~X1hLRQTcF@US_eHTJ0NpD*@CrTc5J~_
z9xGlQ9H@1lOc&IA&PAXsiC8@S{L0cL2x@AfK1Hw#z(WJCo^S*gO6b@aV~I~2tXQ2u
zVt!8iD90KtIN(!Qem<GrgTIuea~1GN1w1Aa;{sO}#k9p>sf$+yW=Je!RI2f$1f4i2
zdF+Pl6y1G`MwehK(YrKvDN{#C=Pmcvt^45s`GAKwZ`A`avjwWWG-W*{b?L(4?y5|c
z%p@rwj*plwj{#7k{&W;~nVX;wZVprmJSr&doKRF$R0#IbQP<E2EzlC_j(o`(49+i+
z?MDd4?%!P}#@;e#^Zho6oD<dB_S=?MGA&=V%I35TfUwS(+5obUOY?&&9|LPD*p_^Y
zaRqgpw0a3d%-vI$?^$&P#eDKUx<szfU6vCBqw=Ti3cKs+6)zp8Ki(G_9y6>~r{+wx
z=NC>n96EV25DWynudnDJj<JBM9b_AuE*tQ4I{bQw?H&?~*7fm=UTriaPg2>VzyskL
z7*KJrXGsD|LzFTk+aQe@pv1rTZ6ZY&wKYIiwKxLi3vF$Qbgwp3B>_e?Q&X=EnjRaV
z^|{NAXtcQLPQmvER-QGVj2#333V!R8(<*q|6tKcjuP=mtb%@%D6VZrDv7S2S1o862
z_HiC<n56YQ(V$@)_zkcPaL(Bf^;)>DR*srO1$Qn)4O9d~`w(HhGe~#HdBjURx-lb%
z%~pG!CuSfBNDVF+P4M6elxo*!p%lD38rwWd2{iinTS(Y2-`P!Dzh*G^xX>GZxsR$9
z@6fe&Hi?QsyBUUwrT!!LQ6=~D1y+Ulw!b+7k*7(;qYF-Q@T$^;6%_>b1m-8=3#@-d
zo~x6J%-so-cfk&pj15=~eoGvm1iOjYwB~;s!Sa3r<FIL#&S5Pzhm*kHsOS3~#bhEz
zXk_x6w-(H99C(5^LiRVFu7c@)V$&sxN1wAZV?W};q`Hp!<kMs#AU~-{nWN+_GyJla
z;M?FVIzbTn2NS7q%5@4w01O$TJ`r>h1O^G4@+3@SF}Rr`@i}DeVm8xjD)-5Qj-C+S
z3TOK+2RFMTO&h6Gg1PnynECOPh6ZuYb}AFwJzMz7b-vpkX<P6ENkbKE=;BZ#wE!Rs
zj`Up~%jT>TfKg8YRSoBFtTlnp88+rR^jb<lOB9KwNCfoY(R72dTSv(8Slv73lcwfH
zNN=o?5@K>laN;?<dn_4tuwV$oLxyX)e)i}*{C7mpy`pU6;Yolt0aV@^;~sjoJyc1m
zq<#^LSyaUD%n+UK&lMl-FCGc}aMf-)0w&(LTfL=MRh(cdPRh%~ae-JfxQ%5*Ucjsk
zj6;M+oU=&xA|TkQoebqbTu0$hNc3UEX9zfDA|k0Cd{m1NGYPJ$orENft>!1Lis;VB
zazvCkpfT6x43J$F0m{OM$UL$05Z@nypd-2^-eE~GF$bV5(2SII)jjJB@6v_U29MG>
zU7NH}LFmdSAdo`MI0ixtv8y0{NszLH!weZarughE<&}1+cE>|tpf&e=eK}>-67N5d
zY)q>YtJjUio>xlHh$ARaHQC|H4hO=)iQu=Ak!&z{Al~lqwJY*kt6w!uj0oX;W?q|k
zA8RbBhY?g3t|P=rj68kPhY9OHc|3&cgUoh6$aV-NlkHlOgVq8^j>Qw<11uvqRTc0Y
zxgp^=5z9EtUIG}PiOFzGy=MLBEBdu&h&EvHl<+gZ0ByubG>t!jn_E@g-7B_Jw63T7
zr&ll$D}~`%{S99(?d@SIoVfX5*XbA%u+|phjZ~yrobjju*@qx3wpA&96|KtfF@<qF
zwzOy%tsVmtpc=0*2$~wyh(adQbj471Azf#P=mKHYQ>p%}*!``XgT}sgtG!T%0u|Y5
z9pm}bFUd$}!5peHrf}{)=K){ittx1WKzQay2lfKeAYL7DIvr~@ds+(%5z0IUi}Og}
z@krSe;-P`ZK{QHQ`mWo^rsbG2osDlkH~GLyY-$j_Od<G6CEj)D6;p$t5**0K#B~{_
zkxcn7V6jCUCsEdM8_#X+TIfXp2jW_ZRv|33sI8JXhU;B=)N6!@g46Cxt_UxTenrTC
zRZJe+w&~X|TC};(?mNse>tQVVs0qco0pi+B8qYyh^a&lp&N)Rc%rS%qiCSrA1*Z#w
zFa*qi3K9J&CW{^1B`;tGlM#W<%obKnPa#;!!A7fh_T|L}5pZog3XuV85@^bkfrUg?
z#72~8Cj$cmk!U_f?S^RAx*IZ{09Vo*BY`RLs0NZo3^K*s=Uh~(5sMH-ECxi`WE-!F
zfX?$u7>NArj1Aa{O&DMh+o}~5EpIX+wzHW1f{g~-Th=S=Q<ta2FhCzq!CCIS;z3qV
zu!3A>?LJ-ZvH?usc%lj0HXs9E3>h}`aS%qgr-m(=HMV)vcm-PqzG$PywmpN!Zm2hF
zm~7D;0*68Fhke~?2fT<e5S&%84_!cv(G8cbCr1`t;h^T^yw<LF0Ik3wzi{TeP=p{I
zr5j(Tz25x4EtRw0#egH;JkWBH#%o}#JZi#>wwpSO;GB50ZBs!k2M^C;je!v~+p2~E
zYsV4UNNgg&@;>*?QNnnsk`7WJJBe~(=lX<U&7%UNHFb1Rh;V3Mu~zGQ0h-|@)OBS1
z?v4$_p&VPn?_UGu_9wjf{*VHk?YL|O!Fu*C1vX}8zIVd%Phdua@P?TC06t+Ty2FFc
zMW+y%k%ftHbdg)UZHPcc30*UY-A}OV!1uD?By3LP;>DodG@V8+QG;*|p`I$~M8X{)
z&Gl|*zXaBHm1;m<#6c6o!4EAOanB^?C&>+YC@4N!brk|>#5ErU97G;>nfV!tv@dz(
zrGRA<gs$cWMF?VLdoeP?-gNRtio_ssjzvmK8sd>Ac35h5K(&W8D*=VFBr=o{RuczR
zf-(}5bJFrl6#a#e#HZA`aw2UbRyD|>q3%0Fpc2v?j@axj<TywOVG|!m+REXQnDX6F
zEhD9QIL1B&48{*6m84=}doQO#q_0BER>Nm4#UB#a9}?dKutOG-*mQZ52Y(S+qHjQU
z?fUNRL>qm)4hV+%*JRc4W80Hzcd9_%r;gMBxaKH;w$2w#m+tXkMIde8{>a!7U8*cB
z-grj3)~&<`59QVx$oN(K_s_%LJcx&EEpQ+Tj6+WqGtdo~^3l2FtoXxYu(Yj)INf~K
zom6VEzY+^x9u*|`A|sX*wMgb39^tx4RutXzSFSA2)ZdClvV+A3!nfd67jeH`GdaXy
z4XGhRks4Uq&tXImJi8hFM2`35JVGFGgcuaprRk@^DILjcE>=`|_VdX}8&PL3-fR|C
zpa@P-;Q`?7joCI&crIW|PXy4F>5hUY!WyX#ufs}8H;Dn!?`Vw~cHuytY2zQe$1s@S
zlYfnP{sAr$&@a#>qY^Q!?8H}klHdPjGZl2QM427FM`A{Ae<A+z#J>y0N%9Wy81-u8
z(0veD1Ol(SJmE6Vn(<5#@cd)(wez);g$ki2xcumjEw3^Kw!Er3>qSt4wXzfPc=IH$
z2V8W@VjdxI=VkZlhs2v3(||DOh)puj1?=!)h%c~5hBLLPHvwb>17buUZ<KHupflpb
zGKTf1ie$+^R{&EKo-A~mB?MiBtvi6s8S7)IA*mcRR!`67J9U}LO^0GUh+2O;L^g$r
zRTdTF&>CDIeL4aH>Jg@qwmmhk?bp82FibW~9#LA#stEW`MsJbeAfSndZpRQ`W;5me
z*OhUgftjExfR72T3P;oP$!GH$`|7b;J}Y=HlJc1L1rFzH)vFy4k}fQb;m!z)nQWr2
zO&Mi~eO3}}F~UueGCJ9^pr@F9##BKu`^MaZ!x)J3YEnVu4vHu?)PmcDJG!7>dc{M#
zzUp|TsMVvko(Y!0)tfu`VO4R|3_A;utM?wCim_v3iG1cw=d#H#{PMZvT<h{{OOO5r
z-XN3@2q8MUPKG4|wMb(wN5j4Ex(y72V!n1i1JYBiNW6{+Qg<P5WjUq;m>lY>?_DC4
zNUFBhJYhgUr+`2IsE_)$>?Sa;{_p&ntoh&5t`H!wiWqql;Te|MSyU+o<?mCzx-L*8
zkxGlfm6VuyVZ$VOGyo3WDe39n43f?f5h4+80ABvhD{%>BfDq(3pBt90DM6<_Nl6{D
zCu5r?eG`IZ5%mGg4dQ_d<N-vuV~S5HU+rdO^uV8M!zzysi_O`VnL0laedLaYf*3^_
ztjx~AqXjM(KLO5)+lErU-p7S`u#y=~!UsZ-A%Fp(#%i=&*(xVjFLxIGCKt$A!25-P
zqB%$D)I5%h`<2#i40aNlQBYje&~6HoKD-I@2>-U}%0Lqcl6?e{Zp*O6I*R0_3K36F
zwGh%D2+;%MV1Wnk(3;Cby9K1A9{%NIAU3^P26UO|Z2;P<0ShF}EKt`AmWA5B5tEvy
zOKyHL0Cf|x>Z%5<qMW@6{-f;Me?vE)8<?L{z0D+~dEmfxVlaWol_aWJ)|5eMqMr$X
zpqRA2Ino4^!$<n1B_$V7pIHrq0|y!G+yDO$z%briJ+xb%nb{n0HTMbsf%R<zU4&TC
z{b+$V86WT8H{(+Jf~VnPMAE8biwGlA{?om`(GF7==!K#(OP2g-q+1}bMs?jHWcc5C
z;>T3~#5?@sSn#IrZ&NyWc++i~F#qjU%TH`yARd~95#nL@lK%$lGxasfeQ4a31lv_&
zydTrCYUN5%Yyt=;k08vPLcip;YdZ-Oa0LVmDBPL@k|t({m;vyiAD(Y*`LgWv=@xtX
zd4XL?my&%HPi2%{H@d&**2%IuNm;`a>r`IeFO5t18tQ*btVz5kRL3H);`ok$)k{jL
zW{S!I>FME@PJa>>_^ioV(*Hc2S5zpD^+3b%b2kOd#cB>!MwQ9BKk9w|Y*laOqw$f+
zatYIc>1OAdcPaIgPG9Pepu3r6u~_76uPF`7yWfCIhg8*<FYJ*s-?kr77obo|c;z_>
z#8%bV$cp5~$J5hO`^Cl6C|RE`7PGUn8;a$e<>TTamNpIp&aG#BPSR2M=%g4qq6-V<
zuv5vpG&eVIZk?d+J*b<3NIna6Od6<<ek3;>syU^)G508ia#+zL!UKU@X{qd3dKOIj
zX^5`;S||Kky&+<i<vqsQ5FPo!g!|!#gNnr<GRB~F=;B&y(dqF4d)4=MrWeU)ua63-
ztf|=mNQ6VS1+L+uVq$(-Kfp}#N4%w^SPYNyb?8TCl7|bh0<Ia5l^Pq%Eh#C<#btn<
zKmAJEn>U+O10y0h#^V^Wg(yK%hB#wg-KPBfd{kL0P=-8bYj0nQDD$B1r|Rl;5BFa7
z_x48T(&-_<2-MT2;Tk^HxOow}spUU^{!B*$H56X0Q2*Obcy6c{o=w)x7C{$bW=6)y
z>E@31XfF;QBtpAk4b1@*?@&!`?R`v3iqGp`{Dq&te5vg0Jnd}*D&X|sM_hp);=;$D
zjp-<AOGhy;<rjnk>#&MZ?hqk?$AUnYhft2UEU0rAbi0R$H22>ex-mJ4-{42Nx5_>1
zG(Ej|n5HD^yFOh-;eDLM9}~;n{_0w@mA-zz1Q*x69eR3jVi}cJxY^z7vR!)ZzAIbo
zIVgOvncrEnN+>RT8D&_(!*gYAb>LMF%3CP4cGh&)GWz*(e-sn{Q9)0+$&IhRH{yEM
z;LWm!tGnAetsxnfd1_{wZ5}Hhyo0rw+C`^-xC4dEjMlxBZ}(MxRr}C8uJN_A<_|8w
zYKr5NWvq#zvjucp1{Rwon2b#?^~hTv6&=<v@X+ux;->Nl87ZAJXAF_F#I_V73pM~4
zB<wynBY_}|=DVkYZ~y(>`+vBr$uAOB&qyDePE9KFyE%6A<fo-eml^<`L^CkeVr19`
zsH&tR%4e*1LFp_2AO;m!3=QwzKfrA?c3a}6b*n$jclO+7-t1yJ=OrE>&4|ljFKD$_
z6qH`O=7$?OC+a)vQc_Zc5eAx>nR#O^E<vrq-wEgLF@y?TDfyS<9+R9KLMW?i1@GPZ
zoXO7=<Mn3b7kxglq(4un+1XjcvrYB&Ouzwr(F`L-`}^x#!t*y4W%#n!tkmZ_s%CF*
zze_Ji6n3xDKc->qfI5ewVsFz+A+%NS(PpQmqrB3XBQ)N;-fHk`=g8{SjK;$qYBu9>
z@0VAq@$2YV7l~6PRX<!XH}`>G>RQf$hK%iDVPVv~7rD7oq`(>Lrm3u~Y&&D?I)sW(
zDKe}P9scd9n3*s7kGRjj6&vKX9@&o<Ei!47fR;DL3~e4d!gasCp;DV2&5xfsb7pjG
zjEeM!hN3%JbB;XN#bWmbNRc6u2dj76f`g(CheQ>1xYyKvKbvuy?a<x5i$uxIZGK!(
zAd6?ugk8-aLfvBQuWYbaxD9(GDpq5!w4VNce>__yJw3e@NjI$|UAl5ADs-2<j#kKa
z+|npH{PveM3EM(_g%~|E|2)mSX@IylrgKTqDIBR^y}F56=K|k>T;-N7=<s$%#`o}X
zVb4!+(PWMB1}Q&knB7uoEC1>fpbOT76>V?$T)~#G>G!`(*rssr;?$IN?z+Z5f5_WF
zsoiz&oxo|+F3FWgJv?;`0#{QW?WSWjzojBhx$*`hcpHv+wE`pfL5LB(dQxo4Zlmv8
zSH9A0QM0(%v0D9MtR-!>d=I1h5Yvls-T|@@-0w5vXE4st<cQfquId@P<+kW+*Pi#B
zfArV?eRuCus+ZNeC_0b#SD$b6l$1=CR@{&GysG@bvc!feP3D|=H~zIJe!(m>w}0Ny
z#aN)|(QSnzaU5OeHP3|)O`_>v{faO3`HMF-@3N(Y{6fc?_@&C^;lrD2DC@c86*Ac@
zt~%-IUD`G3uzf?=ugfV#N4b7A-z)0)IAae}SWh?C(q%Rs)IHIaJc<<N{cA;Qt{>$!
zs84NPcU+;(@$C@r!gsjw3rBcG{&El=k;vT@H+i=%TB3I6=|%-k9WAC4`jO6Jmv1ck
z1<z=+Wp0z9rG?bpRyqmK=%*W9U9(RZE1cp#&ey>)7tP2>PoG_%=OW?DBd;K^ukq^3
z(l7>zg=4QDO=0d)7#VBZ9jGA@Rd>_~<FRd#qdI=1blP}C{4KXkB}M+kMkDi*AveeB
zxV$x4FQnZp*W(e!?0j3}lP)`Yg`VD#?TX$V;bwLCuTuCDaZZUD1#T{0RgOugto57M
zTROVnjT%SnU$jl+dEqCFzWThQ8O&3auU~t=Y^})WkQ5!u+`gN)ad-lEog{r@vH5<w
zrMhQ0AP+UaZ+b^wfbW$@&)C@W5%vG_iTh0HpFX<Z?Uc%I(jv2mceYSPc%(nH%5J}6
zJuS)bhn|3N7fq<Re7?&!wS!ky_t-qW_{nQJnND_>%17FYBraC1R?kZ9$cdp1`AZCH
zrdswmrwCnYzGBF*W)1a>R|~6b{g8vQqN0)PotE`qs)sB5+ms`oh+{ykyn5ia+2<Ix
zgahFZrM!wp?%beXM*c;AncchM!}^CKC+^rKKjm|ZiaxMfAV#?6n)Gw;gNg^1zEIk$
zW_;Gca)!otGF7HoEu*0!l<h@Tb>OOxVV@eKm$B{)QWB*cwAknR`w=Ji*QGx;z{80?
zYCy+q`7@@rO2YWtgpcRD)xyTrL5v-&PeiHRN>tt5!dSlfZkthjA56;gKGQNbq;~Cz
z5$ra=zuG^}4@8y*r_NP-+cYFSTOTIBoRKQ8P!jf*i&amL+Dlg}G;*yxq)q!AuRht@
z4(llvReYh`;lwF_3sTv|%f3k74cEYZRC2M$42Ipo5ZmRH$*uAVe!tmlqI^@XUD}Cm
zJkkWg?iNWnvHf|;)oVr~jPte%TD=j{dzQusw)jkac8?Mi3Exu|E6;n>d6}B723r(5
z4OU4E<c+kSikwmh4L-l-yPCrLtC_b`Qzzuys~A7fy9)2%>Necc&GkM(p7-!yzBb>T
zHn+22Bz&NFx@Ejz;&Z9&CbV2pG0JFg0B8M(9;-&`MXg-sq~Eq0pV+cKKL5rW{?hHY
zgO1}ey}dsjml8ntDOSXEw}JdqTj%;(sY*<G8=!W({Paky&;34grc(Z>Z%eGXp4+QQ
z7*8AB4Sd>2$wx(D;;pjmx>ZbqzQA1g$yNpcSJ4F!vEN76ew%o~fSsBzlJqvGc?2@O
zRlG0|_1pWgcU;>}WztQhmvY<&+`J9--s~)hHQLcS5YZErN_9Zjf+M=@waLF<a#3Ha
zT#i)9o-&^(T=&i<T=!92H`e76p}BUUHJpXO`_dtt@b=bzZg%2UV$7zvOX+%nZ@8?Z
zC$6LwNDha_nLgO4;Joan#tYR=J3<pzR}zD*qmsL}ppEN3xZ#DZ(^M2tX)wr5@Lb8Z
z**8Sib#&TG?%rtmGXt|fUo))R{1NW9(eMuv{jK1?E@G2s8W?d}lcA4&gJy&5={TcA
zarsEQy3^@+I&CyyW5qCk>0hV0*!YLTM;@7sWadUGzPx9y&km`mTY9*mb8oxq{HSVd
zeEc>9#~k*}R?5;qtm6e&y)PY#P;oLjEsTfm@<Y4+#neeAH9myY)9nyuIQ6y`%5iZW
zM`3QA)?#Cmgz319J$rY>#&2Wt_p0C=aDDGNw#LW8%>QSl=-udjb&O$ebxvf`NpiJw
z&b4v2<1A<M0OJ&Opnh}ujZaj$kx1#KzJPJ~^Exjpu}R)n7gpY=mZ>1&HvQ6FiUZ>^
z3**7?%9Y)7KQ7PlPEAc6*4rfCf_DmY`p4$<qlL3gg(doP-}R}ecG3doCju5ib|7Lw
zIlDaaKDF$(3N<{2HwR6`J}zr1tJ!h)d3!X`-3G$anf64=uquAo*k4RTYUxyBQmPL@
zNc~VW5fIGjj7dmJx~=Ncz8=qI%K#pT?hox#_hl+kdGLV<w1ma<5o6(8Pa%h^t7|&y
z9a-owRQhuogdiV;BFYvP7RdnVYo^im>qvQ|oE$XMoX8{G9$8UYCg+qr-lUJ6)Yw-w
zwkCwR=Rp07M21SF{UDNPGH^!PR;yUS>@%Rpe%XI3{&P&>Txnv~7hh1&e6>ZvLTE+R
z#vaAST$<e`MMW3z(fm;=q2@V`v=e`_BOfUDQ+LwaDi#A1ZYPrim9MzgCdk8{zj#4K
zA|4YRy%m+)O=!d`(T&U~EdY3Tq*=Dw=29K!TRviU$1X{SFCn=JgdGk)LQNOmaWm{$
zCk5G~xHK@>2463uf%bCi-f_tt$EEt7PlQHJhD!83|KvaIjQCL)h+rG-$B!SK0|T5R
zjj&7n47!Rg8a6p5v8{aHctqm@SKk*0$yFSn25YAqC>;%#6yxHm09JE-b`M28Ffm+l
z0yo1nGq4c71q=G2qjLbUyz#pz4&wwA{cfYb*7f@030TV&j`hEj8CTpaEivVypr9B!
zbGqi6>%RT#FAPw-cOgA25N;2Pjvo}vAFXC)X7(l%-b-$#Z9V@cpSYgv)XwH+7NNYc
z1J56|tV?1_@h@M-08e_%zWK7knX-@$q1@hQhI406ojTQv>g9w3kol^b8qpt_U%q@H
zwFdT|8OZfwXiuD-!IF~NB&h4~C~?PQTwi1E_SsqP@kqb7DfvMk4hFr8R65oYyk4|w
ztlt5Oy1vf7K0h4x89Aly)du4s5P|)|BvLY>W@;PfL*@qkIH_f6`J49^cKq-j|5lMd
zlOIgucP?0G8ewg6;r8@Xo!srUWX7OQy>Dg&ML!nQ8*b@O+|>65gJCM9u>NmJrWvB`
z{{H@n9*pi2^}wTK5Dqhu#l6MC>pB^=VTYXeMpRtKYs*3z#1Iu<E3jUuDCKfuFjUpm
ziMLCJFZrHUH&<%=<IA4dfy}C0s|pF%gS}JvdL=?;44GF|6Un^9hUHQ`1oks++-+bS
zL*h?4D+2>VF>2+^f5g=24`u+rM#j56u4j?_BTwBJ=f)Ru^x(Nv5%K~*oxXzFcqwui
z8el&Ml48Tc!9PUILbIjE%f-1ZGrhxZbDwlvVk5;h4%a>bFitMjqN%AVHb39BIGej(
zT`$vIqu?<8+oYlA_Oz~DBz1QAnWXPWrVS{zG+43tg4(*VD{>KU>i5wH!p>1=j=P(t
z^`#1~rG;c+nFnuf<jTMxD1ipR@2#qonC@ro^q6^Q=4_u=aXlv|Ckte>_SD?|_m^L*
zog;(Dzf`LI)tQ0vakEF&%t=h&#~L|w*14NWzO4+LU_2m7!_MUdx^R(PeSLi?8miH%
zj|d5zSEAgE&cAWjQ?p5;aVMtZI(-ilX^cc6q+7>jC%?6A`+?LESu=eVpuYC^QHs)O
z$vTZv%mRoqP6Vf9@W^gYG~as?y~9uBVK*x(1q<(>%is;pUFVh54!jslHE)zU^ogX<
z{_5B}fwTL9GiNt30l37EURgHnn$G!E1Bs9GZ-^#obv$?_r)y}~FYQah5c~Wdy&;f6
zohK&Fv`%S0vVOl~Tk`V5Z%5uQ93Fb~(tN^S6Ymq2zWTNP*H8h;U05i_3&!(>B_x>9
zIKe;~t*o*79>xx{_afkZb?H9#ICAAN=Jd9Rl@YDT%z=luhs!$G5cJFVVvW}m-z_Wj
zwX}-oW@A4xeRYg?^oftor{_yeWPd}ZU`c@c49x+}|2PR~VhFmw6<Tcu=j>t*&igRQ
zHWj6%&+h2f_&iT8O;k1A_PIT!L<6@^P?9|r5{45Fz|DrxDgFL$T+D^!8$CD3jwm*u
zSM+1Y1OKr#QCWBkYcY0VVChXmo#w$`j^uCuwtr;95oKrPZ9UySA^Y|iRL3<ZJ<(LZ
zj$s2sJuezGkGDalDn%8Iq~SySg~R-~ed@N$7D1b=(ogjy;s^W>8+S6@PsCh?%_<{0
zoGVGAf#nd$1jWl_2d^=*t(M<c*2EgGq}yAQlNcYHAA6EzN+}+LUIq=e7v?K!7vQWv
zaKH{@h=Sg~`ibrb=}l;Po~&Oei?SBy0Lt(Sm+0;7KG)$cXk~nd_pcYWDHs=Kddq32
zM(t1e`R0$ax}>$p5VqU#q0BT_flDwh1XthsY1yK;Qhn_AD=LL>=#tJfm+$;m75h4h
zL5bvuoG9LIfa%U{@`PS})t`<+@lrA;H`fr0xfOg7N>M^$hijkpOvS*=aWX*#>>MeM
zxK&mJyt!AEHcJLMr`#W$DdN;ANDI`^^WYdxwa@1Vn|{gxgiu4Q%R?A8BWPr0k4Qlx
z^?LFxo#LCCDP1})jZz7_^z^dRTOSN0k??pe8kyB}c6NFbLSoJaiN}9C)(K&idw6&@
z$9U1{g0yEfiXNVaPuvkZaXYIrQB04{|KqUheN_)nk&|~0)?RI)l&been_j(?;<89I
z=<apV#1|t=B@$0E$ehWSWU5<hI$7`@=_|{Qw_yo?{7spW^^q~0+FIkw@n6;6d9r@2
zsA&Gp>%jxv3F+vAnOit@h6^{!pA9naYzQ^|@t`kOWJRT#$;xN!FZ%N6s76v&yl-IB
z9c9AeL{$VtC;#DDxWsy5cC!AYfK)(T_to~JkA(60v*_=wo|?Zg>&yuppu|rRq#>}u
zF-V6Ohra8TJ}f{oPNS<FG6m)=+!bc9F1S-Ze7RK9N8rs4;_z|EWb&oqo`ZQg`1w9*
zg~cn2hc7Fe`a0MV49wAUd5uprS^K=FrkX8QNUio|OuK}vX3xi8uIXHHFx{_M{7Sw*
z^VjgDRBx@lNasUMz2ya%lQPa7x+S@A@Kt2+b=AYi1Iq7q&?S}(NiI2DbKwqvy4u5Q
z|2R+Hak}uO-$DUdB`^Q@i@-2lP_uUU5(QgUHHzmI`Aq(=dy1nSr4$s_dj05On6;$y
zF-Umb)mtv6cyD_L@4AYRyQ@C3@(dqcE4usqrb)_sYGq8#jT?-Y3&Wi=CB&A%3t_#<
z-L(|+yA(8K@LNPlP+Ii+kx`pgRKEJPfsf*$W3+M2*Iz%A|B59}D}O^p51m`4>5GTT
zyj|+Ib;vI%=bC$8K{>mcvis82bK6LriFq&nio6=%hwR$7z`%*!UAFn0C_X&Dam_gf
zlTiB2;q2N~^s9Xh9_*uR4yXK!|4_SE9$^B^<y_FM3Xy?sABXW=o5E9C!PEP`Pd`cG
zZ_n;<v8UU%ThuuxO~~NNV0gnPRxT>wSwrBVSNnV<Z286|BL-b)evF`nRb*@FLwQ_U
zIzGE+^o!kK)PP{r;_?VyAiLj`Oicx6IJ#0^Qpaa$xvThgWIR#9c^s{h2SSEi<|q1h
z769`ufe}D#W~La}2vXg%U;y$|NcJR$$ho_bkhlQKp9fKg`2txV%2;NN-vZMqPomDg
z+@ZkTyf-)L+rI!mPmq3TcIgm$eMWP_W)!)$C2a7C^E6Rq>vwHmY_>&6GtJ-3jM>%g
zebA8Wd`+dj*6GjxG^*vZJFO-u_FVbpR-5g|J(do0-Mck0A{P)M-WP83>9LYa*Ad$E
z9LrR{xmm*0K<cx~)>cl?B{*d-UAp8C1<i?inX%7?R<D;Z6{DB8AuOayb}{-0*^SxW
z6fx>At_{wiL^&CWY#(#I_fBzi^jkqLUw%?*&Q5~0Zx6RUNwO|0kO+&Rwvy@2LBI-(
zl#7dmTQV{-N~uA#2FUrRPnsNdRPE0;?Nw=Ay*gVwgof#8u6F9wCvT4LV8APq|FU?@
zJ|@uyd618)YxgO`-(gW(TSdiUya7WL%qN9H*29g8RuB|=vgO{~TyscAk(ZUI_>YLG
z$$BFfhBivVeoITss-`A30_S4mI!Msl)6a~G#lUxEpL0a5RTvIB2Ei*f$gf^KF31dU
zN-W_|C@9$cSqLQzTk;&!t-rt5{_Wg;e4nRa@TO2P5Ybp1S|L5MAQL;RCw^!9cFAw4
zbt`a^#Qc1j#-=9aN4dzMyxh-bm>fg)LCOq4GNjeqzbqxZTCS(ZxX1M1(Xb%rzV4lN
z@O?kK%WxS(Pd96}lm;mmym5-A>{i6APR_9TR+q|{O{=J=uA!%VKjg90Gk@?vqvP*W
zIXQk?pR{~~Nc4?G>!PlmPtsBr75A5}{~RK<R!!KG*LL)$62qE;J<$r@ysDU8+Pv%U
zemGbj!gxX}*!hFWu`P>t`mSK#5TYUB!6?FVjV0j<udUiS6cIeS4Gt_@LKf26l3&*V
zZe2=AyiJZI>+Vls{9H8*-GXGwBu^bDII6Lw+3Akx`(2(7?EEX=9lg+rW#fIAarh3v
z&J?w}T|Gy$;dJEj%;dD`dk}vqL1OskC4#XgpRNnF1`U8>EVO{xw^v3tTbwP*5F0*Z
z#3BVh{?pt6D=P70HG&r3gz3oXa+B|tE?!>w_azF$vm<a%*!QC%d;!o#H?)Gc8y8U=
z&-+R9zGCj=CXqQ3)pk%{eGx~SSy<c#v#=PVQLnTvBr51hR}4Y>dQeXMU_6}t>qhK2
zpd-8$u4JrxOUoMl-&FHJCsIq+PUnKn8!ICtG;3FP99}{>9was#LP1UI?^8yr50LK|
zz#_#8R4S^_Y_(7cGzdO02)QG)M2|bb+-P~};yjEhGOjVF(m~eB>U~y*bui*K3FwOL
z{;n)>SWQ5g2tNNPz~M)a9XpneDbPmy>tqXih?4njA^#nx9v9<GY$UHoA>rVeGfmz-
zzo&Q>Dt}4wv<IhH3jb8mQ;_igR~0=QBrt2o?+~|`KT?+!AR<J4UfjuDgXkX(R}7#o
zBC{|H1jIik8XQ9n6%}`#&6B{rZz4syGND3gh=D6Y#4UAh@uQ~{GeaA*g0r(T3VHz&
z_K@YQo!8aUV!|@2Mf@;I1Xoi$)_o$@g8y;B?*HsiC8TY_HYUY!D%tg^sVB7-Q!aEL
zK{-z<Cx;7V&^4RivXB4#?$6C75?Womgt7y3Lg`b(<Z%ZM>#y?t(iHWTbgX<~OlJ>C
ztzDSIK99-7r37)4k86{-S=p2$EH0LZFY`Ez-Bbzdu2BZ#9uvji%Qx$n(<SWEY@_(>
zyvG<qm+()jeE+si`Oj}#zTs~%y@i(6Q~sC-Wc8<b7<P60Ab6BOiL$vqSjyk~t<hmE
zrpzO2$!wOCYF+VB0rwp4brL00k7r|VR$b_bu3VVTna|_Gy$+Db>)8t@J4Hj<TN^*W
z>=Ct<3k!irsP{#21FElgPENChxW%Ha@rD}_>h&6nEZXy30cWTDpA=#L`OW=DrQNT)
z|9&YBdI<E8q9*uH->keNVwu5H;o$?nc5Gd?=352*D^zm6J~`{K6|XZ>#bfEj(+3Zp
zk5FoO$;V_vzqr{kHC-5$UQfiba`*q?%`(5+xtZei&S<d-#sAp<w<i#lXltib{)k2R
z!pZ*3q@<dSCoIV*rN|fHbL^^{CGJS|_mr_KD;kN!C*AIjK}?sx`nQ`=C?Mf>twY<*
z=Oeor$qN(H(W$N|T8P4Hweuxo(yhK_lN#h(*dpSqp+^<P*@O)bD(dO=i#-&(c4y|<
z%4HNa4eTYMZ^#buAJ0n}&;K;+0^{NJ@n4k*&;``77#JzbhYufs-fTh<^cElwe<HF?
zy^T2Qe!^iKt+aNO@t@a@fy+`T#s@rP4}dj!6m;2%tnBQ4kXtD0jYGIyvV=+Z7c|ZF
zLlKuLLlLKX`-}gItnS1l1LV0@)hgRdpiX0kWK4mijj^d^CK-3ZwBU+KOA~?;Pp$Xl
zKF|5d$oX3t6_QCc##er|E))W<eW%?D<CMZe{WDsg?`I2>bkzID70=jTw7Y>n@q#gr
z2(v34e2<AI?o&^I)g!J@RGeu88XZzkQuI*M2$nR^B827afv^mHxa_v4u*CX6^S$A3
z1?uACY0}w3-%DklUi7@OIXWdpz{%O!5KK#ys$X}awu?Q_H;Z`Y@SKq<qfnN+sVMGO
zFOs|5oF$A5`&6$8co>XW?bY=4Ie`V;hGlZ;t5@>P@X!~&+>)oAZEd_u`?oEi^RLsN
zc*A6X`04}>1;N5ZdiX7OydWFJgNre$w6djzqpz=zyMoYji{#QkLv=7Kb+EIw{T44i
zTq?6ewf~5*&HVd&(<j}U@4Yu6){Z>8iAgRpkR`)t_PSekG<t0bDCkU_Q<Yu@sBA0e
zk15Mdoe<r-w{&2xe_&Jkn?|d=lm_?Vdy&Q1512p?u!7$GOwN<#{`RV#(g<@;Xla=~
zc4GNdcYm#=o?~^3#&)_KdBpW>(3SbLw6vfSQYAbeI~glC6-%~QA8oxp=EktuTiYYN
z!B{9E<O1+~xKkZbY#}~JL;CYS^{vVTUKT@C(X~$F3>(!(6)Yb2*`Hb9=C)&l72JQ{
zH~2-Wv{}#ldt<3iD=L>fJ746P)#k~@#ntY~cFDVNI=7H*B<Heg&zT|MVkJ;vpDrIZ
zr#ieak()Uy05DAaa?9q&OE*+F^s6WeUUY?bgg-`l_F@f@vyC~!Q7QbYjk(?1<eA<c
z5z61t@yO;lMo90+xr=$|Nfet3-!Zjjr=FA{Q};2`62UYRUsH>*qO_-JX#pf=M{}gw
zn7Y$PaNz|I3zZTiQ_KCQ)@?nGThHyj??0mz@!Y<)OyuhJuQxbn&XHCMu@z_y?^J9B
zRsbMOznp&XR0GDw#>_5V64drnyMX%LhQj&j!pp8dMPgEJys2!_5Tu)0EDbS~vg?4%
zfTJp`5B35wdGtBg{_-QZ6~(4Hf=WaqPm@MY1`$T->l@2x0P19ZFdwJ}38@4Z<`ppB
z+pr$7jN#c&U_*hldNmJ0gfNta2EjUe@S4e!cYs`#qzjCULwNpF5RJ&VOh5Pfx<T*}
zv!0#T@HQ?CB!1qqWN4ZD;IbRwkg#mVA6@XWzFI=2P~KBG8ZWduRXaU~IleXsMLLT?
zpw;X%uh+Wmo`6As(ctw}g3K=X9cA4N@RMG*PNlT6E}R2&;IEMEobowoL3QYV*XA}E
z?Cv<&imJ?Q?X5fh^gx+A7fqI%%C1MxRHH{|ML;-QC8n|?q2AsTKlT=~!0GAKt)wJ&
zugf1~{}L>Dy)<cEh?8beOESru>P{kTGNRKxo4$zcqfE3l2HB0@?=CoeIpHsHK+f<l
zsY|{^0E7)e<0*0ntA`(imYCF3K`=On+Irmj>fHqJ30PftlrZ}0%^G~ODi%u!H1s+#
z-E>scol3XM7fBFolq9lkg%H!RFfw8Rj(xF57O|PuLP=BAZTOy5ljX~ePjx(QeQJ3<
zw4(CLBEh`DHxgJUX$Zijf7lUZesp}C=_zn#93r*YjM)pST*C>gAIh@BcNqulu3n{r
zO5I>MH-`YR9j&?qjc!pKfgjV7gzV_lR(Nt;@c;g-Oz^h1cbaRTMQlr%kqBAkJ|p`S
zz9z3!i68E|<|!?qlXetWHuH6voHvwDTM4SSPJQVe{YI%@qP7a90(1Wiz3@yaxB&^{
z-QC^os&n5|7fOeuu)MgI?pwgvn#E>sA68UUd`96nWFQ<5zzWtIR8%HTe_N=?DW@c{
zVL0rvV}w062>|~cb?iUi9RC~p=>Hu0KTqKQohR^mewnP0;K5I#0?$EOrl=?#S4=%{
H?)v`%N(v2P

diff --git a/examples_skyline/accuracy_vs_fpr-skyline-order.png b/examples_skyline/accuracy_vs_fpr-skyline-order.png
deleted file mode 100644
index ba45fb6549543282de4c380b51f379ea94c80510..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 43619
zcmeFZ1yq(@yDfYJDi$^d0v4!%fGDA~iGYMONT`Sc(hY)OU{Ojdse+`SBAu@RN(hqD
zA|)UoEq&(VeZSiK>~r@1$N0xNXN>)h@qNJKbJx1px~@6rHLrC~QC^B+In#2Aq8Ox4
z9ao|#+6szV9JzD}KC!Q~slty%)<>mPmf}C>r5Eqw?`0OJG_5IW#X9mon#$&DGx+e3
z%?S+~W%Elm_UEk(DAV&c7AEF4CPo*w+Zk9{8=0FO<T=2@w{N?ljg5scFYiD8oyXkj
zGVilt;RcG@PDvj>qT&$xv)NHw&0ubRNXX;q(i>-a5}(|bp*7YpW;f{d)JwH#`t<tI
z_wLj6)o+SC^;bmuMJta7KjmW?+Zbx47h@AueflzUe}eBN_U%7{FYw$=H>n$3o!8>X
zwUqvPNavWdsB=?jpLTa>xl5^s{bk#Og_~(`_pxj2)i>cMRn5N05I+iNY5DQv91RT<
ze!O&}81O?w(v6JDzU`89__1+KBsp@$Qj8bnw)p>!!+2ggPg*Try!ca5kyN(*U{YFt
z)br<@`}bE2TU9-mtD2o?<&`cjE%o>D@wtEhetcFA=UOJFc(1KOSy?M7hNXB~?QcCX
z5w)R*t^Gqn5?IBYGO|LiEMK)s&CD!GR8&-3XE{~99M5X&&3E#{hYv|S?O7KuUfjEL
z=ToJUjT<+rSqj)5Q*|A@owbcdTY{F~O(sO#B}>@u=UzQMJ=?xoxso7W`_k!eH&<7u
zUQsW~nV;#F@QvCf<!#vU@u6|mb1$t_<LDyRtEYWO*y()z!^3%mgoLC%-t+QOJ$m%U
z)L7phb8~Z-`PqrbLgwN6ThE_A?}R_Xde3HBML&J^tR~0N+V<zyGg}2sDx|i#MCs-_
z9Ui;d_afb9Fn1zwQq2D6*C%%U4Tp8!l5sk5@}!Kf-9S^^S+mR&Wg$Z9jcKN1*{!ol
zrM5ab4qq$}A6HTFC-)xLao|>+o~urCmaVeKM()vyW2--zrOQtB9Pr$7@KUT!ww=?D
zCwG#x2D5DYs$ZNHnw*;A+`e7%;N{QzY^TlhXVdWcbMCWgv8Ih_@~uYouMIHxS;mhl
z=XmCR-I0jZ%fB`}HdYgP<fhGVhqTl@Z~lyu%k+3aet!Pbr%(MKKi2A+dl03XpvJLt
z=h3;jx%*yTHMpPt;7xdWxP!e14k(p97EnHa{>erj?brto)}$KO$lSYkPri?nhsSQl
zzCK=+%WKQQlQHsPR#U9j-Jex0EM5-eO%GhDT}&AqpcJKD=cilc?CjDwI5`h+aBvhh
zHa0SgIaSZh4!R1l2;1xzu^-^a`|hfEa%{JlSo4OB8y|oCcxvSRdWGPoCT;7sw~JE@
zN;c|y^105mZ<Ev34%KhIvwEAL@985)uJaWJ&&^JUL`H5;F>e)DR#ZGUJTij&y;7Rq
zdhN=64-YwG<9MTnH*xi;SAy_m77@GPuCkC21@X!E8yFcMB_{408LQLVn%!)s^~Ugg
zWo0FE<9%;$!ELTHLjw6%G&F)IJNy)wtDc?oI*^<{mxWU_nq`;m(|B31S-&*Xy1R6|
zzp<p*-naD0u{+FFJ9g~Yy>sW=i<d4v4h?0lN;lW@d+^|#C*P%!uCl^+@7{?{|4b<~
zOApD($w{cKOVSY;ne2|(s_%K&w%1=@!lWgqe$k>uoAo^%d!mC*`>=|Yu9ld=;m@zs
zC0)V{@wRQUZ*{rURr;uKb9cTE-tpVFZ%hBUNiV0X8{X5JpYJ%3+3m?!c+tS%Q6Tr(
z$EjE99zT7$bL7*Wh+zCJoaPvPlYTW5=J~=$Gc(Qy^$S;4z0#Mm9{PTw=-R#e_YXXJ
z^k_FPZ~4#G{8s)ewWZGE^_R&V?}>1(bZ~G8R*;yj#*_c7Pbqtm-eT;{eE7moyX<>D
zK0@t<Ue;5r>-Wg>XJ%$b4@gn5OR-2f*E2F2s3mHgU&6q2@ltuXOU(>#-tV_>%$gh4
zzUNVYNX_1nm}z_5m_N~Sm8H=O=TQ(HIvzK>TOr&;*Lmm!Ui<WT|3S=5+=ZTX^Ri{j
zx}Mx&Jy5vWdi<wmdw;)aWmT0IUtyDF5zD3C>bS1=_tw{G7j??4sBoPu(POT{&<S8I
z7Y6&TX^K6w3)9K{K;5Pd8!on79Tc<fD%}z@j`bO`ibXh>JmY9zU3sXuO9P+z)M(G8
zx3`uRZq~Q5vO0|rSN7$Zlowwo{SS3r#}BW<@h0k4vWk9v^?7<Qzcudmk()~%O^$R2
zX-$;y=9`@KVeJp*!Hg#}JVeOc?q?8Yi*Q%|Xr=6>pI={j_h4>U;fAE#1}dXH>N6~#
z>Ccf1#KgpK?A>eO>@%1*om4uzlw$mYotfe^{o}@BOw2&6+nL$fR800TQAZ^yZ)QvF
z#}=80a(mg>j)#Yb2Q=s^Dn6(hMLe@@G)<2)C<z>%n5aXPU=CX6uX>YB?Dw~uS~(7{
zJo&UTES?m4ZrR1ln|#IWDjroU&Eye}ZcZ$YiMw1LzI@rThiqkGBI>QqBVJiHJx_y!
zH^0zK>B`F2J9X-=H>+50uo=SYV}7HZ(oX%Ubsg>PyS8oHCKoQM>HEf}N9o1cba^a@
zUd+oh^VR{4@zI_t85x<%6}&myB_t%Irjas|YZBD^W+q3brd#KeL<TaB^*5#`E57J&
zNVT5%+28*HgCxyfXXiTlMap}AH+gc~J~b@j(`VC6EUR65rUsu$`3R3cW6R9W=GwdW
zRDFGY^gs~#Oit)Z?dca<saB^iU%Y7e{_a|DQ~Uk~RaRD3;cGJvaeZ`d#j#E1d=!K3
z$&<HP9Xr;VHoTE`nHqg~=gu93djroi7)OM$?1aE&%a<!%y7YYciWSMrzmQ;v)uum{
zKR-LDT9TcadH7nM!_d=fLmqmDcu859$TfJc&6pnt3J;~JYBBiQkT+r3UfA&3z+)q?
zZsG@zO+nKx*@)bTr!VyKbv-um*4|pSo_R#reju@<<09jxO;-@gnFNgeG5M+!HB+qW
zu9CNW=-TJfQk6TaSY&1gu#TRRB@$zxf8oL#t4<ll4I2zUhaP4+c*#@Wg<I=Q#nTfW
z{Fh2MbL-`4&rXf`u`OP-C{`gtoV;fe%=PA{>H>SY<?x`p<wIjvi~o+So(#G%&W<@-
z(UN<WrK)f<%VDeIG&D5cX(jnru@J+O9)2z<QI?nYt{Rt^`E{(wK1bWhX-bt0=eMQE
z$%kF%a+;c&s-Mekm9ETj9C?G8DqZ<*wM62hgO`mF@(vXKXf)4P#h^P)47@<}>_Z5w
z!@@C0Z%j396qHu7v3VmtKVxlh=~8d$qk~SfV|B@yvxrqloM-To9%6;mChH}W5Y9KR
zo<9|UfF2_kA{dvEp@H`=Z1K%ae~xaoXe3GYUQ7a&{pqc`PX<NLzg;vtS?*ff9pRdU
zl_t`#iCZfQ@#<r9vku~muZ(w;*4RMyASV}>EV7x>g$qv$y;;sa=|oB~NU66$tPn7+
zK80wJZ#@0#)vMR_CQOH|qLAf#3|+rnryq@xn3cXdF;MgI;V$kv=V)1})cGMF*Vx$D
zQ+C-padC0Q#l@Vd^TB-jZ|wTD$aDjmavCWU>g?={&U2X#Y;diPJ)@O!$t%sQIUWxz
z7baZOb8nyD9tAZZlmiD2uu1Qc53MK;;7XppN>k7Yc*q%jHr>prH!h|+MqwL<PYu_%
zWggoq=JaOWx^?_`WB`U<OyjCJ<(Q7PwkKr0k?}{8rlq4xz4&RJbY;PHIsv<%8o(^^
z>dCrH!Ztqy(g*wc5|BdA>KA&R0WNvtFm!=on|KpnzWrbe0Z8G_!_t$JlX0%|bKZ*V
z(ip~SJVI#r%$e|^_M*O&a`7sxfETqd&Wbb$nS71Bdw1#N@83Vt^JiZuDJfOIF{)ag
zJmI@+{cgjG$Rk);%0=ERV+aCyZ%cTzGZddZd1Bm>lSyK9y-8Za)Jo<<ircRJesY^(
zYnKCO@!hq%n1jR|u3kNxG%0iB$Pq1IG~KKH0;xt-C-J9KyO&V0^Ure*4jCKqjyaT{
z#{wW~<vOLRCF{N@^x7KKAOW~TBJA$d!I7AR2`4-@%Jbfz>^FB!XnS|(3T}$PCe!rh
z=ZH4i6BD#<C7F+8yG3XsX)-O^ZkT1XEt&0>m_O6~IZQ{#`{=P_V|Dp+suNAt3SQ>f
z`bFLZJL<ePrD8n}(K19HnJUqnL;MWZS{tro)KeL4Wo_O5=p#sg&{|qU3mk6X`Q<(%
zS7^hoojXH$Mlw1O8`S}IqD{Y=2+rouPs}G7uU@w73&73`Ms2wZ`uYKZfzRDnY*Ooe
zY)To%X51pY;icQLRsR<;hW{sM&41tI|C@C5Kl|Q#6Q%uj_e|u^=cbv|KmGN~LOqC%
z;u#x>nAdilZfw=oIT;POUJO!u;>3w0p2e|iX!(0{2o#}T&1z%2rN#`{J*zJ%yQ8~1
zrlO+4YT6u;TPMrrVS0M{Kvo}*!NrRzt`8~gOGvR!zy6?4zQXbsAJ#j;7{JZXud-Fh
z%;q#bJw2C@kUH>3O`d5a=bFef$&4E}W@X)y;dn7{sqAr2Q2tDGi$lAYRgi?^aMVd}
z<~ZQY6G%jr-04kEU+EXANySmdr*v4#-KOqRUc&kl1I@klDKXA2E+l%__uEM;f$Siv
zv&<r(2mslWh_0D(iF4n+>L9&qaR|`Gt*v@z&YY2vld~Q50o5CHoxeKP->5DwF3u!q
z5)81hVbi7<Bvh{5yHED?^mxzdIS;vCxNw0yJ@B8)(1%UFQS7pTKLXF1S;jloCTcQ?
zJ7=)<eE;C#G}RNs$jqFWX4+^x-0>+3VY12|+3t2j>Xn!~tWH1T2V1l4&E-N5odF>4
z97q7pB&-u?P0eq&9KUG;zsp#SI`X&V2RlJGt+wJJlfK#)b<8DmQ+4^m&k5eJ1%xBW
z=Gz}P!#*GFZ_U@EqhF;`63AVXW_mVvyx~g2%kwu$Cg9e5eH;%($nYgoDIeCFUv2Md
zUK_~T_Et-8bN+Q*zXd6X-?)0O7mJ8mUS1v<$-JpvF6l}Th=50r;?BRjO_UDTsj7&-
zw1T2}HvjYiKK;TU*XdawBqt~D+_`fU12~;Ee(>BQnqD`cxX0hVog1k=YyJr7d*qAM
zwk-+@3WNd*nzxwH($X@!96NgSG^qcl(o%j$N5|b<TqSQz>MIj<vTF$)C76c%J@f0`
zY7md9i9vIS1crID<7cg>MvbZ<Y*b}hUmk&u^7zIL+Jlzgj~_IuWCKY$eE4u-YilbZ
zrW-bF2nNF3>gwuh4bE>pH{;me-mafF*|k*)Q_0FhT3%kTaPtDe##255QzcyLpv@1(
zk(rTl(9!xF$CRO8zxJ1lPba==><1rAE_E3q7k%*H!IgpJqHV5Cr^7|-<G4X~oo9xM
z7+ZTN?SbK;#GU%}^xK>X<Na`-@j3F~Zf@=}azlhO0sj|)JJzRMmdm#5F9o<OLk1&%
zmtlGudGdAnxXwJg=i!l{XlmbNd1rDg+0_(i;GErHi>X$+nW>;@Lx+YB@ZaQ*XFjK|
zUCVPsqCahD82hQb+}=5vaMM>w@9wPLD*Ze%GV*BO`0(%n!vEae-MfG?h0<G``o;!~
z3)Xo{K9m1FT*9lWs2I5?+~FueiLqy%ZCJlv`O-X|P8%S?7^(q54?B+CHvITthn2m3
z9e|Q;33s|7q2hC^#YP_x1qW1S7E+Ts0E}(9*u=-GDrL;*+JWY5c4=H(6$n89@5#~0
z$#~}XP%p?HKnOQDc=Of4Tp4-!KsLbbQ}nBtD=<YtJCu}s5e}lk*Nf}w)CtH3U}6@?
z$j<JI4$}1v3JQ8GXzHXhO`^%QYuDsW@qeHVO|~yKNtR{vhpS%Pm)(&&(NYDT6`7=y
zU5pzdXBq!l|IE{KB^caE(5otC1+nPea*sda8W@&@=*OV)Paz^1ff5>5M%@Dk8pABS
z0!rg8p6fJ5#s~0p*X)3OzF)Ih9p0EGdmm<Iqrk<FYaAUM48Om>=l!fvQf>zZ_mI<Q
zEarPhS67$eie<|<Qd~hY+PiUm7f?AcQu~?uEZYQc7Ll$WRam-3<pA)x#==5Eb!UnK
z8X1-?I|Idt%?Pa2o5}_FBy?(ce7yGF`aS+ZK`}&qGyYv)e(gDs=b>5z+^qsemB)QY
z8?GdiWs9Z%Sk%$#gMIb<(9qBqOtG+r&*5TRkP1usXuoL%6{g(xGzfyexBy9?=^=z*
z>~T#p47e>_bvZ;0TjrG+9mB7F4QECB(k;4v{kpKzs9`{2He@;Ph}_&<aF)!I%mVij
zj-NnPEB^dBkgxDd;G;*Hm*z`u+WkeWs$vvG=0YW0b;~8@b2^2AuPth-_8mBQka5kL
z?b1w#EF$GY4?PjPI==VtVa@fsPd{KwHEY(wfYbo)u?RpgEnHigoH+Smlb#B+Qeek1
z0AtOhgIIk%cT2&Y#xOhxrEiC`sEpd^SAP_1Y}~LRYNFM(HEs|~tQtZTxJD0ziu|s$
z;;JXd=);6<i+d7lJN2KaSg}p|e*b>Kz`!8x*qv3KW-&YamS}sCM*>$I9v!U)UwDCN
z1C{YKw7SHK*En%LLDEE!<K*P@Z8lR!=6U`4^}aejOa>Ql8X_?O))Rse9UX1-qvDBg
z)JZQUqtPE<oVp%ew&|_rU;w69`t&78iq-Kr0$<d%-);2LN_YixU`;9xjnD*NH}-Sx
z-TQ#=BnTAPM)Wuf--TO%>oL)?LB(e2eVsviY8Nj?L7ccSDMA;^lIUpwwp81foS!s1
zJt6B0HvG&-eCm`hKY|;P4N@+b*CICB4m2I^|N82f(x>{2(NP=FZM~F0h=?Q=^5*=y
z0ZD6a!gXF9a;B}*q86+oMCAj>*Tz^y!^6XsAO>Au<u!NY{e7%J?+LH&DvwCU(l!3%
z%MNmQ(LIo1`SMCIwlPf1IK0y3_;agy&Y1z;yOnRCq+Y2_{1RfG7pL|@t4G0g$`1+k
z86izmr-Sz&0dT1P@#R@360u<f5~;9lZxrN5IZ)6#v56-Ap2?aB*SSo@gLp)#y$25}
zlSu=umw5y{;$liYTY&be*N|e^+1O&ia2j6e->Dj_rFvPp^y|T)Mw_h5jhZqn3Dq*o
zo4iD%2;UAcKSWZM1odP!gz0MJ8`G8?O<#UYeQ9=mo$_$8WK9JttJe_TYA}<$HuD|j
z%^mfH^vGH?gP5i+8@O+QGg`DAC36PB)t4>z*V`3N$nKqkgUL_=v~-2O8Jl$tTYsmh
zNcpUh&mtt)^dYfY*mB@`9+Q-ZME|iSnyR_63w4lcOXSO`x#>Y$#QdzR4UtE-O9owd
zcbj4P>ecE%N&L`m61s@iQ{YB(-_ujY#Ka^zDS9!buAi6y>exmPHB=t6qK?P`(S3QN
z{E=MqCiSnQWzy-Y1+?E<x<e{$wyBbv8m&@b+`Ks!aEE=*o&*qVZ*daQp-+#Yy3iXj
zKSK~w3>H?_OJnNg?N_TPPM=*B>R&UhW5_Fd`EtL2S(63?(JH+8u&Jo%=pN__fo#pT
zbvo*0^llB_3x`lH?mvH;#N5fJU$R74EJw!|iaeQT(v`!VC1i?rc63x5hEWz`oBOE9
zVbPMo8(d|fhn0QA#)HW`6B*1=hBh<&>sLm+9?}N>W0Rhq7*w~gu#gMViUE`>X%3r@
z$=*2run*_@(ecSIiRanL{k5vw=D34=TspKEj1m)8(vC7BR)@BP`v(Nr)-?JcKz%#J
z>4zc>S!S)R^YOMHii)_z#dUo*u(F!<Jj!P{gY^y)s-tuAMWMJ=E~*?KYS>kl6$;mV
zZ?&5IDwU+FSD4lNcD_Mp2Y@8>s!FixfQX1>fBy|cM~Ex`<hO@qb&@O8oiT9ZO2%kf
z$!NY>8p+PJk;&hBdg2aScNzZr_D63+Y63FSGw279Lr|x9;j1$2`k}3@Evx5wh|SDz
zOU&(zjK9BzqOUb;<gWz!RM~dwZ_8{U(3#X9@YuB%5)(H6_1K=ab%g>IZitDCA<7_r
z*?$^`#59UhK=#E(206gd4VVE4Seck<g_%x?ewzo4m+R|vqB6GvfsQPaiV8(2-8H|*
z%fm$W=;^gGwVEB%_dEk_3}0{%IKafF?qj>IT}SR*VuEPW^(FgWHuf9W#0QjIdlGpm
zBQrA&Ps1ZNwIpMmzRtS>4Ryywh6OvXJ>g$DBlhe`d#mKTf>vu>@49edS(%)I*qFb*
z6O@0lI6B9EHfO7U`SJyduo^NKSrVMW!dgUrU{ss-tFBgIU=ddF<#(NP790PnUzhtC
z@{chlq`_ysR~Ot-=HP7FCfWZ5HLb#|`F9%X(-sW;ibW2Nj--O*)aJga8j7(@at!38
zhLp=bY(|w)(h!Eoh^VR>)ZRt*pL{%`o~ZEzb&q%~5ysxtk>lqQr<dvL^uJqc7|B#`
zqGTA`F`zqal%g{)EG8z#xNTc9$cT;rub7w)vU&Q>UAqL%6;ROmWH=|Ihkm#FB$&76
zA!$?*1lBdBudmPI!y^nb1UkbWz%0E5YTLE(l!!QlWhc}nSW29SKkezQit%w8{}~5K
zk3b^dH|O6ivF)imfqZI{^JPE_n2)>Jd890BvMWU3>bQy2G*@yCA%e}<ru&&BuH{_%
z9NP0+Hz==Uq+BA|tT~GVB`*ZPH+jFmuYiCoQ!YB<Nzf2-?BTI_$UH=>L)w$cy*g0~
zQNQ@bYRR6>eVhB3n7k$FYTjS+``#X%ljGat>FH}EecvNH&sJjU$1}tv8>X-9M#cUd
zgeES866m7U?LHDoj9hA`e1STiLo@FM$^@r!i1sc(m4%dA1A+P8zI!KNP`p9Fv_T~~
zZz^8UtSJ^Kk$^QP)F4NHrj++Slw$oNF+Zt5C=QT!ZdVor?`X%7E+arw$gZFzs`GQR
zsu~)xSl7J?$$3hsppl^?Xo@?;&+l1hNl+gy4}?+GQx!vEHI{9V(;E&62|b*$#+O|d
z82K5pXuMI?^S}vus!+Hv>-N1}6_Ntdr+)WrDrR?dWbdy_z*U`yBs$(`E`nl3p3VZS
zA=D3Z#kMpX(p$&CKoUl~=l4*1K*N9=4IQDFK%}wwM)Ajw6@;<AdBZ=}SEqoPKGoPd
z_Z;w%<&cG>P>aJeAl`kb-!{9>&(>*AAY%$#c)wHtl0RXxm?wnn1I_jTD5+vKll6N3
z-BOxkGJo>|+}26`8U6U$Sb`pz@_?3kU1_UWMPq?32=`U`G&efh;Xtdn#naZ)mSHUo
zQ|#_}`;)KU_N$2vy<h)W#6IbBgoNIcIbB^{?}+3vS|u<&;0W1DRUg4DXBO~$&~AUe
zli;V-;bK?cTrOuNKxm*TQ!TI9b_B5MbC?M9<h=xtOH9|78{14v9xK#Px8B@myqcC@
zQSQdg7pZ3nXGM9n(1;pDWduxsO<FKSj<BQ7vD>$A6Gh$ErmT;iH?u#j@eTBIm+Ag=
z->8x9@;<<dGeJB$V6RCSZh2q-&`{^HF;XbgLJR<ku>mtAwE=QA)Qe)E4VRdulSKqn
zl4)Fz;!D_J>yyWBul#}v#AuDWp0qF22HD&eUX<%GsjBkl#(1UKc}@22+qZ8bn>K?<
zpG9Bi*m1Wd@zNKb+H}?nm=F4n#}8L4EA$}uc=y=1I7&fIQ3hrs1qjU6^1!-D<l8=8
z*GUhj(I0JZ23~`hpTzywVPY_D*%Bx2GW7#<l5i3vxKWAu5eY)pi2@tbNz!0Jfn-s{
zR<n#Fc+X0MExEQ>5WQ`8wyB|jN@gpjFc+3IxmZo^ReRvb!!P=pGObBGLkzOTfZ>Q1
zUp9CZ<jr%2Zj<}?-ML#Df7fEh^JCpX8#4M<a6RV~70X7Z{E|<RC!2s3l6aofHi9iD
z$nI6w(s~>{=dp3i`gIpPL(YEcetM!VeHKeaw{>}MEJD^6)Ok@I>8VKwLN!|5%Ow3x
z{veiK&BwGp&cdwwasBOK2}Q#COkEB|8tV2I?1vS-zJE`e=dn2j4kIHgOK8{4)#9<3
zPFY!ORM7dv1SU^gTY4HM$rtJ1Na{fpRpb3CY2X{v|I%pE`v8>xgp?E)QU<7?vZSP&
zzOaNuzD#-m<_e~brZ0>E@rdlY6T+>K;JGzkMwU*Zf-&{ogO_BES${vjb_WANeHP<g
z;cscGcOXty;<nhN!NmOF013Y~Wwc3M)e>ktuWJVR2;zS%;s8k#*GLkCNVFSdtWdaG
zhWYxE_=l#+a!*o@b<X)_AH961IIT{tvoV<8$hYc>p^LLKTfoW+6kJF-jMO1BG7fFg
zXOjjvRYE+&tmS2371h9ew0;8-4>kiC%y+ZQMtOO8HHc)kpd)>m-Qm`6kf_y#nr-|@
z<%zHk0;HmRD<rJDPD@u3mh)HiN1{}Aa>^pr2bB9{5q?PJPI(@F3pM0eAK<j=l*<Y+
z>83Omd?9kn83LM`WyMWNq!`;O6?xtGLgN+XnO=uZ#tv!U1D`MMy=I%k#*uWjEXVR3
zZm1#OH6Iu&^4jc_5qu$8W3^{VY=5Dhsf4P~7fd<-2M;1q`d1_x19JW|g7lDvN(Jg%
zMt__E^tXnl76iw_6I7jzAhq;1XQ%5rb~2Kh#6w6^FkqNr&g!JjKy<PF{p&oya}Rzc
zwP38oFmI@BZ$5pI;Ng)cr1Yfc<{jD1N#DPJk8jqbIE(&FVZN{Aw`^grQu4~xaeuW+
zz>8u~@w+9r%=?$KwC<7cb%N~G8g7s|(jhsrOJk5^6cmC`pC<h5Ivu@?FT^(#g|3E)
zyJ*EIMAU}aw{S>5K4ke6)jPeeta+$A9D;&sq@<^E;D_tnWVde=2rrQZeWQ@gNdU_<
zo`e?!Ot6R4uv{jZ?aKx&fs73<U%rBxu5T0)hDJRD6|a2stwP(RFVgNf)A9;W+gPJT
zhLMjihJSift*eQ;xc9CNIE%c{S74mO*T#SSQiP+$lkfiByA`<Kuv*A#J)AERM4{N!
zkOCoQ6&D}h8Q`B6U;61{cRn)kG%`q4UxyjL9#}43`la5>=<(v^3=B13hiuXSrpTs$
z2u!1V+y@EKTaOdqlDugYtw|z9W|#3b>C}Gr6&k1u&<g=J7$HREL_9UgIT0@7kFL!O
zNfDlj8j(8pS#9k^6n%drWe-%s4n!)Don2i-7J{gh9iNXEO^SFIJ}j561gj0O?_A4E
zC9Tyi3|3RS%Fb8&L~sWH37e(l)10l2BB_o-!)o8%-g7p8-sL0u6yn}#NWl$xF4^$Q
z*p<zZa0ld-*d2aV9{&JrX(uXg)RN=q1j&Yxg{41ZS+_3;CI>uULwf+?plj6W*f&SV
z#_k>Kt&v6HI~hD;6fq(mRl?r7<Ol$_&hGB+H%WwAAyxOb<YptfJ|O@PC08Oj2lDC-
zWW4IKTCs^|L1@71CM+Dqb=QP*gESdt_wM<ok7(>6nS}Gb#vQE>|54<fvwyhk5#xoX
z=b4WlXzsZ5Wu7N1sXqcRFe@;b0|N#>9x({rUd3VzhTir4K4;9I0XQ?#H|yrBqzlH=
z4wr2c5B_rq@7vnv8Mn0^S?%2i0f3zQeXQ#GLMv-`6Nw*%L<m3vs9;)|hten(efUuG
zmEmS8An>~rZc7w9dBn;$P`w8MkUXyZ&Cus|qU-)vc}0n#3D4W!u5=7^uH^y8+?MP!
zM0$hzzS*Mpu)ZgLakxjrv=2vKf0Z<xv4rX6_Wk=6H8nMbEx+HAO*2n#-15Wt6!IOj
zxgZ@A(_cIT`GA?0|J)Inm23z}Hkn4q?)~!YPD$G8$nzFs8hiL!p4hdoBQ?VRS6ztz
z*|HBF@&B~L(rt<Hzmq56X~1GuBFF^L21V)2NEs^D1(o~?ijk4A2PK0VKrvF>g$)FN
z@$i@<R1*QXIQm<1&%!=8Rc3yz{i!icq#f3_B;6b9Zg~RuQBg5&%6Nq-856*zE<9!l
z2oKehUHbCp&@<rQ+GjYaYT@uC9*R;Nd7Y91uSiGms+B9N!8$^xPKOA}-CV*@i9GWR
z(CZ2kpn%S9s#*qiOvyN3KR<}8Y+_^8$|PD4uL#7>PtDDf9R`1s9VF&{KT}dx)=@NV
z_6-RsxyDI9y)<<TOrK1VZf+D@B=7mc0s^A(dyF`+Pz?(CYzuZ9U>ZmPw(l5(4vl&E
z_^3NOBUJ)kBH!1XbryT@_;K-HC#xgRO8^~I${uLt#pUJdGBWpXuiiFLz30#&jSxXo
zUp5yA3Q!G+-;=9&v744V?6nLJP%<EC6i&GW1bAYQLG6a$wpWGlJKxc+5c6IDD@7D$
z3VpUELr^U#D=YgDK<CzHxs0CTS0oVl!H#304Z7DR4T{R4|3kEHG|t*b#g3uAkID95
zG4N(>fQeKQ<^&=ZfoJInNI<%?Z?+90u@M0fx#II;Ra&f4#-$bNdpS8Pp<>2^32||A
z%YFLv30Tm&cLWU1xHOorA@^!Jc)@5$>rCN^8XOe)91_aQ|H8bWC~Y@&N4*^)B4<(e
z@a_SihB*2zuAG#}3@f(hLY@-_as}zss4^h8bz*5n06^qIX*Dh;yb>vt)Hq<?QWzNc
zOh4-9>+A3D&n7)JRJ1MN@#Eeb#(myk6XRzNf0d%B3|5+{<q)oftvc@N3&U-L-{oDp
zK8GbRY!y7^i%h%=U<A;y_x4sZi{4ET4u1c3kXFKoL32t-|1HcTAQ0f9vTA%SMI{vp
zRH)y*ckc_}stte+@(mX|`w3Km?0eUv8i|A@?VC3@bp-`d*h2~#JIN5;{@<!hY`M@2
zrl9wsE*A|;%aysAN$_H8%TB!Q*k#&p&)om3G9+&bQ|;^aZh%Ju=LB>%;v2>^&;s!D
z{~s&2WamH0E16k4DJniJCSe7oD230RbNqwU4}?8DbAVgg(lQAjyBQa*HKYRk9{f^v
z$YbM8j$)zQoIELpNUH+}U!>AeiVBdmROBsQrrJ?d?=sjnmy29$dfwl7EZ}i(Qon0i
z%}|7p)xfuYZuZ?-&ky10VCR-}XO=Ll`s3zMmIL20-!CW1nOU40<cf;Ye4Co09<1Tu
zxVLpF4K>E9&_*flXmyQ<{C=T3WQj-f$k?p8OSNEUlt(-*71?FE2xrpH(U>Z`Za{8D
zZ!K=+ZjgG95Pe|HDoWCYq|zPd&)>7MpRhBwwx3Yb*et|AOEGOEw`tTUDBsp5Jv_Yp
zVaIOX<Td@)u8Qfq_k5gXw|D=%D!`OCd3Xs`?b>vFY2@36G%w}GG<Jphm;I?dk7|aB
zx%Oth+TVWq_ILH5oMMqA5ryxNne{+N6gifKtYao{@`fn8$#>Nj%54hhH&W?9x&Pd#
z*XsrvDwdbHuh_pPkH(n$FWl!5+$YC9x^<n+A1C9OTfA}t4!OlxVLXM1SN&ceYfRL6
z>E%N!9)>h+_`c~W{%De1O2tk%ET?yq@#tffI=6)4KZL7e*8KZE0M7DDen*e7+fVo^
zoSjS_PI`}bUd>6K%gRKMLtM9bIzN5wrHlTjWtFshgf^;Oz|t^g!5oYYy>%<X_t%0h
zqnf^ys$LZ;{?34ylB^DtI2hP#_6M9@L}~j}UCyADT9^e}l7I3f<XU-9h<;Nm)zOj=
z8~XnEr^ZRumf>+8qvWg;RKeIGm)l$mlSVrf3*G(JDKj(n2x;j$(J;+>fBSc6if+Xo
z9x0*MxLz%#GSn_S9TBKl7gNr3Xkl$=uZStPSiJD7{~cU@B`%L=CtVNm{{My#1_d!t
zhQZ$3(v7Ca6{DYLUBM3)iO&uidKSy*-8l9C#<%>N4E-#39~(-L)fHS*7p5FRF?qKa
znF>4X*ZuH;y}j*&#};9Smls7%8Uu%3Cbh${$m#cz<BzrByq~x}D!<*WD@aA2pPZXs
zB;}A7m2pU+)wSF~V;gzUN2Ld&^Hm}f?Wovf9^HK|O<V~qwkxm<0+x+)%|_E=@=CgN
zcB)QgsH=(elF7;GcWC>W7O&9On*QdGXYbKD)KhLt-jdg7dp3V%v*2erA(I)S!0hbw
zw{JJE?xQ1ztgjQ3mwxAq39?ftljA+_ar}{SeU0S1b14BiulDacW_Hrps%fwDCF#54
z!^_V&III&|Q@dW&olY_U&=gasVm~|Eq5OffzC|aVnRy(IQm@icULU`BVQAxzUfLqG
zG_b~7*n>2kC^FtC$vE;d(XLvPbeOosKM3CIDAIb08&KP>n(NQc)@<6t&Mv)Uhc6V3
z%SuYpBE5`eI~O4Wa#~t0>RRGPN;Cg)i%P-0vfb{tWd;V=BLqVd)Kx}$EqW+7rUK<d
z)S<74R5J?ZPs={;eH6H_MMc+uE8fvI!-kS%Xj{rp-NCfI(Y`QklQhqiCA^EKTL?RD
z$1Xij9q-lqmyZy+{_RH>o_6&2H+A*vR1PRU6#033C5NL|zks^R_urNwY38d{jpMWN
zqh_c*DOW%F=rl!DL?ZGWdjCOMT3YGUDd{<eWgL&}nndk;R%Lsx{1hB9@G@KUc%w;W
z>zvww`E^1<J+S&@tr*1VR<U~4VPg4u^5~XUARzKEeCZsZqa<%ixh?s|#v?DI5Om)|
zRy^#6{)vH?TrYh1Dzj|WP5Y(?*Q;p%Y)W64W`hkP<4KvTDgM1_xq>8gP0J1}4`bnM
z)D6FafJ||5THOhA&;Es}mus%sAFK6mA;dgP{wlD}LaDZ`k`wv_GYB9=(qAjXAoSCN
zhi3{Ot^APez32E%ilN}f{igGm+jvBIlPjk&CE}G;RHD$=qS9BFOq48OK7i1!>dPod
zk_rHE55PxBdTDs#H(Gc>2Po#U5SI;9e-Lv^PPFdzlU;}t8GoKM=jF@enb{%}whPhP
z{-<hsn23G6a*RANdJ=4D)^rGJ$LrELfEh^e&J1cxGjzKiRhHilsPVpy+l&1E*YlE~
z?ce}jf0QlrQwi5`1bgGgKo9SPvnUkbxN(EI3RQX3-FZ^YVJTj<O*|3Wa`y9LI%D^9
z`F}f~x+C4o-8@M%*}a^V-Scluw@RH$>I7yY{vS9O{DmbrIPlt~eWTF#aR#JGKtrb)
zO-g{VQl?*}%-!>&Jr8c-6#U>VXfI^%{@dNZ+OF5%Ka9JxTh?nU;EPg8U=r76f!hRZ
zDnd?!iQs$i=i*`|NlD2#7iFL`Qvb_L0*pZ=P1YA`KI7J{XQNYVrEs;MF;>-qE*BBQ
z-acPs5dKL8+QK?|HsgyE4<*EEM}}ioIbI!UDp_)4;q<3Zd!kt(3IK--jX+=oD76o%
z&aPrU6iMzLWl+l_0d9RCY<-lqCzNQ1Qd#??-I69iA}OZr5UI-^9h5>Bh#Hh*LMu)j
zKW>P=2{cCqJktZ527}m>&egha8Nba&*7|!>cd)`MBpJ3GI1UhOX?%6EGbrpi?AB^G
z8EI@!Sy+5()R4U*m0JXWp<4L&9MzcGB481h5)#tUqdBZOaFkM%)o{4Dr*|94NbA&F
z6#NK039vf+Eg;O}KIh9dRQ1lF!q{84d`kpx+?0KLP+$FK0XgM&t%)j?SnJFcG?Luw
zH|>dzkwqMjxYfOcOcEK5tcd41Ato6B>53i%INW<Z5W<m~`}fak8Yy^Jj3LYiwXT&6
zcxfXcXB1YVv((^x4xQU$Rdc#gsBNUgEr!k<;v7#Xq3O-d$nf1{;eN|sRB;zu2NQ2c
zR!$0=j?;-zflWs>HxE*gm&j=izw9KG46cIeC}}_9zh-A=?}4tI#KRDqzOPxqIxdAz
z>HPUmdxY9lmbbG|)rn49BiZ9n8X-a^VQMI;7)=g$5;?oYv71FbXbGiye`pnDET79z
z0Y#}2HTn%(w<f@R(chG*jHOwpuICah)aqPS(@7&4@Yh4=MUfi67&?_mAw8@C#uNej
z0qsqyiz!tfaym7f4uXX;4E!urn|QP<fM*%e-$g^WB;(})Dpt$Wo>ENd?Pq63ojWjl
zT0yV;-MeG>-R}GeN>S`@kF0~r5pG-LIeZC;Ft}|}TSV2|{M!YHbQ!Ojz+A<po}>cb
zo{q*^N@aDSNTeC?7Ke>pY?Gdq`j4NA_2UiNI_o&&T6F1Vp|Em5f+RYuCm*p!+*rIq
z8~9n}Fpe8f!EsPdhnEktNpHD)U_-tz3ea_7-K}*3V}g=sxI2R?B(YMnOQY(;leCnI
zr*CIboMtzXvAESOi{t!Xa$Ki$=<&L^i5H9bX`tA!m?9L}{`F{X4csHApGVtOn~jY#
z{Qwoc0_%oTcbNS}Pmpn!DAb(PgD{ZZL#twEZ_M7{PJL#VkWMq@!JAisyRCkDV#O@{
zD%GGsV`W0E8H5vBdD=_%2Q6g(oUjIHZ(NYzJUHKu#ir9_!H()H)AHUHLM~}1AhNZk
zDy2BKdC;waH|c7q?`?LH_|2@8@9N#fAC|M$EUM<o*I6N!$w!F4xE$_y5~3EuQwV8W
z`TNan{hn2{Hn{Je)0RjN4-ZmVot&QT+m!#?9f}ucb~6u*YM5cwjz};VA;qporx#Bz
z#+~iVU!#4OyhYvIoLleE3<1YXtbRjDNl7KI9~jwQuyNRhDj^>yKb}WB94R$J?j;5I
zj=@1ac3ZmC=IHL=OC-lwiWd0?(PVrc^R}m`zFq^jz$nZzL&%v23N^Hn&<jD%PWm@c
z+k?FVjJ7PU9OWfaAnQlHQ~CMnFKD1hcx@9|n^kLlx=lEMWsp)?{|}j+`C_bjuc_NL
zIP3-BNrAFW))egeswgd$=$Jw3LqPLGYg~1Pr9LU@U`FmsafM|C>|QS*nl5!?Ip2#g
zPLURx*d>A?8@BNpTs&3i?_5~4{%PMMEK<}7V&I4ZmFz<a7Ny2ec*o8TSfRTFlho=o
z{2H7hBH7h>TefTwcK*H4?#P+2HMY#b+o)fqerT1Xz^PLYhC|D%ISAPrJDHnvPtg}m
zEh9>egE=FgIU7$R(xR5mzIwHfQa2D5!nm~WC`!Sc0s^VUX*%1W^S<^eZtHa!Rr`Im
z_>kX|3(ARsj!FfG<18Pl{2Yj@HYkY-<W6u7iIJDCuUER3SDo={!x}~Ikhm)23psR(
zM;;l3J|qxT32UDewc0O(Q<F4|7LO@Z2^`BhA=cV-d|X>(qox*!-(~tke0`si%6#cR
z`#qAre$`mf8rdc-CuguN;F;jhU5&aScBQ{x?4B2sDqGGI(gs?{E^Vngaok(>@=;!i
zU}c7oxRF0c_m4l=gG~Kvy`7%5EcpRVT|{mh8j47ZUuMaN_wQi~J4OZ&SNF%Yp<(OQ
z45HXabb}hy3^)_E#IC@+<}kZ*L*+rcXnQ3`G3V*DI^p#)ZgbKGMIuzqcD`4m^CJo!
zUN^_2b90wW$p8;bP4y+0^mqTSTqIVbveMFDGA!F+QB(^vpi5O|2z%Ob+OSi_2BXLD
z{kx33-(RdIR4~E5w}0&qz@I~JO)V{75iEpamzrVyzbF7zqtHv@qvRvy<DV$|hc0^&
z+_wcY_vU<gOLNf))$ZACt~VX-JXrczD02J5y`0||j33(@X+HPA>e$%Pp!z9I^Qvj1
zSK3UiEw_>oZ=H4{cj=oKnzm=lLK^m^?6u`CDa$=w_kP_Pe%awwYmXetm)N~qDpTZ;
z#qCFRBSXLVN*}#@w7O`fG@G~8I!|!x_4*P=Q?~)sj7Rz)y2(8{Z~_9~>z6wVJgTdb
z_fE<gmQ^5=UBj1H;ekK}peyItNh6~;Ks1=23@r`ZHtpS}DJ^Haa;=bYhf~OIA+MaT
zCY9bS-mTXH1BF{IHSjO7<B`gz+jk*d7ByL=xz%;IZn;aBi>aUYd)>&wePB<~&{-_<
zy5{(+Ea4xnbW8fcDb`&cVJ|KLF$o5Dqw^=+f?m}#vxx7ncxc9+&b#?sx@>aJaC@7q
zEK9kjPXXiXs+VrfJTS`ov}sy7&$iCHobZb`Gi9vRHFC%il-$I~h+fH$lcMPSa}Uz{
zy!u;j_0SEPL@C9Cqap>gadF;TtM!tiI_~dQ)!}_sCm^c3HS#QqX0Ee+dh4A_P%iy*
zUx3fua9OA!dT=K${dLsVIb`L)+cOKNP!)JuqMMKu^rer9H93rqI8o(b%!6b1!THTo
z&<)>QDq~70q0NzHY2AG@+oF0UZduZRA$}&6>8DoS*4(7Fp$00<-#-dIfFEcK_3zMF
z_7Yyfjp~aX9tr2kEwOKl-7>{;?)n@4BIX|@F_!#h&UnVZUGeP@9NDB;O=Ntc4gw0y
z!YKGUYhPsTPdlSQ-A}E0_wNrs@Q^L#!SIRPPi|ZNhtnc~V|X#=QlxqFe`buF%zU{Y
zMMAE}Dz>?vx^=%COx<*0RXlq1XlVIaG{vLfrm3-3@oJ~K;igSjzcDrlhWH1t2Lv2u
zO;}!@no*kb{LX{{26U%jh|Tv~&tkURZts)%>tru+G9GU3fb*Mq%00NUv^E-2Ip3F?
zCjZ_cx_)*Q)f-gz`Fb8Ha6&+y{drl!k1hB2caJQX74{`9k7Y^DINHc{ZE5-Od^)$O
z&eUBMD>m+{kP~W-+NA4ngg7x_Pw9CL1o}Y5j@|BN*5ALq@4|4_Du>m{3>By&krq?p
zox4)UlRG~*-CO%2q{e_IRVnXVYcx|6V}tI{F^An4WSYR$qei+y?(JwFAb=O9%V=0f
zuE5;n8-@1Ag73v{4P5!b9n*s>>W22c9UsFOdUyH_Ur<(NsS;%BSly=i!`{t3J}c>F
zUe{O-k0j|R1(vQQ4o=twr7O41{i63spiK%d+ih_1P*R22OA|qHji$TT=r!*42|GWo
zlWWsRHw$Y>Hz&@TE08)HV1q8H>7l-HlT&9Ty^C?wamxdaO9gHvZWjRkI&@cz!M50o
zxSiSJGNF@DLPw=PX{FitJ!@4&mn}_}HhT1IP>h2sN3OW6Of^|I_szLKXfZRX^k?|q
zxk<^0u2h;SB?#V1$UW#VG`8iS6k~qaONw2<KX$HP*p9+t)1Tcyv<rXjzfN=lD)`UM
zDNg^nImK?MA^31TWEwF4Wke>R3KsqSFvIQtn3h(e^b|GbLox7EcG>>EI~h00-riww
zStNY@Q`9eh+7+F9U$au?>qG-%>LHoA{1Nr)?|bVfD*as+)4;)?q#XVBXq~CMyLUfD
zSr^l@zfQR#vG;uI74_oh8w?^o2ktmlACnUBn!|lf|E{GKl5`LJ^vy)5NUMR({->HB
z$G=@ycXLDF22NY9U0d_8%?3XeCj6*z6%`vQvV`4AUOmU!ps`J7;lRM%i<e}XtypPe
zV!pqLuWb$0RKRj`-4hzd!KOI&za5QL5zZrEv~zM&`r$+GZo!b6fSLy)yP0X-LbUrM
zZ#DCTIqs6<v=ckEK3_80%kb2$t+Cv+{1uWAI?BV9G#~i~)VRG(d9V6_5|x>ZJbv+F
z{}OgQqi?gIk#d#|r_zC$DCLLC=eFylg)_t*ewId?$V0iU{>FKVVlBLsFe`a0#kN27
z@Z8L|tN97ZD;VfN%z9bXkI??w`Gz0wq*^RFWBEFY+0Y2@$s)gGFk^>K^Q~jlukGq0
zcbFKqSx3@js86y874tIf%WpsFl-1rVMg5{xKS=%&=W6m0F{7v)W*R29tCmF!lv@xO
zy#+6i0{yui>D*4my>YRa6~Ac4rVbZ$_E7u2r=3vH6l1Kx)pV0uB<M3lS=>%E*)3aA
zXBEi=D`Sb~@ZobQhX*pgY_DUXhgY(H-O8j+Nr9ymZb9wVlv}bzBl~9UlIzB}ciy6I
ziqdOKlii;YNVjlC?{XR{@2zRjISX$3_&;QBxdm|zP9UHgZ4DDX{5B<9f*xXEUthAa
zZQD`r$AbM)KYsA1nwH#QKp075q}4QDqlz07k4*aWizU3uDr#LBT0?K$O<X1o?z63Z
zN<-=1;b5kc8tU|7gx^_bzJ7hwWmX{FnU-?c1}QGVuI=cD^wC%o5z%*W`+}|~V1<`O
zNaVkDnHx{fC%mOO+nyA_Rm&M^2R?kiVX^7C1(jhe7h+fGzM`Jd?9OFQZtfZs?%!bl
z3h?Z`0C^BRuq%c2fFD|n{<7u~I}EWqui#peBtQ97sdM53ubzs`h4_P|`iqzfR6V@q
z{qpnsUWm`k(lN4>y$}~AmIG1<`|-n&bUhHuJL+lL<C=)-mr8@5Ag)RKCgqN6amg%h
zy76Ign><#j`U}&uAs6(kYTg({u}KJ+GR1xUYQwXA^c(<m@7Gt)MjVeHzXeyA_V{6k
zUy;NLN4zC4Q#ZgR>JqdpgTX`QkFemkGzHuZX{l?2b4!Wk0FMEN1hi_59+`%0`($Jl
z?QvNJ1$7u`*of;3UcQ%+4Y#8sBi%++80;gSb%*=i_egGPiMsb*Z*l2PNjkUH@Nm~X
z+$9xSjsgoKE9<~(1>6g|pOQg06HqQBT{GyH2=MpscuhnKrE9^bneax?zEtBl%b;`e
z<dW_<rU7v&N5>@VNcN(qN<BU6Tp9O!;01rX3EP-I?h}DD0WIRh*A6OBA`UZJjP;^l
zFlz|ofuWmm$8OAU+Q+lh{=y<A%9-dmfjm0#XP%ve4^f4<8?vzN4T`G7ri#X$T6Dw^
zT?hN1$dd*l^gJiP`kf$4Z|sFJ=fRi{7^K(DX0=&(z|L|7tq(knsB-nqG?bSsAYfOb
zKfgDdpZaA;zWUBfjfN6;JB*chXX^`E%9+ASoVIwTiXcE%bcB4szGj_zRA`vCBic?#
z%UU;Sj455ePQs%D(FiBnP9{*)wgR0)RK_MDt;LkX;gkRKSpBX2ryU2U%<zQ0XhXl6
z`Xb8ha{JDs1$SNz3sO`UA1%Z0{c0^sr~*-1{%`Lz9Jf-`2$t<?{Z3PgDn(cj_Y6p<
zq4p!P|LuzdZ)tuXS~!OJpRQT(>}!8zOKx~(!FmscBY4#mwOT2!K-u!u#my2tzdDv3
zNzF<7X;PlRFa}9FQl3wXu3)8t?zc~VSB}ko-+7d>y&e6QCb;v{3N-VW#_D-OJv!)F
zOYJZ5+vG6Z(V&=*EY!R5nE+Kf+_5-3KHjUxIL{p~z4vsCJYof-DZSf$e}%!RxvNFA
z3WyxE@q12Bb0+T)sNq|qs2lR$MnDhl3&qQomD&<2Y~@>|X}@Knp0ev0Cs#_AFPG|@
z-zq99omDkUD_ZW6)2@_tr~9?|wEBn`xfSJojH?P7`E<%A_LCy8=>0v($2Hd4>-^Se
z-L%uHO}`(K6y;X*t0Q13SG#wQ@mOy0js#9!7LqJ~ZDtso8zM-nGvgN@SIJCu>eO83
z)?UZ)4J4(0<8vOnk%LcM#zeDGX22_6si)JQoabt$Cssai)D2}K)^~pDI}>HcX04}t
z9qkq}_pi;Jn>qxHzMq}&NgkZ;vW}Qbc;`ZHLTU&%(Wm6nDikq2TE~ib|If+WW-%HR
z6G8OA$qZ5GWE>blr@ow}r6qCfpvyJMDQOLkP&>bB)Bx3I(p+j<|N6w*wQI4XhZ2?(
ztYe|mXrx6ei6dz<A)771E?4{ZKEQ%v#OCXHrT>MXMM?@wD}p{@s|n~MVddy%LDyIJ
z>vB}^ZGU|`!ob9Tif+lu3Iyi9?9;e0jawYdsTSD825xfBUAvCKn9p?J;@$Cq<^*i9
z5j2Zl6?hwYq>1SY8!7CIO>5h`ckef5)tlCs{A0M*vj1ubB34{tN)$BF2&-#-fHq(Y
z@Axx~))&)d<KmR$z4af}zTqF3T}DZo^0~JkfV_;|v8sXkRN?9(_Oio=>kT5{xTre3
z+~NECWh|l&icnQHs;p2cHw#iyiu!5tX+N_M2M1Z-Sc#(sV98`(6~r1D;y^&)B7W8~
zND~m`{MpFouz?`GRKXoF21_^`$fe(luiuY;F5~W=)VO>FOCzR&<LS280<;wo@g&F;
z(be`^Kdjg24>J1l^u(AB(>X19?@U}Y(ioFmH%WPqH4Ymv0Pq4no1FiwcgEw{#nuw$
ztlQocgG1&UevbVWwr9^p>H~b4)lW14lAQoDGx(@D-eVGNBD3STkOct*s7*U%ZP>yA
z11@V-RpttMs`^g#i#X1Gmbq)9Uq&g$oy}Ymy+eN$|2H}lLE3|q#*S{A51T?&(pVP2
z4n^5UDVN;mJrrjm!nA~}AxReLtq-3U6eewYx=3)9O2bl~<3=ABaH;Kk^aPH+>bY;S
zg+pSJ<F!FlWzqAJT0ciyhAjX>sLDQg5#v?x(_r&Lz4cFMtNqq2=;>27SJ$FvVP^%~
z51ekp%=&%3nI}5xl)FSz)?tyxA?@#Wk-yh2EFXzea<hD*L0tQ6Oz4ZrN~f+pQ2sQv
zJHYgg`9i3hiBx9q3&HK@-6J^q*G-I6g>Cct#Y=?K=sFkfQ{rWP)Tj-|0j1#)YtiR1
z)-*e`EjO*g^q2i=vgneE7bNh1LDzUGocrYNmh7iTM2I|qUzab09>pzzyN_eZ;M)P`
zjgG1;>?-$<4+{>ky1KB+u766aeS9g(%V%oF9NXGzYdW9K9Q~tOrRa^9d>C`vqZLbC
zn8>3QtYhe!5)F|<YfN!5v(K%|5vXKQRN%KMQGOK?oz5*wDnE7%mHB`DFhmZeuARHs
zB_xXUorIM=+a@#clBt)TWhc*_eD@AjJsi3~ExuW?=D%H$_<qB}v#O@PT`?lJeIapZ
z3j{Ojo?|4hw;)=hZ81TEFsR&T)Kz5omk-^108CO~2w?;gf@Mc<XbRm{HRuFo-@Tg?
z-c8b!h7#DubMBi|K+Tax|3f%dZw?Bmm1Prm0^h%Y?_bBj|2je(#wfP{8q`EduO}ND
zq3ZS$3a{ZB4E490KuguwJga*I-AUNQMdwv3Y`LV*g=~QW5pzs)C*`mYppfF^M~glY
zVMv1>q~ll+wI1y8k$)i{{fE&pF|xk$;i7<2o=!<`+8}Ct6D6TZ;UtE58=9^L4m=G-
zKNh-Hf?ObMhU&G#P3iRg-X2a-wFNw+5zP%RE&OKIu7Bw6-nHu{u-50bqFYdDEucqX
zf&&gC?3$(q`HFOI!dw>QLb`8foKTMgVLl6lN3q^pcx>2l$XGz1jIOLyB5lef3t;oE
z7o_n8J*sH%`vPQ!_SHx<L7|dY+54KNU>AUVfs_}km=>uV)C-~n(Om~or3%{s^=9v*
zemN{GDC$4QBQaI}ztGQLigwu>7}+PMMOgdo61ZroZ6}Tu8UVb|ziy27-@qYq_Q1*_
zeVO~CvlD9SODOT@uWz5eI}WpO0o~uDWOR<`m0lVuLRIsayj<#YRtK=#Z`ZA?I9qhJ
zwexK$xAKI<9qMT;;-JS4k@8`jEiF;fv$a5P$**r~*W3MG_<m(&RDooj>GkVM4jBSt
zI4C}GM@+P)@@i2+qa+vD<cX6q*RFXNAp-DM_ZCwHMJlxqOa&9TZn_ilC;7o7jceNa
z<Zbq7%FR3>cE`IIKr;-?f0mH{V)~cCxFPgioUCrHul-6g_f<50IE07wJMdU;y_~UW
zVcBy<J)hegUc$H$@o4~E92BLyV^2|&U1S=GECpOPy%)_y6wfZ;6UM(kpLf5T$O6z&
z|98Qx$p`C>qLR<2tPMUh@z1fA`rpV1h;G9Fdh_lXt4YXL1#)FS!MV>ywEVN3Xi!Ga
z8y^DK5f|g)ByM+G5xL-(G`U5$_>PX?##PA+u`5<4M!qZUUEe~Sj`#fs;lHD>5UW)_
zT=J`|ivD;CTL}yWhksstOWbqe-c^b&?b`p2EDVgmKV**o+4NLDhuMw|6jc#3(3s--
zp^ucpU`zB54vu~Q{&+6-e%q+B42N=AY|Sq54-7O${}yR2fEy8hxhSkknA9_W3PCL6
z*2$`V%88_Mj`3t9ZwjO>V6(lbyvRM~{{mW-iZ(YQ?@(@cwoB?^iv|T$B5lNd?%q|@
z)oo#%exZ}ChPDyX;D&0nQUp7`T5uGp-yK_t7^4Y|>~BD}4?{$f8_b`dW1L19;)1Q)
z*FP+btn^y@IEG@B17fi4f`-m*sdi>fUhG%VyG~R~SC=(6Y02+&ZL!Y}Q-9d1jvQS1
z?q$|eh1YN$lKL`=9NuiiI_?{VP4!;Cot2f9B?U(~#?>90aPD4^C2R9ny0@)W#1N)f
z=DliHcE>q|?s7Qr+(LUdh%)veB$WZ686t$CQk4DE`sY`^?mA!{m2|V_-oN9%pk+&!
zR#ZmIl4dG+7J5m`7rHD+uOD%egK=xUzIcyx%s~iuN6%Deaex~9MDV~JbxZm;NJz}Q
zMYOxx`>~9)w56<Ri|4<;zb=xGsND-0n1m=Dsl55Qs|y{5b5nf_%_ev~0Rvaz)(7dw
z?wzdhMay3(H!Jm!(kp7kgu0^=Io<KXCSF9Uvo=YG3$?i<l{*GMYy+Ovsru;NxqTn!
zCCyE1cbz1C-MdbD#=hp8=@QB(KGfb<!hQa{0dwfCyMhiFB4v&bwhq27N8fM__TwZQ
zO8AHl-$iGm(V5iR*iVOjqtJ-dizXOle4jgZoke7ws;92Hm=g=r0a_cDdhD9=TrOm{
zH0drS?HO&?porHZjgkE>sxLbR#(h%i2@209Mo83lhoj@B1a#!*gz_i7aFIB)B5EqK
zf0Im3*qoc>G%C!0anTQ9D>Ub#_fOZi8XH}*NL-t_?}BFAQEc5|j7=EFbixgcU#m<-
zPh-nVvgIJ`yvcRKFI4=jN0VJ=WwEv47t*QS+DXIn!<O-{VexGIs^bo&pqKQ9T(}CO
zN{Kbb4n18+z7cbydr0~LE3sAEm7}1k`H1xQS<YKlZ*^ixluW+4XP2a&J7Mybbq#5z
z38=rv!O4tWJc6N?G;odAqk|j9-scj~*s<||BC;hsCh?)izE-%SuLCLlb2q33<cegA
z2qb;7B^}CCY8o0EE<whRdOwq$CdRN&gW=bg=Y6Bh^CnMX`ZZv4$EorDv#p(P=_~q5
z|9q4sWs+eIKVKDjLIv7#`}}5vj)Sb&b(C|0bm(H28PBP`WlG50CsEJz!rr3|8Q4%~
z?Y3>^&e(&8XL=vv1K9xM(&pRe6SY!{gZH1m!5FRee0TQ`B^aay#AipVoD$dkUYnbA
zVKl4vOJBtUKKd;0QuH%n@5C?YbWX-VD5P8}gJ~M=C|MBBxsyH%2k`{7V4qpEYJSO7
znL?Ilmqyo31oprqI141facr;Z31cg7O5w;qUUwpal~i!SKS8|zFnwQVGOmt&xNiOW
zFjq86`#*g61nn3Eun|%-cx^&N+N!<<4RWM&d87JjO5j*BU3`(uElCCe^A_!^<3BCy
z17Rfx73@R79p;U05?>M3WPGis0R%}y7^(V_de>7MDwfZQMNtgcQV8_M9f~{gk7t^g
zW6uNJLqJf_<ueA<H}O$nVCL$tI6f@Dl3B>Ci7^?SD}5-*k{&Xhv<PgsiI@addCDbV
zIhA^n+|-6uk+(UgQlGP-)l?2f6JNAHk<DAs6~zS>3OBGa*}3oFLDJ){gkHOpNMmaM
z9bI}ANGICS!;`K7rV@y9OIB`<#LgxdgfBOi(|p6*a@)q|P8X|>f-MYKDGOS@ERCD+
zvE-Y)@coaWx0ZEb8)9r=oAB2^B>(y|3gbmYJ~>Bn;(z!GkXHUf0sp@b`*$^s{~nV6
z9umxf|1D(wKfc~{c~A-0bDd4CF`#1g)0^!=NK<G+No)W*e$kHzA-lR)$&G2~Z;|3$
zF80|VE)24PZ||rJs;B^|yhgBj+D?f}>YQK1B;HA?kHg}gPbrJ^sa`!QUdrucuE+iR
zm2cfw5VA|Qy8(SB8v(JA{a*e@dv6}k^}4nV|D<R@sgyB=N-B|*nO250AsH%DNr{kI
zW~)`nSQe2?86rZ;kXdCYWQqt8l6juz=eVr3_g;J7d%y4fdGF`*{PBLC{#h2k@%z25
z^E%JtJdWcO6cmIkQVpOWJc2I_jLgr^r@{sZg5v}Tp6p>4rlzjGOObeX!W2512yp>h
z1q)sCf!EivX)D-9XNh5Y0rmQBm;<ZjDS+Ea+}6s4Dv6UPWRt|RBhMLrTiv853b*A%
zcy)C{#8c(6IK~Oq(kWs+jItGZUw88oUS(Ou2+<WjZf`N0nbntJkW7pr>R!7hl}X|j
zh4Kd%V2#QjUaKLk8L-qPDrrcI5Qc5O-0ZH}L2Nbf9SOgUbhHD<Eb+^0p+Z;7tB|ZD
z!ToJws8v_jX(GU~6{)iZ074^J%xrTUK1!YnNaAdo3oWHr9#W$%`!pj!2u(Ar{-p{1
z;8bZkw!`I}t$kn#ynvR3*n<$FAjDds$wXNL+TUvkDv8sb<XE{7K|jK7{bV8j0g&yS
zIDS0tRXQqfQlP^=)dP~4R|aJv$s`}|DDkZ)i6`_h@b&F$U0TfGxek(4>Bo=N4JrB%
zgGJ+aQ^X4;JUo2NWn;YfB;dt~&*6*b-oAZHuA>wPk_lp!uGeXlE1{bAj*jMmb9Cj0
z%E}{IHUmcUaMnFcy2`)Zxx<3o@i6X@7fk;?FiY6oOe^{Q3i)#6!T37f@Mmh#M6c=+
z^W7APSEJBkDFzSI%g=9P)XSGu)13!gu6uEae<6>Dg_Ay7XvIOo#_9n>BX4OrnE`Yj
zB-FR4xVR7aCjbqsH|P~egqYF(7D~$e$*K>kBVn_cBMI*Sd@Tga6E8?inPDy=jFc{9
zP8vSg#aQ^>&?R+Xg~RGs+zT>QZdj15rl%Jswjoj2E_TR%yn%9bKp!p)_Lmp#ahD)g
zwh_t0@Rr5M_Pt}nof~=g?-zzAgxQUz^iQ4hz{17)=<q^PGe0&(MMoFDfrq3Q$bTf+
zYjv6mWn^c+L45zv6%>Ri^bMrJB9%K`XScGjT<788VTPlh+9d6J0WS!f#FlcCbn{!@
z0Ezd4g%KOl!eSV>88qK~PftDq9F@E<e1HYt@DTY3#4>CehAQ-UdMmn>cRyWCP%!~h
z;jr`KSU4L<I7f)ZauoJbF|eP|5|mH^IxY*=QE%Qf*dZJAhNhSNJYxx%zv$Kr^v>A)
zaDV(*IkEv=6<;*VHPFH!1KgEI)%h+<36gcLdsZ6OW<$N!0iy_ODWtL`-r=Bp6vK<o
zSYmOyqhbK2;=%y2z3^%pw<HvcaY+Xp7m4Q}QRQD^_Mv<uaGEi!$mI3(BFwS3y3(<q
z!?9h?1q`Wdi&(dIZAcS5d-cjg#p?y=6a)||;>Xg!Dg0pAl%A+^;X-fo;aj6uI*8FQ
z{?CYb*f*f|VHy&G8BPb<bv9pZP}4A@G(SXcC&B?MaR~B$_wJWKr)OtEjZTG2eG{?x
zar{nKR}IApw@Or(%z5xad*r<UT}Ob1jv(^2?~aKtUX;7Ly91xx0&kZ?5Hg!G%hxTu
z70{v{ho6K|4kBsA0=gyMXj8C3yju*zZ4GWg!IXDX3D;dKK7JI0HryMd97Hl;Gf)<U
z+G|WC@NOeOw<tKw6@z~IV?L8_U6rutB$&V<Bnng-A|Ff>rXB=zsu_t#RU}hWQxhLU
zvh@!htjUzb5rBAFT2_`YX-6;tXATyYo`Vw|HTBTYi$h!<4_`N!yQ{#(-86SXov7;T
z@)W=~2AS29HMbJY$~xQ!&_}cT!(p}ui#=juO<LVdLk|HCU_ivT^B3`S__`;NA_`>z
z@vYXeYg<JsJ{U)5gO9#laTY4>1MdZDlXa!7hg!6ule37CR*>sjtO+hd%?8W5$i}f1
z5vBo4*)QUywfLbx1eI*5qxk@AQ}15l1*(BM@cGN7Aal@VQM>brh%{s-=9FmbWbokg
z?Zpeoe;zfm=r5boR8e8HEL2HeT{X;%0{G_RD3~IO5;>rRL{8(7B*cbeeK-cKyQ|*L
zG*?zu&S3Q?lCd5nxDuaLcqCcDW%HW*i&7#<g;fh{t`fAfge2Le)jzs`2@Z8x*2b`z
zDK{82&qTy9LJ+NJJQFGwgFec2+#0`V1heb#WFkc+0%JV%ohL0EgZANIBsQc3`-MYh
zmFQ6BGr|Hw^y#vEFQ64h{Zjsx+4Cf6t$18edxRG4NlDP;Xc*Yz@ILDH#$_kxG}VZ0
z1i%^{>?CW`y+h`m`~zj!Euj2ZPnO5G#O^Agc_Ht5w<PGGfd{585L3Z!9MiM#T1X@=
z1AKh(jYFi|vxYM$@pi`v8d4@_6-bOBK=7UT!0k)?*#mfhXZOId;kk-;{=$G6=~)pw
z9a3<?sGhGm8OGmhAs;3NvS`cmqS^t_RfAYa%npbPT>flbvK8(cAQ>Z+u*43QxKe;7
zu1Me^><cwWPW~Vd^iGd-5a(7{;=}@nK<WKK^xXnjoZ|0Iu`b~Z)OB!HdlOk#q@@QX
zY#&0lJuD*#=Q>D?#x+Y-&E-uqaeXJOiq<?Qzi5C>eJFDSdB}JY;#PyFSYw(<D5#s^
zj-&T^p5q(g$g`LUU-Ga5o19qn{H`SR9H#)Nh6~}?(MeE~G-Xzzo`WT$7%BH*@~Vzg
zhVZMq=FV+tH@3J@V6@WE&WoTDL<{S(Q13!eg<0rxQ|%{uS)GPBY@46L#)cO)F8K-w
zLU|{qv|sAE!C3bYUP#;WlQ1zY%9|f8GioJ!pKwBOueU$>^o%g!@SDV&t?;@q#9Q0L
zu!bx{GRcAXinlkr08b$z0>tOdK-q8_#J_lPiYOC63t3K@0P+!zN0`&(`GA0c=lCo5
z2WPRi$>WLB2#K;g6~moppA*^ztPp{?5z`b^D=RDS3hhjs669jmZ&H$Y3X%hH-o<xR
z!^b(zK`Zku4!hSBPJ+Rrmd`M{<Arfjro5#6comk?enrgq`GPPDcI+<V!H4LPD>eO{
zTG4`GEwY8@@QP6T^4h45?@tFx)+44N&QsZ~na%YxKH`H_;HK6B)XK#Lg$ZmZ{=84o
z9>!l2@D19FsTbtXbHYv77yT;@wzU$fbR<8|Ur!-7sDf$NfT}7hZu4eFM&faT2JA&`
z23I3GC|Zb*r85kEFKD00-Tz^Q<{hBCXUO6t8yYX${rFng!%NH+x|u76weH-xLwLn?
z5R0JVc3eS~wE}ZJ)NkB)LGGf<;{0Bn%8H6FM6!XT7|fKJ+PG*Yf+oR>cJ0!e&q9?F
zs%;L<Hccj`PMk{VfJzZit&P5Ar2$}##+oLsS?Ymz#C}O0V`z1rOC-L@IQ*Bh)xMOy
zeYwbqE&Jvu^z^DC{Hhl&#Pe%s86woc#6%f7e8OggHBOflytb;~yG#s~2_2HSJ?S`2
zsI5z&p;&fXQ<NRNWOT1W{)4Ne0>u~xctjH21DZ$tbC~<0OAce6>(j$+8aWJ<FdrRS
zN}7Ue-YHDIDErikR(z7?2ksp^jv{)3%{fxcCgw2t0{Bp0+8N3l)^RaPTrMSHouNB%
zctQqv94BopKudJ7lF(ApPSFK32>tk-ipbF*#0c%puBC~^SvtPp>BQ7k%8G-5*j&^Z
z)ULDF{?^`l^V7|+7i}kVlMesUrnfyWx#!4{DUZ5kqeAuobjzwp-)-YYi=q^$y1*po
zQUna2Se<*<3Dlsw*aTOtS?E{bwXtT?vQ}n9)Svftc`y9_)~mZ2<5^knmTI*B6XuN0
ze3y4$7yKR()apIXQbl}OzJ@Ei%aE}VVZ*X55oUP+z=?-nOt7RwI*Jb7k0UPck^U3v
zk`JeJ5<HMR|9%6Si<82JjXwpqX;=%FWWJI|yqfrWE&ROp-IB=qI`*Zz#&^fA(A#U*
z1yR!n;oe5#n1DURd2J7Pj6Nrmt<(eC(y%d}0fdB_t^q}zy+ZOB$9PsF;Gtb0wd173
ziKv46@;dbq5+HJN*0AP)s7rhkOilSua=+hQ-RDT1n>4xnE(pfh*k#hn${DEGFoo$S
zW0aXIQ@3@Vn~sJ98e}|D%8sxr8MdBh8>XL^tY-j8Qx6X%h^|=p0ZeVcI2}Q!5ZK}o
zG!HYPNEwLG9{_j`(J?EfkrONlLwFYWMZ=_`7sz;4uRf7hV1$-VOmZ^pM`2%?JhDiU
zpv@gr7#FL2&u$puM&>sm{>0vcjc5GNRiNLi0tI}?^UbnSl6W~l)iIQGu&)R-=>$h7
z*szmdf{<dADov?8nmostTQP<%dF_sDZ4{343mm>Sb9J0Tf2ilkA(CvjA1(xB*|&Fe
z5F<I5WgfS1G@NhWnqLJu0?cm`^WjJnu+0UyZpFo|p(5tL;NtqRbAw&*QOSxpjv+RX
z=-4(-CIe|4X})=z*i2!<MO;LLj9==lv8%+q92dtX$xtHz8Pjv*1BBL%<iEdp5guv8
z4T*T$Jv;R{66T^}E=i-|2FNMU15t%7#x{q3!3Fr%*8-HCKypv~iAbecwn)M@Nfdtj
zZ6(5p5|V?&bSVljbKGBgc=K_jklyF>msKMYgu>v5i5lXQD5{8w)$(u&t#p{HzbD{7
zPTyz5+Btu^=s>FTg1ur5fYnr|IU7Pn29}-1>IR70msAhLix3Oke!A@*$)Y3);EWqd
z6j3k?lOY5*a#|n(>Lb%E&=kV~cCIr8@jelThut`85nKs(8v><xoU29vNN3^HpNKke
zV({BVB4wOufzWI@$UvNj#YT*S@BH=~oRTx&RX>AB2kkIL*wv{Z01~%fq}orUI4RxM
z<6^Cas}CtFoK<z1m)WDC+1=orMk+(n#Yg|z{A(TpFfnBHizeQfh(#6kgE?boNLC3#
zUkQ9b>gIf)SRmy-P;%mn54Tn)lr_ZFi69rGr8io-7wXe5@Vg=^LT&h&V#j^PsiWPQ
zl7Z|PRDY5kqtd8Ravt48aAI+r{wk6Y04UyrO^S3cx80%4I18p_7>gygiiSpgMu#Zd
zNPdViThV8XWkUh12-(KP>_ZQrU1|N}UMuexyX^C1WDm(wh@*<(67HQ?ev%PY4f<R-
z<rQm~G7e{rlX983Lm4pNB_6*-V};#ACW)ju&R7UZm#VEp{|w};Lzs(1OfQHZS>MD0
zm9Bhpf;9FZj178^lkVvg(!xWO%Q`fS?`s6lCqEvXr~u-hNSOlAfKB^;MYRgfeMgYf
z5PN0h7k)gzPR?MI!Z{o`_)sx$uRP%g=i_)jUm?3;Jz_bGla*+!F#Ld?7`YK68hjO$
ze9=%lkWZxHXa=WBB7P&bVC!9`Z!e;-IO+H6?b~}phPHwTii_CuSxGwJhh|(G(!9FG
z7nCH~<f!@?E;uPFpjBPwlD&nwyJ3)@$j@N6-R8JhBE6`xSN(-DwiY=aiL2T(-<bw|
z-=0dB#n;H(IxYH3i5Vd|!2I{9l4G6N48p2i84H5=Dk8nnCN7|9^*Vi{PNdrl-s)H-
z{x`v4FjE1ci1gd6p;QCwROfG0xsHzI(OlT31RE+#K?F=d!-p3JyJY(V$l_6id-*Gp
zt>Ge<&9v=OBwb&``*FA=@5R{*EJ$eG8cN#ocjTd-h&&m{uO3*n-4V{B>Ve8J012Kn
zE~VAeaUZM^n4f3&P0sz9cL5uuBk%UDH#iPh0UIDR3!xvP-w$|4rzImYe+WrxZJEQ7
zbK_aTV?Ue~l3&w6z@oQHRQVo4M*dn3A{{`dF&P;$X&d9KrxPQ>i26)gJP?)eupQCX
zB=S>HdfCZS8U8`}`8VJw{c(8slO_xf2#~AMViOPFx^-*zL@pXbwNVP|70=@MMvvUG
zuEY_hwBuw(MoqoY!Gq)E78TD?51$zqvl9dP^w4(HyvtU#JwynP!F?oLdM3nY(r&Yh
zC?(Iq7=mc5+b@PLl$QHCv`Qri4}pK-!J4{w;--}turwn5RrH~eywM=w%_*&e<U8Qg
zRlpN|(I9*q11%+%vxFUkRVSLUn}>%NcLX+ij%h==NMk2(K9SrP^9@3q0<k-hrja>}
z!5CMGDHuBxYXHxr!W>(%2A~&-D?kBBSi(PIy*rMHct>%1qC@$lY!Q2?27Si>I06E?
z%YKo*s^Bt-1HOP}<M#d`tJoqIb`7eElN0If1;o4#Sq`ysXFX*2zy@O}vDN$Fhe<f%
zWG)v0452!=hmv~&Wlk&M`rM^siG(RzPN_bpr2RQS616CKNGti9McMuf`Odk-A2c@1
z{scyb<r5h3iD{N1X7On_5@?ke43g6T$sDW`+_%|%rvlh4st^o7u(!GO>T6ONle8Y7
z;5l{mpDn+yim)LeniC|G-V;W>zMQ<Bgq_&eb1Vi|3JL849>U_nsSFsGxK3KZ#POsb
zipDzPC@5wPI!9;?vPvFBq>y!FANHh3FN~`a<d--K{xF>7bQ*uiL!9T4h6L~s*%rc3
zPWv2&VUTfNkofjih!4FO?GhpR>*(kqxDqIy#m+7dp`lpc*V%nvr^^`f5N!Z88M#Cu
z8N$68H46XM2Bg-tGuFgs4U|)syH9bkjIo>eOUNL{i+%AGz{g?`X^x{NK8q$biq?zC
zL+cApgDVkPe`_~51lBQxS7}V#Dgl>1;xTD{qXJk@ZIBb;3`ar5%v0pDk+_w6P~+e#
z5bG_Tk2;^&i{TsDL!5&ViqD}608UK4C8A3WGNd@-n}iT}0pvBN){E-u&)^_vh-Cf(
z>P-?t5mQ&u(h^~-Mn5O+Br?cl-*+pN#!Na=?W+oq!`p|hC+E!0{ioN`S$$iMobtr(
zkGui045CJdJA0hP5{~q4U;x>K1%N-`riq!wrUKOc6eM^Ii@3t`W~s@yEH(7xE2Oc&
zgQZ>a;MkQ!Qk9<13dDLAW(X5~x=R-jEkDD%p7<@HTFhck0PCKpQU7!(?jcfUs0#>_
zjwA#X>x}s)hzAz*+8(~XWH`sN0K2W9>W$IzNqJED0GSM6y@viB8A@WkMva`@nPzqa
z!tdR^4$*<8*kPQUnfM6n{mp>ppbe&=0O2EGEi*F66dcB$d*eer7W3v1C6Df;N09p|
zk=a_^XLMJmD;rsJ?an8OE&P!qJji^jQ6I!2Fp?cprz1ig;kKNA(SgO^6wnGMz#Nv`
z3R^)!+*}TYL07bF-mb~n^v@Z!z*Bf+BSVjEvlD$U;2$1d$wf1K9?1;26_QvsyJVpw
zfkQ7rkn7;Q80Y>>;~ox%pITq(hK2DsH08lB*ZT#uzU5N2>;Vsw?l>Gj>VZ@R>(E(_
zM`!u}XC2L9mH$$Tqq6}PG~oikaifoOY5E?xpsro0j{eRACGOP!%%4FDk%(Q@3n%dU
z-ns<h(S70Eeu^FJd??cqI0*K@rV}J>Vn$404n&(b{T(1rV+9@eBfAUM9GbA&vkO59
z42P!toE!;E4X&P%z@Q@nI+aP0r9y1!r7`AWIqku^ndRR^k=O{9zw%S;=AJNz!D{08
zhSZ4wj3oX+Xg^z4_Rm~Vx^?j;cE~J6G=vLo{kj_)_TMOJ2q!2F#fM@I{gLJftWQ$A
z)ay7TmN`Ro4w}<Lo}%JnBh)q47jpsunS9r(qjaqabB7S+KT2m1fNM~Z?No%PvI|Mn
z2t`bVpJKOp`9dN81byus;^SE`wGP1<j-bEDI=(FPNtWjFmzIY=;n*Z{f6#UT(_VwK
znw})cvaY_Y7+**RsJ=1*23wKb(cYelqjr{m9Yx|_l?(rm`NjX|o^6mp#z~ZS`S~5y
zrWsjWwNcA=lVYg+&*Vd%C!eyil+BiTjRM@l6D7l1NiVBXmWiPt?#UAx{jeC}B2t9J
ztPf^z=-eYja6S18!JC=8v-S)8<FR)??Si+-!-wd<bqGp^f{DG_wvFu4GSCm_1pcC_
z`8S%&f7;DS1*n1kAy-@SsJFDRIOI5elX$d(5jcyy_tMb-Jm=*Ks@a3^n4Dc2_~vv|
ztMk^8#gp%tH>?!+xl{Jl4R$wu@eNcqYu@{_h7>**v)e9Qd^PsiFPnt#mq|Qbvxeie
z*1L0pbo6d|tM|37QHqV16N^q$yv-(1_uJWxM-B|GzpUkdfB%huz+bdz^$WK5O_gx5
ze0`OxaG|F^Y{BtrP+g0|l_W{(DQBmlg;yBEVTUqOA3-Yxr9lI9`45a2XOox2P#m-8
z&vzb(di`2R3ac#^+mX_7+O?$w8A;Uh=X;(hg_EJfUYJQ%h%I|hOI1~MQ~unyAug`V
z0DVXDrg9H>pQWN4cl&+yRCQAm8&({PV@p-l214?>r)3G}$-)w;$&@!8KH}}j+&$5D
zvC1i1N!h-0#rSGatN5f);uqueDY66V*rBxeL#Z9JuzVUDpHk1xI7*GqILgZ2Xw8$%
z%E~$$m;S21IrFm<pP$1jO3#LfDHUB^PBa}k9P?27#U&?mT6{voWh>lByxrXqXQau@
zH#_8F9dOpzW}$6tY@C{&E-zb%`us8&l=>GPX{@ZRGnM2lk}2Cce##8#V^dUr`^JHg
z=Yxb);o7xp`?TKN{2=V@?Ts@l%iY}$6L#9YkF>b|enxDMw*?i2`PzzcDYVTtphWRn
zwQfrxa#Q`0_98)4f|_}bw?Hx4#l@93;%!!lZXFxvY}LKXK(@?LxG0=HeY!M&nj*06
zR2#EAc$}%JsdU$~aRAf0grO%Z0XO^(VtHt=O#b2_bbXoS5%>SoAC*&meB7^p<}xx~
z=gYxy9V)dBvtJrlQ39SuaYv>JU#?GjHx?H54&-|+{@g5zUJi>JjYrLw^dl1M8`2W0
z7$|kfE<Hw@Ek8C=ZBKw5G_SY!-+vq!V6?VA*3{d(c7URMn*1?SOIC_g&^ocr34!|$
zE`X8j>C<h3;Tf(QTzmDmGDH|=h7{0I$X?t|MI>1pi4>ar$js5@K&@=Bk87oT#Y_w3
z+z#%lf)AptY_5DfGSu8%j%Jkj2NZg*kg?pojesga?qjL0t=$a7>9cdPuERiDb#pU2
zEJw|YS*{g=mJt3+fU^9HneRBs1uWi&4skH1SjMh@bg;g@ek%Z9Po!ZYz!;g)8}R_u
zjUn?V+&{10Xn48f@QsEI^;(vv9osjBAAf#<mzvsG%HE!9c5W(Zrm9|TCq8SY+gR>y
zrEQIirF*3`OPl}wK3Z?LfW(q4f)33c{w&5)c3bh(P%ZxUApJ?Rskdjp*gH7rgGPM2
zl8u!$G9$yb!Pvpx{xWEc=4f8|YTQzG#uw1Op-<}-w*JwnzW#nM;2=ep?r|LM@B5cs
zQ=F1jng%##Dt@(rWw>d+<mIWsg(hrj-g~pm8n_1b-rjd+T;HduKI-l3(+BW;yE4P1
zY0JdK#Cx5luU}ap<?NzbdhzAU7el9OOGB7jUkK5`i3`T+*4R24a=cIGYMaEZTIbx~
zu|9B=ie&HA*LwVp%B$t3=bAn&0sk1V2<2^UZ83sf6fg!yXlQ84&9-+7NlmHORmtA2
zkY1YK9RJ(FgD-#T_tv@x2fx#o8mfxp+bKgG)pa|j9pjpIs3&Q$z`wv9{g+~7;Q9b6
zQygbTHBnN0w<fQ6<)dHR^XG~w7kD>iY85`(&*(6@dE5G`5^IS%#?PitwxsDMcwf&r
z?dcP{FRp@BRm8RCL+8q$Yd_c2Y}~(p|FK5XoRQ)km1b|+?wY??&8J(U%27S5Y}{|3
zAW#=|O+hJD>ReE}zql&rV^GGaC?}##w43zow!Q32)UHmiNRmBInC|``EXD7(I(5qo
zoZE#p93ji<%jP`w6b2}du?wuk#JHZF_KZv1FClh+L)^iEv8JfO;iuUf>o|0_4vY<)
z@ZSkQg_VQjj^tSfvaHtm`i#-UpBgOJ`f=c6No~h7`O~$9<UfmkIjeqSA-m>KW8>Xz
z0vl4MM9i`+DQY~nSI_8Ghya$nE>eLVf{ZN2JVHO4ql;dq80FSKe%mHu=9)YCEuecW
z9LMopP1)7{J{>t!lY=J?y?mJ!ySu%qwPkL$fM1SrwtTGt2#*o|cbftXb*ZyAt31T1
z>w1>YW@<<FX0?6fi~4(fbNaw9$8-|92Hu!9o1Kk;+8%tt*p|Ld8qpWs7_QXa%--lU
zXlb*jbOWgW`etvKdekGR&tKJuJSDV(-Yd6FojgqD&c+D06x)b^pZKG)+e$;deh$N!
zOr~zix$W2@k1uao#h`_mdzvC+rsNS%5f%LQURC*?S3f?r^%pA_mrJYc9kN#)rkvx!
zMX)H^x0Zb{*ljq+z;l>RrU)<o!{hvB!?x`d7A&rhOs{#X-W}a;EN`*BcC~X!rfR2>
zD~%u9U{Ise)>T_9>1AS=HS4QCs|NWTemU8p*JHCXU&zmezT@wY@{QU5lhyTvpda;9
z`C;*j(0DqVHj4bSAbleagLPMmj(@TC`fj_A^En;XjH`!VDSO+o#6ISvIK~&NQXHcX
zLir(6blk*e`0W>K_PDr5j|z>Cl1C%aQx-h?gxm3@H3uzU?>*^LIa>-xAC@Bi>z6s7
zl35cYrXUiymwjd0waO~$rwr<~ZqC<MF0qy(1@hvU6QtT{6Omc4J47JQn(k7&MI9q9
zs=oBDh{DDu?p<e^^A~LMPmT=z+{LfKfvYM$slVdHzCcgAN{2}^m%hZ!OQxoOKB>d_
zv>x63WXAgp-@4XM;$gk+WD7MW*A=HWJ(I&8naf|4Hc^>--gUV#(Yty?IeR$795Kbl
zF>#HT+sJ(Otv~nCcz^f(#ctm)4oU2)K0SA-+P*8l<HLq-bWXf3v|g{kZt+`=iT1^Q
zd*kj4KH6pj9_iL5a=B*?i<Yu6U)(8D;h-nv>o(y|{p1$i{D_P-%k)xSVxoXaR7%Q*
zu#!k=JJXWkv0rE!*H4Ijy|R*(*WvIOWl~tjVds%;N9J=LxJ@3zYpto0`TF`X?Sn^?
zGXpBN(^WHE*7xe==H@3Ob#_0&2JQb^PJR63zuRt(KPt+Z$V)RE5-=4LlRtbdCD*B&
zy?gbOTa=F7wpRxu+@_V2e+LuV&0H{8H8)g+A|r~cZZ6n($F(C(Z#9Jhg1+KAcU&n~
z;F|5VZXFfXhqbu&ThC0;Y~Op3ErQd&`zg3IN{t3GHajIQI_T~AF8sUrWy@2Zd2!xb
zeS^1T^5zR=?j2~jxxi5{U3$aS88qc~bZLDt-K8uAyrI-Gw;L>Nk61Ii-OX}aSU*SZ
zHj0w=xwZ4HI(x^JL)mwt<y>e5^6*wmgF}n_&NF4Y#h3}cIQ?U{Gz)+&=O1G;cCbVL
z%umnPdOV{x&J}MLQMk}O6n8aX(wp(?gRc}0Msu+0i@{{M3}ArR=OEwhkS-SO3fTm-
zL7_3`r~zBgBUYxDObV;~MfNRl)WvPg7vy5YyVk!rzbR1L=6v~+BL~3$O4Q7Ec9KA2
zWF&lPCLGC;Blo~)#4B&Bliq6^Jfc>dX!c$7F;ctV%KVW2F?|)lcs(tg<s)e26rr)O
z719?AS=rjUx*}wJffm-vZ606GMn!SA7j9aoK>MY3v$N^-J#&d7eC8P&@G;`g<_kuS
z8XT5s1*>bsA%Du@b|tu+6Ac&vBr97UE-f7sznz=GI^9M=p|DOvA!oF-rS(>4fcuc+
z?)LbE(%DmkCnOGQxhSN?7n*r}xLZy`42L9)jHmK)-!q8s<IkTT-Msf#FVTjtjV30H
zK%JJv&2pJW-g7=$c-$OdbK}ilzYWCOHm5>F%!g+{1k|6p$=NN_!_hyd+M4(%<gbt0
z9cpzJ>EqSm8D!rbg`1R4rEtY@(atT_+9vFb&5sIJYdj)<<=9o~aWq9Zz<{d9eh@^9
zYa=jhOAH{CydJo!?rE-MyEb>^Z3gLP+XQ_X7ut^v;=$Fy4{WIPI|22(%wq-IM>&kV
zSg?fscMCaXIhwn!i<L`Xvme|X{o&qZc0}S3&A6>{j$3XNZF)XigSgl)Q?ettsoY$P
zAb5xpB}fq_FA6rK<4neN;PWe+ImD_kE6!W-cGBmIfy#^WGi}qbC}LYLF)TyIBvrSz
zaxM+8U)q4_W%!WF7I<t}Zq%MTxUw!wQ!c*w3r+dnp!5~#F4XB3p`XQqp&B<D4XFG>
z78yLQ4H=%;@tSHDlo~JhcZynoT9-JGx!Fm;2yTzYDrNh>zaBi2e)yeiCof9aik_Ag
zqPvZZjM|05XHNa?<&FqcX8t{#d~(@!|CEo+(AU?yp%UT*bn7@j!ntHVK0aQE-bp(M
zuKRkY&~jk|HP!vuvuAH$);H;}fGL#TWG1VrsgDrOU?14_pUqzt$3A)cM4jQBc%1yC
z&(+j}ED@+e^f4CecBQXKySvEXn+}iJZ<vw{(&l-xAyUxR(o8c_y8+6_2jN!elJvnI
zNCD~L&Cg(Cq@L5VihcJy1YdHVQ&N(3S!!8`_^sPqD0p#D`SvPC#@4x(smp+bN%`>x
z9ki2Yg=AvNuDJTR2T8j)OMuscOCR1n1tk(E$`3&_0)s@%FYi15iU;CVr-D9>#0yO5
zcyO3xDnDjO=iJ<Jm1jX)@b;>2-@buR>R|_AY&!@@YJ_YN!_N`HMRpaInmOvM5=I4f
z#*%FgH*&fphwkgcd-6ek{a@?C+}vDbm7hOvA_W%sT5qmcACpVrJa?X1(M6+N?=7j2
zND_`4FDF}wZiB?tA*u+!D>jU6n5icAbIGToqCzv(P=(#5;nMrOu*xp(@Ja>QLO4W5
zB_tS^8^>E=jH*?*$GI(2y^~Gc$T#zy+6&!NjT5-Idxq|^?VE<fm)MVBx5Q!oH{XBl
z?(6eJ-)V;w03%;;FCu(m#UgPINn~_n8cHMyHg;tw44&GQ@OmE>8-m@)%K>nP+K##;
z=jBNOfft*01aSm)sC0iCvQTpX-`j>OjZ93uzt~pLh6?J-Bi;>dQ2Sh0*D1yHeBYp&
z*-({`IrrMRqn%@n*Xz~g<QrQ#N`bL+;QChA)iL4aJTYBZAk4n^`NEbnZ2OoWznVUI
za~1E@+;~&H3)_JKyFE#wox$#^YfTO;^u|mtrVo{@Q3!otYVP$In+M|}4TihW>*6MN
z81rUKC7|`8UC8}Q7LG)p)clK!Lrc#2eZZGCgCy;Xf;s<3+OZ-mtUqtZ(PPJsg&_Yo
z2Yyg;JvMa9wGeZ8D_h$VumSFAU3;l_rL_IPTc(KSq0b~AsPtJ-S`4vou5B2-v%f=k
zLkAZw8i)w2qmC#mM(!MNY<rM@<kGv}G&3y>Fim^=?CfkiP-Y)=y-NV4>7$26Dg$?3
zFR24qkJzQf(av$qs0>41Jsh_!qT8Zn-xEVnjOp4~DJDXBlmK;_tYN?U^#T2v9Pg87
zM3lwJS(TfN5pJYd_sB|*iP?GK!iBr31KF{?K<bp^YbYn;gBFv5Fn}2ayRY+j6v(#f
zWKEfazgh^UT}h9&i~&}FdH3q^rHn(Dztu;%DCnP>OFGc`iVt5a*f@!WoudUOHmM9N
za02=#tl-UkIB9#2bsq|m`{+^#g8}-bW(yrlja0tm1x_Px<TZT1b=@2Ew(IM5Z^#yQ
z{%u33FitnXmMrtPFpoLKs;364a^2Y%tMzVl4{evE&NfL+H1-U$*5#zDq^3zP-Wu_G
z=W8l^m`Rxbc&NZ&DoDy!v?DwKkofPV=8C*|;c3ymolo;`TXj~H%ORPmU6d=lS;F#_
zRoda8!-G<^lsZ5*<8*|eG8x01M;~t9yMjh^MNTv#{5Dp$-_ukx5!qXToQV7k(x(aM
zv@91)F>d9)=Vm4H=c5{%)?IANzo@PKH1x*D8?I8Az}P`BobO;z%y*gBswl*0G75ZP
z{Jt|L|MuvEGNy`$!D7OZ&z=njRXz!QB=<O!+NsGTqX;90w+%V8&PC?uOJi5vu|rN%
z!@tg**D?I->!r8Ud0p%*3mpFIFLr3$EBBFsb?6bYSPAQzu#J6S*Sf0$z2(%hUj;u`
zGYXnqkTo%BOL-ByiL=|G;?X0<{QDmEOFN(CQ#0%0GGirQ%fTG`k+<ue1*E3mt}g~B
zvI8??PGs+6-o%u#%k0gNXmS=``_e)_WA1$qhkra3ay~P5(5`P+?miv7nf#vz)7ysU
zAB1oHqu6wreox-eacO7tpKs{o<b+@={bL3_EY%mQi_h%(^259R^p*An-(MFQ;uVCI
zH1_Vs2Lpm$(&4c;$sMfOA`Ay5(7?z&m+5;r8TGj~Ir#eg<4MQT{(KTS;<dn&HwVSI
zgq;7l8b2T0rrCb-V9HcCZ;R}%mL=Zqp$M^ab5>~<eqSlt7ca6QKIvc}&3+Oa5S=y4
zLqvrJUg@2>a(Sp)CFlOBoF$d2)}z*OlRZYAJeG4Z(%4>PzmWqe1fk!&h)F|MwlIIO
zFaMbqi&x^Qen~NpBM<LSg+IU-^4vdOrtnifLVDP3eIcU%7HM4M_SVIL)^_Zd`y^2s
zcttsJ#)Y;nrZF}l;P>)Xq`@_34svI)+*a+);WOpeE<>t#Og)zR${oDo$v?jBx~f1O
z`|DOVHpS!sj`ZaBz^LZ^ItT{S&0SJkrDv;sYxvhkTp71L7iH91Y|^@a{7#qE6@HoZ
zB4llr?oB3o30@Eew_FQ%vT4T1QgUgZaY>z%SGgjWlkbl-cT2i?l-8BQGWkuAb3~@4
z2?K=>wAmxDi?98d!Dc!2!nwVhX2oBXA2#*(j9*>#z?x~~=|K;!z3mHB`W5&koP-WW
z;1${H=)KNnhKd;(ouAkkv4gsF;Fy7?PcM!9^N@6Np|lrSYpNblb&fx^koix)JQ1Pw
z$0H2}dClU~nausuqO#n)ZoaUwMzKtJ@Zh;Y-D#BC$3N7*>fHt4!AmByyky&+reW_8
zv=%68SM}KP+(Kn((%U@aa`IP4zlC;GCLo2rwVRj?s1+Y}jDph?MeTWT7e&0y+6o1I
zMDMvdmFDy|Hk5>vII%h@+J-+K)f4}7NZGA6h+1o=X*fpFm8M2@p!MmB=Pa*93ySF0
zGK<97tvK``%xH;$>G-Fhaw`6-$^`{ScL8m%cf7YY`pT)Fl&wO%df&Hi3XxqM9h;IO
zFndr-ynwl^^Mu4uM$l0zYTCz8R37(r@^{_0=j!D&`XHgBu6OC;gC_P&s;qbw{pUm0
z(q1VqnDuwtV?*7i%2j$O@X@^ZIahzWaRU>RJ9}7wV(HBfN+ksu^E#|Jbfx_5m)P!b
zs$xyS+bd`o_`>$EKM?#9ODSl;9npWGMCF>YlS~8+UqR77>3Y{IC1n-GYrp5}9k-YN
z25-2(Adr+eUM9;%zQ+b&Tqq0j4W_Cc{D=#xEXZKf$ma#@juOXw=3)%{ayfsqk*||@
zI`ZB9tiFd9fX-R<5^DXA*Ax7y0etM#x0Vpt7LtCV5%xdie<*c$@QLDHJ4!qpFZ{jg
zMkwc~saSbg5<1*Yy%Gz?_^Jnt92|_h!!$JmbW~NRMHM-9Jt}GO=W-TJj&GkgRk!Bb
zTBmN%uIqnC#ZpzxXg*O#>JJkKR4+<S@xx<JSKQcRTlrFK+Xw&Ki9TPxs_o~-0yHr>
z_I}JYqlM3umM;4mh26ETn(i1byNIT=&MD5Jv8Q5iIXDc*J4R%Q=9qdgFf|LR87HYp
z4jnn705>_uF7Z~yrzhg&zMc_E7Yn`w|2mec2F)NF^N4!+**PiBR|(0<2heZc0>S+8
zLx&C(B5%+KpIQ*4vK{r>2qq~Oou&Tq(HGcnu%35vvbUC#y~H%0?dkba3}!hGRsl)g
zw0p1>6UK`sMr<nMq+rwTi;O!+%E=xDl;ER%=Qo1=b0&LV{JHZ=+^_K=aJ{GylSv5w
zefN;b6A97w?AHBBzEd;GPX7w4=J!E7_Ex0t#p}yBP*V_nNA^ABJ3IOLgGHwM{)=3b
z{d**O*%Qd(VT}5=@NST$zt8U^|9y=3Ehd>RSWQY=)^%zlAT|xyYD<OtI!bXeGLmE}
zDJjW3DfJ+l`JZ#`O9nF5=WnoYmz}yl3efu8QzSk+xJyZ@H|w<P$r<N*4OR&3V>`e&
zAARYfPWlviLYJ%oWeY+}U9&ahsJ6EDV4b+V!@cmUA(aa3mR@tRGw3x(to?*TF)KS;
zAK-8b7_!~Kg28VDYCY2x&N+Vz;uk@r7bE<?^Ic+P2`FBDNQNm`;L<NIv@#DuB)0)a
z6c&s|2`CHtu*(<3NHfhhH!nH%U2OG$7NZb?g!bPyopGNV%9kE^^-<vShd7=Uhhzeb
zu3x|Z=+UDcCdf29zE^_l<qqDVm+g6qK_xLBI!^Y0D@eTm<;UV8&jU7vzbl>bOiF4*
z*XR03yC=u>z?Bs4%WUF$<+;<B73S+*(u%_*@8}n|IKJMEvu5q?`cMlGZ(XXmeSla5
zr=GN?(%<c(LW$$iI{$2cj0Z*U2f)H|5ZWy7H&R7}tf?vp5^r}emK~rNBt~&xZcKG=
zY<#DGPne?)wm-$q<2?f^7cU-(PE1edf2L%n^qk9;vA5gZ(8RmD=YIFl0g3eVwbg3N
z58V^{<8rZ=BfEKFt$W;kZl549V1Ml2zLHPr;>E(Gr^aGvmi!{F_MV@;;EhP9Ke`y_
z^12va+H}YaT=4_x-Sx8YHobvjx5ke=1S-`e-H~JUR<a~}iU&YT*}ON<^FtT+voqdt
z&-Y0D9J^x^f0smY#UB<cECWcuSKf6)P@4r2gW~Vzw*MKe%}2I|%VO_bZ8M2yjWP}X
zDoQZfgL<AlrQ{|3sV<(<0Ed_oLDz6O%Yp=bgMuxe$uvX?^Ye{E;(=vyH@<F6xu(>n
zV#lwPo<dDwUNcugA_IlOeW^7w9`Go|>o=#a_(v2;FDMJJ6on$u5;TP~{iWoYmZ`k0
zfZjbw<_^)8Xa&;>%JbJlT6G6$?K!cgQb^Yk32Nlir#oQF$Ogi{7k~_UozG~lk;Sq*
z%#I8Emd!-9>A^%zd2VRNDMl;yF3(c!-Rg0ve_x$S{s=}x{|JdFn9m3%6JP;RG$36M
zeYH(8T2nRGJuu)6D{(q0Gebl7mX?+s^`t%+fn8his;ycZ;2?ZYA4J#=%->UH=oi-r
zK;O*`vJr9YAk>e0NrHg-?vez?@tK2-x;~R1<t6sDyE}wT3K|+3YwOe>0V~P+vt@|Z
z-@kTEjwM~ko5N)I00W4>A_J+P+r5M;+y@5-gHW12=(}+7V$?+F!uD$jt=P0|W3zj;
zH8cX}&hbpGrPN$nTOvg+7^$`)S!jR>9Z7qjGeSGHCnk`=XFeB7mgTPmO_6Hc7K-c0
zE;KGl^?+ci`BLvFN$N;%7o-3mOdT%)g#`^;)H;H<Mt_h(N6V^O?r(*TY7#qML2qq|
z07|6Dm>7fOP3UflOj{!CHv^|L2mji~_Bw?UT}})r{-=${|JPS|8UC-G$xeOJne2@D
zx1GrdEieS`8X7mTh0*pmsu|K!5C=FbB?Dh@P~tIHjIO#N=<h$aY*W!VlA4&5ga+p7
zssbrQc}J<se``6JQSG46y<A88GF<D79*ryYvLDnMvUeUVTfY5IejKX#M|l#(J>bE6
zWK#`4yd<=&tYlNu<?+ewT0X1VZDV{`7NfDXvj~^EK)kVW5<xE8hN<xq`2fQ{U12A`
zoUz`B{!<J2TA{r-Oei@~xss7driP>(<DelQcj!YW*XWG<b5(95qqV*>=oQf!!1&J9
z_G9PDs|*$YM{eW4-sfX1D0zbp?~hDRA8x3AXoERszoo9D+^w%>tZrmYG1h3irKHpb
zn_T`t9c_y8fS1g%$3}m-t2NyCh5caWl>th@^oEG1r-TphGNz>zqvK%^|Cb9uIS%&^
z1HONbs9QFykZDAEWUw!AQxH`|Ga;4y+q3Dv`CHeG|MF%0Ysi<OeDK5C{qn_KJ9}-V
zgZ)}qYg%NE_eNSRM{ebdO)bL`dgQS9N*ek8RhE@Ijblg29otz;Zf7aU$^Dfb9Ave>
zRi$TT{q2Fu{97MZDx3;sRZ&&7EAQIBy?i6F&+^yg-z{+MF>%lO&!_Fbe53A5kVXHu
zvN{T-<~z?jr}2r2B1qlaA1Z9S4?qPW%oKzO0r2aEe`M`Vdc%dr?BlawyhIoM$FLo0
za~_{|4V677;6FiwQvbJI^z<nsW&;$;nmdZJN6*oNJCfaMO+fw->o4b1467p>PzKvQ
zT1}DhAR+oBe<lz7f>{;V&1DI!ZC<CQQG^0Bz18|}?QVk35@~jJtp`-DW76g|io0X9
z<>hWB*QCG!DfJ|;G2VSPi)+T^jmV&?*tj*V#%B!j3b$eG`~jfRi@<=3ow|!VxwxS2
z1gFwvCeTH+e}ZI4@7C-tgi@4YwyWZ?&~E3+*Wm%_4RjXa9$h*rKJ&KcC8pv<1~+wn
z*zDs3@$`QyAKOH6^$&WN;tLT=AxK*K2+!4kaxjC)x;#u$9Lj@j$lLBCmn+W5n@B{X
z`!wl;pAl*M?G1!ypZVD+uI5}a>!bb#ZhZq{b6YxDGv)G4BZU;4GAGYo{F13uWSPvX
zu$a9s%t4KP>b?a!SlF7&+;N0*--iTC8L9SlK6C~Q-|21zX;x&f(%;hv%ZD7^i}RIB
zj_i9+)j#z)=>necGE9LCTm(|Sczf~CW8V^}$bGA=t<_AdcvDl)5xkfrGAqV+pCTzw
zk_Cb~6UY#_)A~jhM|29j0dL-9PdHg$VBQ0H<V^tGg3#A7!F-JU4+0N4+(|dX9L5>N
z8hN7u{_&?Z1gH48+97}U23RTya&gY)_~_uCK87<c8thIS0rIZfxKRrDBqU|=K}MI=
z8$TWEG8=m-c@8wXZ3ptEFFW?@bFx*Ao>eP1?-|N-friEd8NZzHnfe2hhBL`%VvGbN
zJMQ$682#{HUz)r?2q0NT#u4N`M_}$&v_Sl-l^jPrb-nUup5(I({q3c%;w&~tlcQ*F
za?Zgd>qoaQ57r%U9`VGQHlDBMT&T7Sq@B$=R@KU+psNs*;{A=+@dr2OeUf<OPvYN>
zN@F7<pk-I{`-nZTAyRE@J;8*BH=6K<^F_MzcXB_lsLqPjJWGyhUR@P643*fn;WX7V
zqC3Md>s<udO9#?_U!+I3@{PkBM?cyjnITA*2XdEUpivz``aYuF?C<u0HRX$uxC~OQ
zu)e6RuV3$Q8lZ2^!ekS)Nj^amw?^#(gR^j#u`q?hoM(H2j+p7l%H|BeTSuT;vMj-O
zx|UP6GiuzLJ2ro>Po9s*=gOezn!vo;eDn)2LSk_drNOu^1evtCi`lIUrb~z@#ZjrX
zfJZMb#D>~jSTb`Sd$I^b>-Ex$*A=}kvpuo*v|b~fS(Ki8E!{VSf4V@RmH-%rfnUG(
zr%rAY=lomF@ud=%`4SQJ<7uDkIYRzec;2E_`aghe`ojE#hpcKlj9Bj=zi@Xao(UtQ
z0npi7GcnjvgKCEb;;3yB{7wjHxVL1<V0dY=ESPR2`%x1-D;#opkSSVG*Y@UYc6)bU
z+@UjEur*#uV)}pPVr}ku`JTZO-JrKF=}5Fuy^#eWhoYGzXL*xp$#SxE*Ov-tr8el!
z)&!nS+7^LLBXVdr<1Y_=#CqdE8GZjC9rCEcp{2#4lJ(BRx9zYQA~Q2Z`7?j`xx2re
zgw$8E0?r|C`d^i`|GJL<c$AfEWKZAsDjc2n*`3C#%?C5PiedNGw_`wo!}=U4+Re<&
zP6SZ-`{XgKG~{iBRfyDdz)yof?ex|@%!WC!+zG^@UQa~c4rJHf*{UO=6SKlAD35{R
z(`E+Vnwz@D^0LHn!W=?(1rrk!Ep$;yuc^DY*8{CG390d?>*c`qaco<s>plgB;GT2z
z54`a7)2??_Sc8^qBPKsRIZ^^0Mw@1iojE>Ltn!Y6?w&rq1F9l=UZFOP*Ud3!x&+$u
z5Su*|eR*94VaA9{?2ZV!-UKN9Bbrk&3(bnf%bj+?zeYKDSH?hvi1XxsPrCL$Px$`t
oEBxPk@xP&N_&>3X3YOMuzpkDeT19;Y#AS-2+!@)VU-W+cUnM$6?EnA(

diff --git a/examples_skyline/accuracy_vs_fpr-skyline_formula.png b/examples_skyline/accuracy_vs_fpr-skyline_formula.png
deleted file mode 100644
index 172bfe9e4b8dbca7a624a56a758a5f5fe1fe0d92..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 43487
zcmeFZXH=Ehwk`URQ3OnYBEbNXL6jgFBa(BJpdvYE$(Tz8Q6wl5mLeGzIY&W3at6sD
zk|ig}Z%nH8uCw<!_uPAbyw>i0?NzI)1&bBtH@`W@=%e>O<}(!~8Or_i`!Niol)Z6P
z4Z{eVF^mL9P72SswK})K1F_R(Sq*ad$BX>#AMpD=`y0AW7)Egz{YTK+@3RF@UUa^u
z<E(CH>g;yg(FC)(?QCyt=WK0p=cKENqmzZ5?FHWRy!_lJ&7Ga?Mfv#t{yVRoqZwb!
zqG%6>oy26XUedT1@vGlmKg48bcVX?)LDqvTif?9qKDkWD`&KW`Wa>an=*#@q78kyD
z4Q4ASzQ1OitD_aF*lqm0%DVN;#Zg<!*fS=kA|vKTA>2G7<f3Ci`wzdc-`*LT<~aK3
z#Qn-e4@o9*sx=q<W~tOTDc`_Aa>FrWVi-P7PQ1?m9<k<E2XO@ONI?$sfccRSo5Mpn
zA)x>~+#(>Lhlg}Oj1nGnr2Wt&ai5eXf(P3FGv3zzV6-lbi-qNCX=!OshOta)DjN@v
z;^)twL&x8XIK|nPtZ2!qnVQCVd3$T>>+83T7BGl7UhOjSE!o=IS}rG|Ztw9(I(_EM
z6`}WvW=&BpR*V=uud*8*h6S`dq?Sx@UmBYndvE9J#u2igl2X&w);1M?Mnpu{rYE~)
zvMt^utJ~7W08{DxPDF_t;IB&f^yy2#iT>gEw#0AWzF9W(9$WJlja0gJ^{NJ!xNy}B
z0q*u0X`+PYU)@<9iRx_Y>+2k+PhYMJ<8v!n>%3`~=(#qFU)x!V-n|h-#U!4<Z(NmN
z)0<nkwOHq}*6$|I=~?UNJbZik^FuN&4i4EaL(fz%FR$h8wLz=T4@mGX&cD8^+Lf-G
zzT42k>nFZ_do0JYLoQO(rFCOvdS$?KF41jnz+|HYK8~*c$ku3>QIhxOs)^EVol>uY
zj~_p>$oA&is7FhBw<l?G<ydto`EImDbLwkdx^#cq`HzVSbC;=h*`T|5y(@iAO=#e$
zH}g7Y@&@r+TVAqK(N7k(mYR~`6AQOjd--%`dvdI34jnorB_%~dLSoSvAu@{J?a(&T
zyn6MKkljFv>r79`x<XwjkGiw-GRac#TGns4@D!`G5-lGeA9^9{w=WsR#-?_^(ixeT
z`w{FXCs%QH&TegO<?Le$5^$N)*xsJ=?V{i-K51lRWb-}!5`0VHT<K=}hbv?*6U~=k
z!lL)@-w&_8VAiPN<yDLp1I&1$Od!+7YOfu-h{@+%c$FJB{*c_5IMXFM<60zS{e380
za(8VmjB`_h$dBR}mQXIWvq1eOGScSxix-NPmPrc>3;a*17|-wiu9i}OAL?^$dKF}4
zpLOTiMb~`$b`F-2psK3sSn1~Mg@S^DvG{18^Y4Z2gsP4$eL151_R%Bau1so%i?>+b
zef)UZ*q{F)DdqW7r%su!&GyG7Bz!wcOB*g|)%kXNbM1Xl(BU_sp{9Dd))u*TgSP(s
zV|9FG0;zgA=b62iUvu#C3hC<Vj!mbP3Bh&6xlDJSm5}Jq!NZMT+S!~lTF!F)^}WSw
zxkUy)nPFTVJ|5-Sc+6O#%xAm4Xs*=CpZ_cyTeWq6VNR<JTg|WDye2_LT3S=s4T`X1
zM0~dLkDt5cCo3m+zUo+(Mcdi76x|C;^Oe-^;dZ|X3v>It*pp{xC%L_>4p+gZ6k%`k
zRPnvgI~?vL{H`py{%Nv1D}H_WDYHL+mGj&{e~Lj-XO90(+k&t4?=DmyQ(qld9ccP!
zT>TO@SSi1|ySr&K?$X$687BU!V{bx2ObCgH8pxQvZo#}a$S`}~h2IUPQ@#arsOv{}
z9W&n98kc%+g)gXr9sOks#uewgyEC>>&75ZMDZcRKNL6odZ;Dabz)>2ScQtR=P89Xq
zM!An2SJ%-o4|?+^px>4+t3+z&w{h3<SIN1rK1~=-{MUU_l@Gsv-$ToXiHV6`%BQ%{
zdGaG{!AU#c&G$?`zi)r7s;V{3N^}};+)qK#2DjdMqWM(G?;kG;e`VISf4oLv;MVV=
zX*k%Ps7`bIIES*bGF;fCaM^l*sLK?~`-Zqn4@oV)r{C@=cF$t5*x6d&=%4e6|8V7L
zZDr+^OP4N9b*8>AE-4Z8*}BK0mwh8B+pLMn<nCSbFMk}OGg$lb>_|vZ&}rG)nwl2a
zUV;ukuF3|5g@vuo4W_|F)^v4gZ*Qz5H;%RUyA7tkdUcS5l$4W$qb5gEc5!~b`Hiw9
zhn<~Wj&=9q1NV~R;^3$#d|SM7GOV1bj^wau31k1A<#?&dm_W(aIK}tvxi&XtThV;?
z4#PKZz#I!&w0`33@YtNmOSbC5TXtv0Rt21hj}&v03!>%G`;?!5@$1*GHE=0co>FN-
zB*}K2(Ub437`8XB@;_eq<2e`lObNI<dinOl*Yc{%?-L1n{kC`a@Zgk`G=!bmu{_x}
zIyxG=@5otp2+M5BQP~<#KfTzAB9L~66veqbTQtXc_Uv^60u&LNXZwp>etqX>6u;N(
zVp+ilk1Drs-(Fl!8Ca=-O=S|-m3lMYXM2O=+&L9^i(Zjy>a@m#=a#SJ9*+F@`ohG&
zrQ$J#pl#npcTZ0v=l(Zm6<ae+>NHeTa3*!3iP~v~aF^d`(07(?uik`NQq0K6(91Gy
zVD&9cPd~>f>ikWxA1)}zyybGx@~^C>TDSp_iNb|!!dWaPTVv7pRa8_6m^Yua=`ZXM
zoP&iJ5)qLC%UNS*X9vG})2>7_Sv#%s-37DdK;Nw^L}XO$A=-wzv3k9Ec5@p8)z#H5
zuRpX+wtLQ%jBi8&C}}o}_KvH3N`1437J`qX&0uYAwa83Qj+6IbripxxMceg;a6u<U
z`J~RTujo(=TZBv<TsP@F)hIUCUhKZ)y!G25Qrull+sHe|pMU?peJyT--fo}8VY+JT
z>$%rHrq%HlDZr}5PwwulmC-G)kCK=|Y~7f#^G$~Rs3P-{p#`pkhK`QS*x$-vwBw{K
zEIRe2v4$M?B~u>5Vx2c^%J}7_>sPO`b8{<9cc!+%RM-7VoN0}Hp{Ay`TuCi80o%Og
zdxo*gSp7S>F!$1jWYmSFP4)E}a80$_zXByz-#R)ug}i^CEIe@UO3=|0Cz3<Bw5Ogj
z`!w_9^?h_0sVdmWv+dWj>P%@t8Hnd*Zhp_h2M^lIe0||(1qwRelvdXS1&Hlum7`K%
zK!v4{fnp$M-_%orFZt;#YS2d4%haJ7GANw5ICueLrzW76E`S%EdEYQ5%6JPhYEOx0
z9?RrQvDxMb5vOeLP1obb*}l6zB9TffGrii{hVHR(%F%d8FYHoMgESL&?%Y|PjF0BL
za6vs@DYDh6Ny4pVg^rds35S!07!nd6AK#gJQ+w=m#vshBUiz*3LbiPg5adnF%&g`H
zO7LrTMKePE-fR7fi;GE%7A$Bv8tp97d2dX{KMxFa%i)8N0GlM_>C>l#goG?1QFm^2
zK={X}q;N-Gbhr#Fv-S7FNV0&@?b{z1sGcfCin8+Y@c{~$day#xB(6jC^ontne+z^y
z{h7Re*Y}3^hDl)U$itHv`b+DLGaWmy$pI;)K#Y%#jpg9ts%v#Ts}zx77Ug~&ZBNJ}
z%kApQn)hBCxqSKZg<Vik^V+p*8*62|scVD2Wqm#&+kb7<T1fsM$e8X)LnhNMa-GSx
z>f$x<oXH82YYI@3lWYAX`{w8Pcs%5TYmhutVC>l*E0!~ZWkw%9eArJ#)fvPp&ms$P
zqysLAhK43~pu|(#Ln$*lI@-{CMKj=puoA)_u(Sj%+u6G|0eEygAfa&H-CnC395jNo
z%9$nOwD_|I-NjfHA&8aUi#2S83somN;uKG4ZvrSnAsW!b_ClbqhNY!t!8nJuDGUsO
z4^h`&2{3<?D4{GC!+Hwtx7AM8N?ooxVfSs=UccFFqWJ>~go2JgZ*&=`UJqpA0od0P
zBU1;P*0Eun0(Mqy2-gaPg(SH9XrsO0D>YxweHbXUlfSV%8IzE}zB18rrNm<;Gh2dl
zX}L{#8uGF_Y?u{53L5uj-Au~NFcqfv!xw4%u1^4vXr<BC|F8>*Zwj_jQQhl)+pfW#
zRpEZ;Hbs}xSv1@R;X&uRLINEB9clr;>yv4|+sSu~ji;~ObB&Lc4};V-3FCKm<G`BD
z=h^MBfS3KYVf@#*v{KmHn2sOU9TNz28t__DWMyH|(R*4s+`PNhw5#9DL`xed;kl~A
zl^=QlmeJ(oB*gFAIgQ=95b^T*Rv|Q643*!{DfPh;xFRZ0L7`L5!2l2c*Hsk%SCl>f
z{~-UbELZ7M64m3UAfj}QHN2-`Vw#<G#L82Bwx^Az=X`gIis}+V>+2QPH#U;;^S}4U
zVw#uzu0IsEACmT7Y7m~~jp>B4Q&zao{^8d4wn<(u*G!qi$aMruBqb#aho78)q-Iri
z1GCMt?H5&icYZV|l;3Ggq2FUVRZv6q>#x7c8>+oR77ta1*7P)~B_RQQKu>p;S--*M
zoM7!WlQ4t0!R=W$kA<&vT#>twqv<c0`bYU}SfC8!y|LT^;n*rh19SQG;iZ|h@l_Zi
zBn7IKsSbHmw*h8eTv}q4wXm>&Anv-7)jQ*{vp$XjIKmZB9d*zdd3_)tB4s~$Qo5z3
zMNVGcX@yz*UOJQSmILLnbJ75y=tUeqy=0ccL$QQPSU_F1-@e(e>_UM_bK(S-udgrQ
z@TsLHsW28E-OQ^nDh~Cfz5*wd31rovUS;x{zx4RYlO%&8R|Lj<Ac3Bb9$p`h*4|v5
zae=69G`|i%B?FF02H+fGn+$l5<NSFwsO~IVK0Y>Wj7Ur?-H?NVJhi8aRD~}>D$0Ac
z+l-E}Zf7;GtRQaCXVb=IZomMgp!wm-I)=wRS!RJxs2G#V%cW86+$6b~(wSzI25?PF
zK9u`)T3Xu2m>532UcmYSFib$^Nl-<=y~)#`Ylsk0hhd;hHwrn?u*73BIjuB^MKMZT
z_uad9XgYwwSgD}kL!embeZxxhquNjjz^(nRy*4o4>UZyc+JSo5sMVE}`+_u)ii*mu
z%U7;6K)q-U5Vir}qnWsbgrb^SP<OV)n|=HDJFZYtP+WldKYQUq+t1O_cYq;|96M$%
z`Fn^+U0>fCmBi?GpWnk|e^sf#B%I$|n-d05F3?}(CTvjX%+Kt*mBGQuc^(j!5TG8x
zs$))oFOM8Ncx%9W%|Km4<1Q_qp*4V&dRQb-_nG;_M65&kZIA;<zxvasPsfb?3!rST
z|MvP!!@@{)13<KfKMtL#hd&4KS6TOD*9cg3){8@hcNS0<j8rq9M-u9`Beb-xEi*22
zC2QdpZE+zMFZ}%l=E`<`I)p{Byv@;jS|J&~Qu78-@p|X${x+an-b6*&K7anazD-%m
zW@By64(e(6k~jbkaJxnJ3iWv9Xdxr7g~$9=Z`q>mRj{zIXe8#|Cn5{qUEAG>UI%nn
z53Gd3dm)>8D0~d+7M7O<U<6Sd+}r|urN2LB^v=Ykr#G(@pb>UuQvFruf8^wWYP8pD
z{xjJaa6hJy{l~(gc5PRr)Tv-H7a&L92}5%p^j@o}tCNSdplNMw{dgNb+wk$DM+vYx
z^|H-h|M>CaD8O!NYHCVGQ4O>ip$atEwkz9qMcfRO6dNmPzNrWZ#J#r-0>JrWWF!Z6
zr)(=AcO0)lp*aAe#pUI8sOLITZwjXxmgu42u|D>GudsxQK6-VwKUp>Add;zzpFi(5
zM2czOzkeT=SetR6L|l8jhBhUX#vXvAMrUSHLwRl*wyxGTG=!9u`Mwi^jXC=Br+%v=
zL3@eQh^}T;rqjT^5lV@TiI}N@k|J1vE8mR+Eyf!o9YrIdkP3kBOG6kDYTZ`=?C=F+
z^zwqJio2O>-2(eN4XR3ZDEHj@9G;>Dr5z0@(CXLs3_!@O(L2j*uwAABRwcYPF@nMy
zr~-8Wk^=6Fce@5`aQK6`mXi#k&Nra2&g{0v=ccEp@85j-=uv*Dn`op5OzzQx2ZuH`
z0HQhnp1-Xf4V+X#sa<#GC`lVoH+a!TBRd_Px0zX4tg?W`9Z4p>zMz{9y=nV8Ffi~X
zlSEQ+vG~etf3Id`n$HRsql8C$P6S}<q-!r|gFc~+520*vWu+tAqAjWP!%iS9eWhqg
zgSZ)3b}*w^#oXAei>-Kat&ZzUsPUM0f6Hi>LHsyw92^;$3N=v@Y^gAfHV7H|VnmJW
zqdLFq3S(OtiGko0b)HZO;4Xs2i=XrM@(PZQZW8&CpiT^n2^A(VW)pvh91~Mht?#>G
zm&AY~g)Q0%@y@`=tTCbmM!?h60v}8<Yx?@#yF_4ZWLuj*Tya^RP@U;3Q0L&_2p6yj
zVo3x_2bKA&zq&_~hJn?<-}3Qk_}BWZ(lPS&>(?f$-=Ipm4l6hkC?p;}3O~=t6DLyq
zO`x#!fc#arybaq&;8r;X3wmO6uB>dLkP|MA4D!mydeu~;mygvZqkcNc$;kzWgv3(8
z0xyh8;yz(7OejGAXlHw~uB}ZSVsaAflWgnm^MI*S<(38@fa;Za<VbGKUxA90OG?TJ
zGEZ%7ZKl4AzJ5e0)a|GuMQI1>2}F1tI&=L-P;qhbQD)|}r%WC<p!jHp3ug_P^)6Uy
z6mx<4VH)bpS1eGKO+%@gvREh&LyC@01HKOK4Ez8kz2hd_f4w5Nxz5&D`4D(6KFUK5
zXg|c?j5Zy>`J|r@g{f0)l30s?xjQ2Z^iMEQuoAMZ5ZuseD_N}N;$&x6QBrzYy0s`T
z8!Iz%{``3jH8sZ;y~~#`qlrTMrnaf6Yj<aE*AloQ6WjZFP=%rb7hMonTJdS1b5^Fi
znsTC^U-r@rrOZD;Ku>uYmEtf%C2%?W4;*NR3l}i0f6Li$;_%^*sEqD5i`E3Hmj-@?
zT%6G=b#Sx*NZ}mQ2KtiaR(ZrXqg-PSOoC-wTtIRY)N0wcD@ngrSHn$c>)1g=FC?$2
zuC!>q9Lw1EwuLpV+sRTu1;Y2LIXE}~lb`|AuAuezb6uOTI2<oB1j#9r=@vRC1_cGR
z7rIz^uaAbAH0F^u&+8nogW@SwhROZvk+TXaP-8i-bZA@k7rF$o0D0`awNQ=jFbWkY
zfVIUb27LmOPBzxqAkDCZS5H|^?m5gH;(Y_)o*?GYws7)n*QW5GS5_#u?nfxPt%fTe
z1C9C)?mLTYpyWmZq5?^XiOnJKhlWw+zi-%De0veY%wOTlc0BqWsu>Gsa(nv>o+BUt
z1*}!e$7{2^BndP?5n{Oeb#--ZQ0}HB;|ps6rzsK=6f#VArD@;0moMrzn*wOcXucG7
z^s7@>fPRbxzFvD;Uk|evt}3hoLexhEu!kU!LWnFBEd1r<MkKmx1$RxbkO}u;_~^?e
zi1LVF+xR^aNW;L8^8LF`?59ucr%zXOumMkY=DL3~&~nR7zB}p^wA{_WVfE_>Vmg&N
zWLQTv)5MZ>Z>}Pca$40d89R#IofO}myTql9=Wb(@kT3v3C5C14$qC#1A#W^U?$|Vj
z+aPZquW4-Lby`{*8X7t;Y!~(k*nYIkg0j8V9U-Jo05WmWW5e|KSBYW?l1;TmLtai}
z^(}~IbQ^G&L5j))pkGpSvnV!L+1R*1BiZ9F0fcqBBvQ%&K`t)-TdMzSdO+VCd>EA1
zZvpQf@ps92>RT6Le4;>Goqjs$A?6*0!{N>=E5jBp9Ga8?j?I21q^6;v1NIrs!Grdn
zGY=m*67u#fyX<~y>OAMB!8mqMd>~f3x*cDk@BOw7xpidruqCl^r00TNfur@AHA`+|
z1DoM8*YRfo0Uf2@o6bY!1T?g?oWNbgTo2%E8wZe!KhY(%*5`!S*zuQY1h}4m%nM|A
z#F_~JONsE_NL2vBTCklyXeMjPa&8+rz)df7ok=<)^O7B?)&@-?KS6&NOvSse3jfkx
zC(DrmpUm**xMF;FHwOX758O#frr+Vwerfa5(BFlwTBO;J1XvWfWHle3(wm)gb8~5F
zWjj4Q)3%n7_CE{_-2y(Ypy(*BvA{Fs&9S+u{l*3>!0S<DW@ZA3!dBGJ`@TVIe7dQo
z=5=(mQDj<PB7YUSXYi3fTE3YF^+|Nf`n$Q@&%ke7B%@D131KqK$fuxp2Zyu$ZY^Zs
zvp#Cuze$>Z#Mq#%Tu0lR*u4Do4o-k_E25=+!479JG_Ef}Y>vC1(QWN9be~MW4c&Zz
zb3A-zU3i1%&z`9&DEI?G@6w^E%>nruC<%2+t<9|LY=MAd{6LNZ`8Wa8ZRt;w#<%=e
zP6$~i%x}|hX|YR4l+<n$v@?UKaVKR{%HL&$dR&@_(myXR53ofXB1)m?NnCaGSf2Q>
z$1}$%MJ7QEHwO?PsA(E`Zz?gKcVp){4VS>3FXTvROO*24c$hJ3hCa&^cVF+P{SDWm
zpTnvfQ}+5B=OrX0VD%&vIE@?58(zQu#I|_hI?TPrw^xY%4A$6?8G$g};YUE!4tiqA
z?#?C+ukZZOXZ%e%Mln|f*CGSRQo*3iO=y>GgdlkU_VahaCKxFo@J%3d4S}lDwzb8x
zSO6`G@`1<7)TqbD{;$OlR|_|$l3gIYO#{p8kOL934VH12OP4|Xvm<Pl5RLFnF0)w*
zfD#TJIs}=n6{(O-QoH&P7F(}SOMC)x2M9tpla1YJquo{*=21q*R9MuUY;5tZ8(45)
z#>)(R;%#aH6?q0K4t~|zI<81pQTcdQj#I;}q265vZUEPl>dSd$VYvs!DZKm0p&mB@
z9BOCD+CVbg5|AF5B~ze$QGnv7BSAGLDTmAy5F5y=PeGZ0vK8++TOb3`q5VRWcnswU
zp^3~o-Zli*U`gpxOKqjk7B4AngP5lQZv!Rc_X}Q62zZF_M49fmv2%l9mpO<%qMoZM
zp}Yo3sM3OafHVk?wOIqxfsyJ!tM$dvo+7t2YoYf_QR0byepmp#U{YU+;^W7UQIUxP
zJA}nZNpC$EN_MdXcCM4ZQ&15Cw56<CYm=`^V@(~GO_>I)tX7-6{y4;UD{Mk801bmh
zw)JPtn-$1|1p`H+p*MdH4}bXM4{}DRv-GAJ&)@w63hWLLb6`1r1Lhn6rizD0A--f)
z@9UQ@6R1{?P>M%584`dTltHf>OxvK=@Mtmdg(8t&!Xulv=+}8D<Spyp@*}Ay*Sh;7
zs9d^#;oR#l8Qeb?k6UhR&xt0-Yy86t;F|_Gzoxo+{NwZnh+@v8A=;cmLRxB*tINxG
zeF;?<{ayTB6b}*5C$R3iUCpSS(SrEv9WLdIxNY^=oq>S?{2bpZrUtT8NO16tR+2oS
zjnVT^>BnW6HPs%A0la64Ze@W}d}~{qpzn?s7xQQks4tcP2k6CIQ#w<0V~n<^v|S2)
za2&=d$zrgH$k6wF_|Vytiois^!$|FcAtKTPDxih3)PS-B#9wHbRbFonKoi-*iS>u$
z%hO%T2*AUBV3kE^GjGr**|xus1BzFK4_P=^Sw}zC9lvOg(@N3#04O>VkR!<8U7!cK
z%;a=g0-FH92kjylk<;g5K;sBNLH#4(HmmOq8wg>8GRLMh3fMlwl~1AU*_KzEBs$R;
zC2@1ojKuy_m1tFA+#eWadysC{r=>vr`tRV}u3%JbKvr~8EL-idLMsw<?`-?w%iE)T
zz8S^!NZ;@UKBfca>wF#l229Hegf8<RU!K8cRNLLz&T*SFM0!t6P0dDwolgv^)j>sO
zzc45^R~!S8x~mJ3v_SME*$tK!LYe1W&FGc@c<ni4AJ|R^5E$)lbWOFyJcVUe$IZpY
zCIcB=MMEPPUo<O<<oTdapnpw(@~zPnC1K#($C3y}1ij3=&t?n8nXbvmSONDeegjK5
zi!SnP4Nl2<rl+HywcQg6Q%ARbA#%#fdF{R}r;RHr{4=`cjOcg@vsXZl+i0b2mWFf=
zHMfY<nEzydQC@ecw`XKZUU4*d14^dv_EXXr8+5d<Ej0EUNAfE=6&t5mUA*{)A>rD-
zePKQ{qMyEexf&ZEk8rknoPsI<us(ysenz)al8^hlP{LA?F3G#MaK|kS!*U=H4G%H=
zw}uk>SFqK&AYt(H^IHy<A`GG520L7^5by7QzyNM^(5F9h$d90~b)c>%5aO8MhUdJ2
zRnnZL#yN<BQa$~;+utQ2FrR{~*-sym8Cu~i79cL6@(V)f#8^_2ZBMoZowpRA1;j9&
zyCog6L1Z4mTejtB)17$@xGVUCFfekopRVS8`YY}G=)$MHhwq~a3oZ(gqB-@(=HA}k
ztRYm*05>`TG@uF^nDTGO&dBzZdKZyW(tkqb>F@brQLo?W$r7TXdO)yvY_D{28MpyT
zML7(39lb#D#Xpk00nktN<RotZ4FyX_9d>fwr`BdEvqlElR(Mqs1SFK0Djw|v@~ioT
zUGWD<T1fi@Hvw-Oh~3+TQ01fW4{I>Qy@wVPcDl$?WUx5T?(MP{3~f;`d}bVW<yrEP
z5ba<#S;r_xP6wBdEF2$X`o$e?!A3x=6_k=VWV=9HR<KJy28DNkiyE-_bMDq0W@Tla
zMEL>fY{(C@F`d>0naH`tgc(X}bS+>`z$1_iy0%`%oyX{kK&;T)U2pP55)TYr73P7i
zUdnUTn(?B&G{h-U_r*`Z3JE}RKkpe1I@rWi^s84mAb+fuCtA?nmz9-mu;s=&M(l4q
zd!}YMslyhZ<*b*K6`y>r^8@}6B=(?e{u-L7rW&qV_13CSq7=<g3~Sokr?Vgga{T!5
zzXW1IsFY=)D9OMVFUle16Zot4l`bPb*>G|9Oc=NJaON(=4-3fh5C>B~wO-l<iYSDI
zl3w5ja36cshBt7AboI=zcY%}acdX+ngtWoCA>!!hh-!6^edL1xs@*oZfF&r(%1S*Q
zcUoMO>z!tozVMIHx^uRvdAdNbEnHOzhJOmmkwsy}8wd;M)nB^DjgKBbo(OzmsBR;a
z=~mhJTErJ^fGm^)!Wh(I@}Qz6wXF~(oO(y&Z$XozeZ-h?>PRC>cnQ7<v8z*=y)$SX
zvRHH^X>xLM7MkzQ&)XwfVnrr4A>oGS>MvDkX+N0uWypN8K@1@gmEZI1jF2pb@Hnt1
zKmvUPhXMy=ooKJcXFgjCGLaIVTJmAMCAS7ZsYKKaBA}yucie#vl!Lg8*heL$Mw0FA
zt*xU_tL#AtFx~{a?J)6SvxoFPztvLp6y)#xAbCofDCchHE8WF~MUJ+yk<l34E2bS2
z9tbX=F+6|yvJ-BHOE+-P*}hN?t$>SV8kAyHF#dqJs-LHhBKB<IG%uv#eBKHXGl1iW
zW%foE2`IA>363@d?80eCpjs*_ub4c4@gubsnWcaM-;21(`T07Kt&`DyjFRx20>GyU
z69ZW}5$t3|yeaFkk16P;kXZ?WGU<V1IHZINrmtZ7nYvjl(V51PUS4v&RtI7JX&#EU
z<XCCDg{jQJTCVA5)!+mZj@s_E-J)-g8YnO1z7skkBwWwEF?cO3dk5SWMy*>eP$5-=
z6Gv6`^>O|?k946heb1E?wde!mUUR*|N=f!t44%?4qvbh&mb?Dg`_yHZme>H7Dgplm
zh^v!77|e0BtjjkNvM5NW-vF;_S^0z#aRuqOD>+N-Nu`OxUa6ST9q42O#6SmOIY7)!
zN*yz=KQ<lC3rprNCm-#*CEkmdM!qy1S7h8J4JY+!?6~4b%54r5IPmX{er?|XtpTCS
zbjR=PiwwL4#Cb`0H#a#Im6n!PC~zQCp)2F#;}@DcUISH^A7xD{AUMl3%~DlVSfQg0
zHA6WeD9{E_7ueqP)zb_2Sp7AbyFz46|7S&#FnIAXvJ?KF>@ocRh}edErdQ@uQc+pi
zQQ&0O5GAqsVC}+%3oezE!h7HZN)6zW`&C%6b}2A{nm>SO0A-1r1@JHcC$u4b0c5OO
zb{~6ZCIG_2%=e5diHV6DmCQRsah3!)AK2W|N3v~tg<NL)bfHwzKzzpS3gT4#f=mTw
za6ZA>mk$DJ8c}=sARr~UiT8sC6*mh+8(28M2r@$S;49n8RES3F8Ip+yKtboR`s3j4
z`+%<__VaiKPE>X@6jTHvD0ef@wgZrSj^~=K`@LX^d+)$p1FW@dD`@Nq01|{XdeeW7
zjb%o1eipjyecg}L?<_kz`wcQ`CTB%66B7=0_5_9K^`&tuDAj^c{f2-Iq{TMCaH?@e
zckX-ywt-c4czBpz{GR;`?I8K5*w~CeKdnln8EYm6Aqs$tx2XM)9|$cW;o(WZorg>w
zp{Gv*i-KHG>CT1~kkrXQSI)B2VN*0<trfTw6%`dOY!~%}ni&rkCab=Dhv)wO?f%Cv
zWLtJ{%C^e$m3)E#B@bCM9N5tT_i@H<YawB6Dr~kIWcB8&;IabyA6}tXU0GsuLlV50
zkh~M8GcnwNG4=nhcXz@Od~$|xjZ=LES*A@<-k&!?bV9MTJ<CiPVG!_$*d~Dg8Jm^%
zlT?OOiefONDkX3N0JWp);o<R4vPpc&AGyKc-XmLx@9#?HJ`Emt(-b|yhB$x;btpZ+
zKq4b(H=qwpUo*HP>Xv~~5cS^3W|;(Td8+xtLoqoLa1ET8XP;ZT`@gs@;Pvkh;*w9X
zvR(teyn~2Z90y~w{LA_wE?zs^tgNJ@<>;8nGFi>+s{vu?M3T)#Fc@`T54m-M02}fl
zo>(F#A(<Td{1870LKxVcLMAUjd6Wth6NF4fNM``Vhb(Nh%kl~e8n852IK;*E0Se;#
zf1JpN#g?ne`u#eF<ui{<N3%f&g_^ewv?C22odnQLQ2DA5-Q28V)0eLrv<v_Qtu<H{
zE;0o$NeRX<hI>$n4HUc6OM2xOy3OIi=7coK3#RqWNt($lvV0}WK|px(pb7vg5|8mn
zx4T*0eflD3k_qGOgU|5tDgpay(i7z|#nH6}6haS}!VJRDlzR9{{J&`NR`I5KdN)s=
zi77UX>S+-dA%^R8FDwF=eQ@O|*0XZ%?y+r3W|P@0tStFyyclMA+!IHjp$v8U@OngX
zajPqa(eo1u_)XQ+XLnLz(q}IDk*Y|bqLTnKX92N4>DRYIPjhW^xOf__+}sUKK2x^J
z+55R#CS0&i;wu5BB7<J@6JArgepZ!)L`n8KNmTVp!$Tp)<GuxVvt)1IR&zX0caRXH
zmzb>}z{!>FO3Jdbn()zcFR-_-`7Ub~wHw<e?Z^BItC`F(O*Ivjov-(4a)_{IokK9v
zXL)YxbYxoXFf`+9Ff>+1vBsdMPnF*ZQByO!{Q7q7A?6qN7^B3=ftyO@-o4D?l1i5D
zpEHbc{Y>i4igH2XQOmEkX^I$%?ga~DxP_6^mod#wyupK>9IleOCjil@Q<XFfRVN=m
zZd6xd#q)XxyIrQ9{Q_5CjJ)M~K|*H>Dwagtuer<he2Xb2zyg5jm(IBx#ko6s;?Hpc
zAE+Up;JfZNn2^AT`BgoD;pmxCzGG!h%XAx^Us!BOIn@^eSCo&gNd9slF>9cL)?44V
z)|njjQVKO^-W2#9#F{4@=%}!c(B?Du%MW1Yj_~cUy8g5IUn9EmRQF8KQ<=%+m_-?P
zZDU$kgEYLSiB%-1o(8{VAKe~oBaduswUCu#->s;ytS7{zODbV*JKh8@I~|r}<4emf
z88apD+qhq8_M2oCZ1GX``<c28R6SQr39z-1`QwdyFF&I9`AiX)W!eBY>{ABGN@N2~
zGVA27iT@Q-JUXJ@W9XEH`E5G<J}?d`C?V<*F^7omeu_8NxvgD4f4*R9OsV?EJAwL?
zUSVa?Q7Y?rOM>-*c`$QaTAo4D>c`$y+Vi82-}_PJI92p8-QU-)62nl$kXd#otojWT
zdN8J=K7ln0FpB(W?o4%!jdeGl)+&0*a9iA(8vu=TfWt3Gb||zKpUd%OJM0x*la-{0
zJ?tlrNVhrs{Nu`?*BdbYwJxON$`kEMIT)#XbBx8UC*TWlRV+3Nj*p>yc)l2%{rV{-
zX08W|6%{(bE`#cDfk}uI613piC5?4z8g#4alOH@-tS))p(~=Hrp#81N?jkd%imDhU
zpfxZs8t6%r6LT9)4S&{?`k+AVZZnt7d+X+%2bqL0+%tPrGH&Btk_L(2<G;@@-DF~t
zocjKHA)Ox@Jrel$NgNIqKd*ax=XPdI+_~Lzuw=#vFJp=iD^)kXT#i>4I+Jtp&QZyD
zfvcXLPaz(@qaZ)A5ACq)08USyBQg&U3aXWyy{VQ(81u)q+pQOl&nE86d9W9X1W#o>
zpdjY5lz9kS3L@aX4>bBGbyUUZxl#LxwBlPq`9%)TQ#lB0w<N%&LEN}}`9W|5_v6R1
zLN_O^933sh=+T;kH*@UUM?7}+e-h(x2cBNZ>}916Y}I~R*-w$=Nty4ANiz=dGGO|M
zEnOIAYY~l%Z43JPqSng#l1myv7!{Uq2FmE6AXv5b;%L?OOZ@YVFst?`Vi|7!%TxO(
zbN>CQcW^BW9KB<(UT4)pxPJZkA;)a&*+LQ`>iFeD>dhHB`6aZHU_lLVttMr|o>QbJ
zp!bvi>vBWJ2Gi49ExNusjpNguL|Kh}N-Ji=PW24G+SyraU2Dqd{=}iqo4<eVUm_rh
zlJ}D0$D3E2V|$^~=O=9Y;McECn%up?${L$y^1xUVpU#yQ$Pd7eE@gUi?hS~KxpuO$
zQKh>BCEC9XoKIlpYX83O*K&wqTr`>m6^#{|_ee-a*HRN01rdG%B3l~~zEEaT43r__
z{K%0QpsQVGdNjd%4FJNP8JLyB+HAG%wLk@qbHu~Xhd)B7zjX<Ql4Va51!2}37DNMp
zn@$kNV!_<R0h$%4(SSku+RSGA3J|s9I@_0sSPJm1I;iLPdBT=aGgR%4x%+2)^yYal
z3dtlXxt7M27GE*_-60fE!0O8UylVAJ{S0luBE4~qkz&aD1NpY#tioFy=%QNRr2s-4
zVN*(}+)uDpC$72Ek`9~783k5@KL==7<hpM^wQ^{l+gld!=~o_Qih@R#i<9}e{}IIg
z`t{3k6Z(D-*8y^yu|Wt~JH>~cQD(~I1V%-&z(vwqikB~sYIkVHHtkw;@!l`jZ;FkJ
zvjj=Q68O%7IkvwPwB?aNDNY`qRH|ev<cIE73Q3^uofw@D8x<ebc*fz7<NjUz=oqt>
zUhlIbYOc8&I_!dWyd)&D_wUbg9dmgCXju`QK(C;eXd=83)Ft?(z*}8ieFZoi?fewr
z@>tKFl?#HRK=9uD5PfP6Ke(}s#WX*bTAzVi1xbfvNh9v^_7y>^Z`=Ws{A1(29y^Qp
zE0U6ZYnjg=Ku8}!)@<NiCV{zyqF)V2VKDr7T`Gf0b_#(lpsAp!81g}Zq&X%++AE{i
z<1p@o1XPiJXTO1@4`o6dIH)>GJl&9|7sTRle&Z)vsd}SRVSuKA2>s@|^M>_)Tw?`~
zSj+EpGI}gL;m<~c5DQOGe`*z)B1E#NxCrTAH8;1$XM}dWydDSv#?t?v8EWKC6aB@m
zZqqm767oh|k^8WzeH3}Wu8yDtBlu%$+9CoHRboUNLDL*`i9LRY;h02(ud}j-MCBf$
z*Q1X*UKCD@GSnfuqNq5z$h}W^2Z`p)73i+;W)&Acjv{hJ>xw{#!QIMh7XfYgL-xI{
z;oPrKF!RsnQd&qSI+C@M!S%HG60>|1J8e-rtv{)dKt95VwWnS`feQmw14Ql#5Q#g1
z4CptYYMx+a36}KZ;!at9LVx=<o%_iyJ~8>$0-38u&Sy0qEl^=oS}~N(FKBsM!I=ag
znoTiMR1<lw0q+7k)n`E8Y_3}vHKuQ7{3**!-l%{)Uhmi(hO-sy!FXhmIs&*B6>5x<
zUOK?RgMd}C++&qiP$YmQsQSBMDsJ>Peug=gAxNbMDLWacB-lZ1g)Y;}K#@$wh8a4#
zxnnBEe|;)hBDX=I26BUgmcDU*2K+@xqIjP~=oemnj|_83hG~EbZ>p9n%~^35_$^V#
zpML;1^W?zsA62y&&Xj+EK)T;DJKp{a7!E2&#sj;c^$5%dMH#2D1V%KZJQ$KBs7Fvl
z&4$~;7KYlZU*{0kPE{p_CEN@UlFqZ~)dJ1TQ87)wKn<7~2jgp)irBF^zbBtF>P~HP
zxw?x_q6ZOI5n!;B5?i8(wt-(}KXvL7Vq*X7i~zxMdKQYRDj3m9^nXdDlC725nZsyi
z+Y{fE{{BHRIyNR31ZKT{{Xa*8+mY{i`*VNO_)nTq4h!SM1cd~a*#vq)$Y(dxoh1XE
zTln1L7fkd2x;E+UP(&1pP32wpq*?~&P&QDoB_LLuubKdb?oHdIvzsoaG6Gk!!PhR;
zzW2%#IH}z&r;^RtB1>rJin{k>AGDijf=4m6sE*K&9%jAWcQ5&e0b*t5V{V{7U#Cbj
ztR~Qd^@GeM8>{>SNZ?oCjzG>7VQmsjFmQn?VfU~H3%mQ>%!;kbyKfW$1IRJGt6y)T
z)QFKF^Z5DgH6n4DJSI*d?<X`+f)OD8AKxc%05&{%eZbY21I<7n(}Sjlc*ows9WA7F
zfXod{TtBSMA-Dm-+5EEa;?L7eF(&~dxIp$f04=3i!CLi;;^N}Yz(N$nm6Vk5Z1d~>
zJB5K9e?SHzj~IA8fyI*!Xs2v<8=p0v5+4nI?a*Hx{FETSDfXoshE3C(_=SEtb>)dU
zwC0?z(g{xk9uOQDVL&>@FE2YIiv{S>3emu`|3fB2f=j^h3vp-V$)-|Y^j9C9RqtP*
zlIC&vC)1w(=0k=>Ho*os?@^Z_NTk5Na-l{;FbIS+!aDq~7?Fw{<gmKva7Qtes{LkP
zb+%8^>^(9Z?L`K)J0E{K%T)!Jo;`c`<B4|=6r3C#Av1rhsJM*OP7x6iD43mNm;Sqq
z6VO4~Zua-5varg!n*_FV0D55zxE@pdhe{Ci?X@}fg3Dlo{-z8>?W&iI$)*kAA)7uB
zH=sd|RTi4#VxZm9(z-uD)z+4`H{tN?z;(W-W!#Xs<rJO~E`7o=B)Vs4{1{HmHmUnm
zUQTZ`{x4~$qt=o>yAW=rhN>xv%uh}LdhO;!%b=!CY)C@wizopQomRlcvh;BYYVLoj
zzk7v&M1b*mJg$@b*Lpj5eqVUhpY`@srI6z<$BDY}-2{u>i}<?dJ_}C*0|6njw)ZiL
zG&1hn9&L=!(@}h=y?58hDS0GpVv1b_=pjNtoGYwHy|gZ%NotB3SzD*9&h*AZ3l^e?
zfZNh5zPCWKjbvVMxvD})P0I+s|K#KcnFog@<%6ET4D%l8J5@`skV_|aSk7-p*QHtv
z%emE=vbZ}g^@5E`=xo3$!^9E^8QC=I(aDYmJ0|$P<3X)V25FsM*fs);a)QQH2OtR7
zcJRUF0xR;W!8Exc^ZrKSODa`ImTZqC(E)-4l70JPpVm}VumBAlQfV5X6w^=S4c4;!
z3L7xiAT$ILkfz1Je@=_nr@ri(%?*rvOj?q9Sw^<-<Q2IdC%HMx^ZoZO$uEV5n8+1K
zEw%paNu%gBow4h+&E3j%;b|4}#pmSSRFAW1Na{5;oC-5JYl1f^D3GcX%pfLi2r52x
zntXh%`pZxDyocw^c+<v5hKa`NhwIH=uD^3LDjoM44l7>flJhyzb8ftC4ig|MiY#Xs
zA{3IQrWg3O?QL$J0NsUgw=19M>Kr;K{g0O}i(&J4KgwosTB?DL+2{5xAtgn_!Xgpu
zcHH;_<XFcUE?r&xsUNgF;&yHASI_Xd7G=9<P3J|2gy6&8@jtvDnbk6fSzUCy7?Lwq
zrlF#mk{_eCx@z8S=bKb@cGS60?=EyQ=b3kRM-_FfiF?VHH{NFgb36D8CVjjXf65xT
zkA~1{5fdLP>hBn?tk9p-vRot~ico?S=OgZA5<AU5edylDL4twtzG02A={(L|=ZpjI
zgPA!@sCYt`sCV6!-Xz)C&~zFuND-XC)z#J>1&_ZE12`sHv-rlz?0v%BKVzL|%#Up_
z4teXI$keb5_O#}*VKt#lyBIIq(M5<Wcl?JJpl-rthwT=TGiW}@=he1-54H=ErykP~
z5G1^_d-K`gBK;E|vU^6KTf$GMUh<4=`!qFevMjoLB;5%|$E%-!WH*S3A-9hqGeI_(
zpp)we3FR*go_$zkFK}R+YQ8jo)uC3(!}j5Gu_UK(4et_v!*O4@O|VeGd}gq*SjWd&
zphaj6{?D%f;p6_}x30dmHYxQ<t`>Ni?#SV(UibI=UH4ZbV+i2r1W>@Idh^o2(g|&)
z+ci^}e%GBuV;RLhy*kKo>az?lFMrW3zqU!v())G~R>=1*C%Pos;N>$uPDuehp)1d&
zPK)du;@K>bZ?sTbSvl_!$$Xv=I<=-mdLc4#Cwz$oT|KlUx9oeqLre$j?-XrAe*g4y
zm2ahdV}=W__pba7(N4kFKbjI&-6q{yR6JG%Fg2BTn4z?NZo-A~N?9`V_&2@M@6%qg
z<u9eRwX|x_9SE$3{!^DN;-MnKANyzg?j0$8=0k=J(7>#wfjt@eDy~Uwb~mxeB5NnI
zguF;5q-Q&cZ>IV~c=-K<8#ZJ>AV#<i9b6OO3uw-2l1xTcU?`B5bj-1Nwj6_JGuqnb
zWMAITD=v$bpFW#5Ej>j39b*JHCgdgrKoF}9j{SH*fHw_jo7p=kb|P!${`n1lBs!Ak
z5=i=vz62M(Rd=QdayFuVHE{CW2m&qD+`&Y8MAPWm_VyZ{D|_UJenDFrz8TLe#8oF~
zQ^_~@$nL`+0d3NBA=QRnIYmC!pRCPC$t<aFSG@3KkMBm@EjNDuMPJvV;5TP9HFXVY
z+|zs4Z&_T15ntBtI@C^RyT2Ov+m2XyI%K*;{%q7c@8BYe5%1q%ymbmcO>Rj*hqKEc
zCgm|I)rV!=I?)o716B(prEY*Uj0e0>J++A8YMSRT3I_uj6&Q*^DlA!~YopF#FzF?M
z&h7Ez8Cwc31s2Y*O_h$-k&)&=lrTeN@%alEa=@3GEKlHPdaaD8_?vCm;-WrX{Lj%t
zQ4epZ>>Xx-OWb1)up@uA4DH){vYZ%E4A?Dbeu0N`_xDb?c>CuG7Ycu#aB=pZZ}W?`
z|9jD{9K*1C%mnoFL`w0&?>Iuj<e!`x9WP=!@{s`N5TVp{>&yB9ITV7@&s{oIMxk0!
zK`i3d9Q=?3y+v17%_4C;w^gmn74sA6czAeyJ%Bmbe%7AT$tG?waIEgthpm|PTK)I3
ze^4C;?Tq^9(YL+8Kp3bWkh^R1P0ix*_$4_mxsaUuuLEwuXt$*GBMN3PzaGJ_xT_Ec
z-0(JHgcRtTy1UhM-j3&{pX+k<{j7fpEB(kp%KGuL)qn_I;>?N9aHPj?f+7RQi=VVo
zS>7u?4|J2r?M=m}+<wbNULF=PhkMk<!RRS*lXFe{2G`BEnbp5rzLC-gz`OPp8BBF4
zCnW_fE^e`j%h%Ho58X^6WBoC5LYVz^2&<-bmb?{X>+5?p>9R6EXfj$(c}huHHkxOv
zBvW7qf1{7@`aToD{D*y0$iQv4?e))Vxl3sjJNFNnV-5IRk^ATa1M|~Qz-?;a^;F2M
z(DP`dIbc--*RaEJkoQ;BI*=~r9%eiXCc9EK{GUJkUR?lc3jXh?a0`DD+jTB=N*LaK
z3rw1YCf|Z)iQH#ud9Q9ps&3zRW)>D$vqqyCmEachnPx1UcKRg|M#um}YTKQD2Ta-z
z4(Bk}(~qJmziejk;FOthmz8BE(W!%AHaI!d(L}(jOm^$_l>E8HCDuxNqX)N3#|69Z
zA50)`miAlP7{|imiPCSpF?jGfhamx+Fy!{3UR2*_@GDi@4)Ezd$u+#6hxDl0A7Yfm
z7bVFs;umL&h#y*ss8ACM931-1fF0xx|8@8PepFfled34{CY^4cDSs|MlxXD^>*bT>
znroXW++EoL`4_eC*-kaxe@GU&u`$gJ<sPmxb)-E;CYQJH58x#SI_`22XSSc5f)(;W
z+R2xSL42Y;%*Zq;sn#XuN22;V1a<}4(^uizsnjoG5qW84jgC*pEj2Y`H#V$n*NHK5
zahU$`nfAnRyGhkv*WXVN4+AHTJbU&GDlWr-%`IpKR|lYo+E|hMyN-rX@x@n?7LB(R
z1VhC+wv6%W@rzb{RD&A2x`{~5gYXCbCrvOvfK@v<BI4Hoo3OApc*r{d5_N!e8$|_L
ztpgF!Q`cgX=3W?;Bm{raFne1~F5tJ+372;Hvbf#Kp~Hvcm?ZK$o~HE_x+FuW>WGn{
zMjc{E_W}E$#Hu^Ef&`)I4ISYS<UIVCPEVDynMFtcpdx68L&XNpuMK2nsbGM7HE}R$
zA;CX~2rn~hX=r=`e_8=^byZao9KvB7jUp?Y$bmY6c}tenKu>DyH6c`daxJzw7_Jc~
ztMB|8K+P@a=Sc1A(+5vgLVCIiGG;>+4n~g2jOfbYSMV<jDyspr(`rl`K_0xQykb7(
z1TEBR?&}4;H*<o1C;REbWRqV82NdeQ*>r&3Z7@5CuOhcH7|7c}!9mVr=)ctfKOBg#
z603D!uOFj+Bl{*O2ztp9Ow$GjVFdP3Q!VY|FnlMf?4NoXwZpH_hbs(YekW1m0}OMf
zzepRhn;x9^0=GeYwE#4}zB7K%58nzhB$`Z$B$7tv7Ze{k_z5YoOY8yy8i-i{hCmZ;
zBs#4CH65J^6yI}+fSnH*v<_r^1JM<^;*j|s+T2ok8O;S~>8`4yy{e~Vwv=|_<yn4E
zje+55H135yk-ouRYat$;bZPLf&%0V+Lv0YA!>NSzNHAi8F?g3~pE(V7koZ453yfae
z5$(Fx{u~=JtehIyOo7N>)?ERbO-E_RdO(@+gTVcd_kk14FYarA76C@^5<&}{Y4o2W
zKCwu=y4;OdmoMuZMJgqOWL2K$aOl**FFNigFl_1AebNWVxa^3qTTHN(@1zt{Vt($#
z<|qHiSQNsrF%%N2YJpaLD+8PCuNS%vV-Fbi-a`f91vIz4WHY25p=41_HK;%I>}&i<
z7&hidooG0eSge2EpPBC7;$xhi!A+mixen5~({KS={+-yl$}srDXwQ&={#9~Zce=(!
zEX&LNKARDqa^NfM2-8Y)%D<lh_Pp8pSSrD-P(HqS4Q(3*Z3<qv>&@0be(1LeS;Eu^
zCPodc02m;Nf~N>N=;FLJyIzHi(vv6P74H=4rK8&<1hANyGic7Ag<#&`@4R1NsprLc
z>ci@!CGkZ0m!niz!;{zw0%D@LWD@4<a0|FX_mdr*c#Eg?f*Hdw$**DBr&DCr4Sr9|
zP#h|l&sacdx7_UaqTHh#i|3P;1BRooG-;aPumueJeZaTGx455obhLoitG@1nluJYJ
zT(416R@z<)jy{w^_tSmTn{3+MmJ{{}Hgw^EOOBLX+1xFOCPh%&FwE--U)t605yji3
z4_Cy6k+zP_AMc?1nPW+#zZST*>b+E;yt_UuiN+9*#?VmHyVfu+xg7w3<zE1R!NdQB
z7zdxiS$|(r-ozvpHHJXv4HxLVAcBKKPX)rEXEbaw)G`MaJuTQ(%ix*Q)8U6NDt>lo
zA`8yi(S#6+oS&#|20Wt3=*qMy4vnqha9kB|FehaT=PM{u*=gWCv?rybk#7qe7*0|z
zA$o#es9F;B9KmKqtu+Q7ld*bvwvl;tpn&*Ydt1>$4h@W25G8>4L$WjS$%C5#yamhO
zn}+S-+=4&d-4PKC6iF-0K+HtMc-X%L{|ytt2KWDN1Tr;^V?kK~nkXz+H@0<^l&HoN
zm;xkZ02zg_(j3hXwP<J@8;Hg*0b2EojY0oS42t=;7&JBREcfo6nosH7WB9ap@9;A)
z^Wucdy3o@KVdiuIqnH*D%$;EU=!jQlmRS6D26Yp%SO9|#E;1zbBeIe$+Di>A6tvn@
z?c7}QoIERiT6$Bk9k6=i{W6rus{jhSyY>4EW-m8A!J$GzL2C}jehlzZo=e4*_kk#_
zB77ZpAETo)vlYW|79#%!>&`uG9E3oCg+=l_e(Y{bLE!|ENq}#hD*k+zDd(<v>4D?5
zQsHU!Iqx2sOF^6<jOgbTVNmlbzMEB0D$39Y)~DiAR#!shxWxIca-_0rc9Qa5T`;SM
z(Bs#ybc+<@I?9l`{@_hdZl%;Yh{ss7!jGOVug|hJy~E6VaJGTvPl#mr$KlTN3f;!}
z2fs$62!n+`Lv_?Xs;_y0VTwZN|Hcoltoqa^FQJczMyKnmKCXmLez%GL2Eg{De+R_A
z%;RRa_mjc2Q%5LqTu|rS9`0-as<QfU9?{ppy{-QHl2&KnaDkphqLr8mKttGi4<G1t
z3Mbl#Jvf54_Yj#wPKupcuPZSd+L-0knOU*tnp~9k!nFoqEezW#w+#c~P0}#4Ti=M0
ziudFrHV_Xn%$^h|1OGylgdoKJM_k_iICh@{Eo2A`Ykv~EPn8IQ5a@!Ht&bA+j;q57
zaV!2o|EE~^&nI=N#?zXNEiSkFy&#A1%Q{sk$#=G@uuxpXi2ZHIXn5`+y7B!@`qIbR
z_pU5-{{8gQFINsez`2e<?~_TO@uw4ajej=Ui{GsIlGHoT(Is<#UU8q-M#1S8AY;JO
zzwgzUI(fTXe0%4>V6_3Xl=+$bawWX5Mg1Q#f?wA9)_%V9!vMZ8tWl$DKVQA@{k?CV
z|0&E>T)r3V{4QivP)vb128RtE;VQ;4p)dL8lYS?2pvfOLFb6-sDo7||<Lhv~2kI9E
z1xFnmF6d|%iX>w0^8I`XWRRHrPW}&}SjHeUlV!uUdI7Kh0<R~6u^xPsYuld$3K2S@
z1gQidEg&B%a0pl#+z|M@e+5kw5!;5IGlD>(n!^z)2+9b{6Jya|A!SGhAcYOOv=br5
zdH|&=0Hha?*`DLS#_ccu1PNO9DK)bebV7!4t$+r{8W9gsX{ZjC_`ydA3tbPUpr8Ob
zaU??-te~5ly>?{=)RMn%1yrL(!!hdQY<QDJdxEN>k-Z-{^M}Pd9y$>lRl#?YUq;~>
z`p2M{)-*K4qw@mKoT)U92A(g!%m|_ew4BV$oF^z}`(K1S%2M0es80f{JjfdbT(VXe
zB_@e#$V-;s<U+(=buRep^UI>e-0)+Kk<&AoSot{rp40JvQftD3_!Bk!7l7s8w*vNe
ziTc;u%4I|%)h<6cdicl2l(U|}A&i)V{(R$QKT@w(Z5$@1a_j3sxARL)P2~oC?OaT-
zgLf^hzCD`2fiKQWMJrT8Z>k_QTZ5A`GmZ%(*UJ(7nZj)~bzd2hv19@cH#RdmI*>|)
zJutMPEXbW9-wy>U+1IZf#U+|tbDad(rL^29`mota2nB{7!d6c|*JJ~>8wG#C6->H0
zDfN_@*-dq7FDSX^FF`|={f&i%poTH%%DSPds-;beU5d+nVx4!>;5heDOBVvn(ljr+
zH`>XOn}Y;<a8r@;_@#jV0g(Hz=S8|4s&>%mSiY=uOlB9JCO$ojtSp2N_GThB7GM2m
zICHq(b_WGP;zubd%#}_#s5r-QplvQK&CY%<kI{V3jzCWC;4qd;6tHJiAUGv<b3esW
zA1wl&0<0)Hkdyz*HQ5uK3TJ|ChhYSjwruBO*8hu8_Mh(`y!=m`D(wwX_td5R`y9Mo
zL{stPv_BAp!CV|fW9t<Xv)flLN)4{){D^D2^c+o!U*TPo?PF2NA)$r2XV~`=z?7QG
zEn6`{0Iy)N<V9mYGxX;lg8t4{GAe*LA&`Fkw<vtz4$4g!PSREE{&8D>^qT)>u!u$}
zB!y#()ZusuD9XU!SL_sJ1iik{vxC|<;3TQZ$8?72I63$no%_AtBV#i&)!^K$c60_4
z2*@D5fP<S|M5Noe3`i`nA77)@+y^gCK1b*s3dI*EL^%nMWz-yuj(&uWhn}fS`1{Zq
zI1mjQ+SK4Mu)aWJV`DIEJ2(2Im8^!tNn--g=`{oA8op;Ggo(YZFl_z^<YaT`1KE?w
zJZD83pcxU40T>Y36k(u(e`QevCiHRqoKf4Z1sxF1@e<WFc%nSQpbIwUrm4eTdaA02
zO7H}~^Yo<}U*}A?xMUrFa$CpX5g+hyFTgyR3cXG@gHV@}Y%6q+>yPY&goL0*K4=k`
z7N5mkdWg;l6SB!5c;ciX)p*E@aW_Edz<8oh#F^jc<Lo0?!R`RAder(>TUVC@?YA&9
z_kKA+&+6J5IF?jFF)1-P3)jwFar9-@qbE<`ypn6EcZ-vY%LPsoNm7oM^!WK&1|2#F
zZ=HZ|>x2%lqhLsd1E+Wl3freLA!81)5>L?kgWwJp`late&~>u52IaVm@{D;xeSQ7e
zM4nd;FQ=+S9;A(wwLF1<8i)ewpWQsfd+8Ugi1#&rwL!~;={r8d7fx|PxtrhV9mpsa
z2i<w-EHlHDX3-U%5TB`b4#b@T{sl4chRT3ke5sQL7<>)iG?<Rax#&DG)E4tF*q(5K
zbUI;r-fln~bhqm;6>$EHLS(=&Iz#cg|FPk(Qz_~t8XHiH#zH?RoK0YY93j<`FX;FX
z<VFW3_5*q9G4bIkcTy)krHD)ra<EIRw5uZ(l0=h`ceyaUWqL{`?V;2eY2i1Z$Xmhb
zgh8J`g*6B5PH+}!cBs<<wG0Z^WhnqclUnLi3>T1v##NPbiXC<nj*DgM4R*`TSE$hm
z3A7sf@|&kEucsw5mr(8aWBm%~_*M||Zv>%!aPWvmWhrZFhCm0bWJ9#nVDG4I+$s6k
z=mKTXLQrG-j~{o^IxS+KM8tp=7LAf8Y6Y_kBfE#|>hL$zmhxcSI-ZvyMvcyPD23BH
zoN~>~%sj<&yRG41i}7D;K(EW9-efp6B&*|(a&_QxP#<5PFN5Kgw1+e>t0v$toreK*
zZuDKFGx!AQF9F&*kZ+l{c)r4E6|G+(o1RZn+mp)RYEXrSWAKGcM_iehgA|-?Hz>I_
z0um&Mt?l{Bpp_x!BSt2WQ&hC4H&!pM)CtWpYRaM)v<zmMgl<PwXs<<y5*kX}i}Z;J
zJR;qf3pZM(hq?|3ynq8B2H!$@ynXBmKWvtf`(#WBPP?q~p~oODEMK<>h1GfvOt7|I
z&^}cKrILSe@X-<IGF|&_1`EEb$!GOK<+%gKFzl)T;Q{k&iJ=zXL`2)*7?}WIA-=qz
zA(Z;~_uM(bQibO9Y@clp1IHRx4``u<Q&zcUTcJ@Y3E)KkkEs-=F}2CRs)~O3*fia&
z>p{-FHhGDl62q#6){{%pfodI}PK}e#)BL@Jj9D+kW-KS(!=e8(mRUdr=#=2l(6Jpj
zK{X!AupGNVLpZQv3Oez8)`|vYgQ{0iM<F!6<mG!{=Epo3Riv~|6Vsz}fKhKxbuM&E
z{)4Mmin#az!YgX;59;exhBjGnup?(MqOE9RiKHL4O#5v><`H-eC!3%X+EBY!^(xq(
zz*VFO=!I636jOX^NOll+pHvLg{7vm}&~3!W*I_|=!@;prZSnAWIBt>)*g9ah=Cu;x
z<hW%~oDUET^a*iBR&skMg@QvGPW3Q{yN%uf)h8Gi&Vr?h6W&&~HTwUw_vPVKuWj2)
ziY6o_v!XO8Lx{|kA!$HFWeP<knPqM(sSKsiAY+3NDN5#crY1v%Wfm)uF;isx&Rcsw
zd+%rZ-uL~!<NfP-j(r@DmSqjU-+f>Ab)DCFo|h=m94tv>pg49pD^D-1C^(xKHLKtF
zosfh;b|)16Fc{tLxIj5G)8$+c0)$QS8H!xQyTd=x{{@uWOu5Xj&QqA&Ufp$Vd4%jw
z8JhS*5%T~@7+*9!Cb9JOldOg0PMFM<N`I{y{x7H3A2rZ_{_L-?$UoFy|GAR?xg`IH
zrDEUxrAq(bUnM|k&}S#E8QohclFx{qmB=M(j!W)=9~@3H0N5v;H!AzpEqyJPEP{m4
zzz&K*5)h5i_CMiu7sAe5U~gIQ<+(HJ?EB@{g_6I#g&;L|NH#EzVrAXZZ9?f*_w}bj
z@@ob0E)l*!;vseX?FyuL8lglJ2|6W&mwO;SZ9quOEqMmN40z8GWF0=m+AQ+ld=nEB
z*J|M8t$|X{3ArZ@&?8Wr*W!pH-lf8jaI8X`;__Y=;D$eE1Q3BFgsV@>sHJb-Xyfc9
zR(a(07A;x?Bdz{UyO$m8)361lLY-foVkAc-CUcTS7|p4wsgabB;A0-M@PO)uG5J9d
zl}OqFd$=Z}*TlL4&H$SrH391d-qQB7O>kQw9~uS-3E-;qbKz3OUlUX(hVk8XY4Pxl
zXt(#qgR~Slf!xZ19%u#ZUeUkqf^^woU=pHvxS?nu`v!Zf`9A)%8x=sp*Z@U&c=ztz
z6>JCOV{xDpr#ybxYCoPnhU%Sf))+=;s<6~L3g4FsbRFQVG~nG2#!!Wxz7ep{4Mt9t
zs_t+EBIt+0%s7qcjd63KyE<ZN3v)Vs<Q$On-+;wOXS7etbQA2>h~WcCJt2pI?Avz+
zT3?M&;@tboFZ0OR3FQ-v7<^K3vO;AwHoLakwHFH99^_Py1Ox=CfiMyW-cIKzx~ioG
z?_jAGgn|Y3t)&}a!47wi<5Os7Z$tmvR7G8#8$1mO_^t$kJlM#g=;@YQa@``3OX@9o
zGQ_e$nj$X?-H<0Sy%q~3`3^iG#E8FQd|ceG@UoHe{_)O)3rrv}DDZ;NU*UvAUK-Ra
zy-u4`O_`NMn`w`A!*%@laWz0Dtnh5_ZOC{hw)FK)5NGXQEm*~YOB$<<_0A*b@;(#h
z<Uzf?r)U6hiv5|m1EC2Xn<2C}aX8n;!>vvR`l~cpH^(O@zlUS9B(7f|ypyUxzmX);
zGr>)p-j2hvF9-r)L3j{(1%eC39qlvok|b{^$7pFKd=e->g0Q@}K;0RVQ9|C{FN<#!
z2q9iw0#Z3*$z=Za#zy~R9K|TCELBugI7pcW&#Ysc4KphiGBG(vYf@TP@$=i=d+>k@
zbTH@djJQNtI$5*1DWFKreG}yTTriU?hDW)zp-cU7@_Miq3<RB<d$_Y)8f2ciuRY!s
zAA+482!Grc{>`WDv}}>ZNP8o@<|h7yij%z(@wm9;iZqohVq%garo!<#IZbX*A_pNB
zke{dXVBt8RNT|3>8pI~XFBo9F4GmmBN#wt_?;$_p*lJPwBg(p9#XQLz5k<uoHyB$r
zhw!_7qN7<VP2RFcNybI?y~iZn4TAt0nWKVp<~tBNh=~s``IUy>OB6$Gh)XG~?AlZN
zVcHOd!rV=*e^!#)2UEC<*REdG>a@{i?7|Hgdk$Vt<xN1=$2d|<2@`P4k>B&%Ir~c0
zRh)=J{kFgWrMxE_Te<KrHQbFrY{KYaS%8ERr!gEN#FmsbX-;2t;xVnl3Z@H#aBptL
z${Ve-L_fEf*R|;wrU9fD4?CXl%_BBg<+-Rh)KJ{xLk=Qe<BTwNe>C|rF>#aNwr$}Q
zCW8(5-Uc~tX~#I6#~5FKTUtRibPO7Vg;?fT5*AUzb6Xo2H*ykZjiB`ObS{(BSh}p3
zn1Y3i7KxK74=vk}i-XM1*-wk)v=+unBDa%#tx2yM_84x5UKs9VMo+e15e-C*Qhiin
zt_!Q_)ZbJ;?FEZIbXJKoH{l4sfH}4>Vk+*|C|@e5H2niK8cpGWcjS*IQqh0>sKdk~
zqXnKhkz5kL1GrD7v`}HG;RgjQHHKQB?~@O5^8@(jhfH6AD<draaV@{%-4#2fu!Tpv
zlTx#6U^rO~@1F!xNrALNA`v`74?BPTsEU|(7a8|D4H8#GUyDh)eKQ^XTBQz)NVCTS
z@edq*PJF0=vkkFkRuHA!)*sD>GtldL^4cb|qbB{+*n<`v{>17I94fBK5Ka=?AVx<#
zYHCdMXsUPC)PcD<UVvM2OX?c5g+lNV#npL?Rek--yWlHP!jnrVSSVi5*BPDZg-1|R
z>#7wi5<u<hFM)^Xv=>@KHL$Pug4HXbU<dPTk3&V5wadbYSODR#CA*x7wH1Q?%T6=>
z0e%QNaJv#w@$xx_kg);#AnJ9cx<Ggb7eC@Z9=ewFX5fWJh4Qf)ydvVI7HY^Z1P`Me
zGNu4SDE5=!<I&={i(D$0ee{?3A9IsuFb(Q%(J}FnIS4Fg1j9Mb_IiMmM~p79=aNQi
z!!5t3Wx?n~uhnNZ7wYu>mP5_7#JlwoNr{PQc6IgNa?ahqE11auy@+UW+XhKu4GSB6
zDo)lBm?a(A<-b%L&U(2@lb^32=!WKi)Y(KS8FHnoYz637a{ZaRfJ&zl*FX&&blY0I
zC;J#`#tCZ~ZFAs8*NL5Fvl}f)s$bfobQwPUS8SeKocTAdUysM?vF@&ZM6ABy29!t&
zaHIkGh>RJ>-?0<F1Y$`C*<TU?iEzL08+a!KgS_ti?b8lo#!0Nufus?0V7@kQ_}nxf
zdjSu06#0DdtxWdK&eW5*Yv>+^k@}msvbTB;2R$;USSdP>UVPSkw)5_7RMmuuu0bFt
zvQ+nk7KiYsVwqaO1&`S15Nln;WrI*0rmHpr*P{8%*nEVU3AxzBzzN6Kh|lbp4;Zj%
zBxZ~$;!vvs-^+Po<m48^Of5iiwc!1Cd*5x?>G}ILw})R*ccnUijn7nx4*?h8b(IPO
zxgmz>UdUZju_8#>NjAsn?7rtm#=Jo6A}2NE&bq4%&zvI$9mE%KG*694g~19T+96nO
z5i?z!vwE~V(uu$>9VOFp5Wcf|QnE)P@~4V&XQzG)*QA;#5px=3U@QtFSCMH||K@Q@
zEY}Gb36jUxH|wKy5n}P%bkkS!$UEQ+a!2Yl+Nm^qICKQRw}KlOkuxEgAnOAykoX;)
zDMxVS$@T1h#mKLVJ_O<20^1->jR(Q>F=wl5_{)gRyaHgM+p^AyBAK7&5$g%N+|ls0
zWL^&07WnX_7$zV@14xT&UotKLVM5BZz*n}17(u|;v>JXA`yyqu^Cz1J<okOF0Af>g
zc_ksd8GHXY4ne&REU#{?Re&e1dv96^@eP57{f^=`Qz>?xbFD}^(eTg%YV$0YWC*x}
zUt(^5)B+|1Ob*ivz}Rc(cA!+7W2y)l)-zfaJyYGSlk<G_64RByF*0-UAc+V>tq&28
zq5rW=pN_=^t|-#G!DT)KX3S%e1?T{DaV0|q4yzVWw24h#*;zvDMkLpUNM9|8Yo%^{
z=%G678{}v<Q&l&RxDIle&>%0d{YV`=P8?_8a+HjqVhxQsS=QvNa6W7$G?G>D4c&9%
zkxcB5Vbe1tq~Kzk>k9wJeG{_2`NtNQ9LD~EQ_}&W(uPMgBw|!d@0ep3bwbNg4Ffzv
zuO^))VxB}U-jbv_5Q0+S7N{J396OD0%R~4S$jBoioPmGa_c}|gKBDwqtATK+*m;)n
z&3QzYVnrIzh{*D%XZX0eJ<4|S5svsvhx=>4Vs%q-x?^xcZ;dS7C(jC)Dh}>^EK<M^
zw9ka-Nr=rEy75G(fP7EJQt9IBHTX|s0*4CB9y@_I@5iG2I=PBbxGYSHm@nYeNqYBf
zvAfZ5d+V9kXQGl>_vWUE1MdbHHE5vl=*g3Bku@QR&)s}sXv0^c<avOpQ;xx5@N<Qa
zxCRh#9lw4lo?hTF#Bv(1wj@&6W+_P6-OF#!&n<JG;y%q$c3o=TeX7)|nTY<)|Dsif
z0QAz%Pzf*{@e)6DV=*O2EtTH4SWr27-5(60nrv4)Vp#n<LkRfVZingCGsqj@`tb-L
zC~nug_C&e@9(P`fB;Sj);pN_s>&2!LQ`%(oYdIvXu99Iw_>_l7vixa3zw?gNByPtT
zIyOXDLTGsUbZe3(sk%mC!q(Wmb>1N)I7EU5v%n)pi$6J!WPRF^YQr;6(sU<yR5&ef
zT)n#b&W95xx0HnN>#L}ozb`d^(`zwZjJN=p8Lp<L6-Qmb=x#(4XXFx#O-|kmzCgH7
zOw3tS0+88H7Itx)E^s$G6GN3kD@E)Q%Z*P3trEY$Eaew^pYkA>rA!6F{0%sFr?)%|
z17NJtIAV{vO9YrB6EqNuz$Md1M;6iTtx)oJb&$b`f!*AIm@G(k$|r9U6D8OiHy?>M
z!4^PVssA)hH_q(d9Ma{ScM<lG(XipjAVa;NdhE>kvfw)$)^M~%tR+@N<kf`s-LBR>
z?p_3HRkx=%P;BxbWE|Y-N|e?esgta^Wn;j_BnAIy@s|xNodEm{4(;fBiSo!{Lu2Ep
zl+5<Uq=O6FI(XAw{X9`IUL@)ri|R0ujLn*Coi*+mR1f0n$a5y<!Q|q;kqRBU9zCT(
zQa+KFpO-6)fAZ^(#yRBb1S6l57~MriZs_Ug2xt(l3c~n(n7?Q0rzgF`xQ=T050My&
z0UJ(sK0qAdTXzp-sUfy`6+(LRK+;4m(rit1by09vV%Axz&lHWHl(T%uu}NzQ73Wah
zEQT9Z!p9etU?MafJ=puBwTv+*Mr7TMo`B_GCG?Z88=l423A(OkXtTF>NfbyoXhA(B
z2D{(uyus)ubBc)m&=_^-bo;9OpT-M}%sdq&)kB&G;cC3Lpcu(3K<fsF@F*F@M!Xae
zxspowJob`w5>fsk48(U5cUI!yNHPo5Y*C~;cKq!P*^pec7)X9X{L`1dt^m(Z64e;d
z_mHB9l&Fv5;&#GjuL6v91@Grb<A7~ovp1{*QH6MoqFM*cAQikVEF!`Ng%!y^K@KG2
zkTPr=Ye#-ilvCrW3)X|caw0Trl*YJ+v^KvFJEK6fLY6{s35Dk{yH<r+=k>KGk@}Mf
zPtE1cXI?WQh0z9y23?H01S&5y5(F(k=8YEJVMv?cOB|B!*ts9(k9&b$gN77Ae35Z^
z$OyZv&!L*|rEr4jGxcmQ@(C@ZlDc4kLk^J=J@t7#Nu<aKI2<L$kmhM)TY?~PrmXxB
zl>qK1JKlyNB}y93Aa;=?Nkm)<HSCm-yfHXGM!XA~+*HMv;r1JreC<=;T%Hj#5@Pa<
z+{vn92C0dQj|p|&J&cjT86SGj@wBbe7jfdFiewdKGBG0}P&#ofBORo{_M#IlZ_h4a
zVyeO2Bw#@<7A$e@%bA!^7K^8zWyvjb1wwveIe0+XhmkE1<1lr4*q<eYi<*dN0D~cx
z9baieK!%Vhh3a-ry|N}n68V-OZte~xP(9)Ru?Iz4uL>pDIt|eF58o}}#9u<~dLMNO
z$)U+zCblq8#}L~Uk|9FkgS0cq(fj$mdn#lI4^W<wi%L^F;3Am6Rs*nxRM{9jsy`yZ
zBH7Dshb`4-a68>wg?$|>d%B!o!3I(Cdhe;Py{#!1F1&m$9fq)T0L?N7r5PIkYywb-
z4NyK4?;VovRF&jFy*-x=q0k?baj_+-6X+$@xN&3?k{J=?UZW0xu`>aGh+DOhTiXRb
zr57kH;fa$qYlwdmlQdHMFmY~GLCTvA*ZtJrdcAl*V#!5zA5>Q!JY>U?n1wUT2`L~M
zXc&4AnR_=$*NBLi|8yuhq#qlk?IgQR=iotNHk)>pQn8!tajmVNR^m#hXU)%lf~gou
znh8~?A@}voz7_Uo<+tP5cBiLDI91YU3z9gSEWaO_KvlQ_9$?XkF);I3!7}-SaO%;j
z)-i0j+@Xe<%f{3*c}<Xs)W%SMu$%~`|4@2bX*Q{14LsF}IuLiJ3Wh!8n<r_`JHy~^
zR7IW$E<p4cHGih2dpRYhzOrYE5yuzc(SIdB_$CN1pWiq;hz^A0;*Ch$axWG8`XP$8
zJ)z|IhKuU3;uj(&A?=h5#b}fpBn~pQ@GG0fQD0Hv+EEY!_{j!`W9U5uD+8QlS-6i|
zQWDz$AePBS!yQ-~X2lp%(GW9DH($t|*OOPULfO&b51z!Sp@>Y}Gx_jXCKi^6&(;;-
z_K`_Vq>`)ebK&$~qk%AYhd8BSu@S*KX&`!dKtPBaLe7!u$p#pIAH<MB652^B+}}r9
zvoNxR#@uwyS{tixod&@?5#ZNp5Izk~Dq=l{`@Kn2^Z-fsXTDujBF@e@--qga@Ye*F
zCBYqshg<UV1qEj0u3sgQUF5Hk-T0{eCtRp1GG-*jD4Q5b5ertRi{prA0Jh*EwDpM|
z3JBU)q@UG3Go3zjU38eS;ZQk;@C-14e0K6>i9I5Ym?Ol27Y0G2h<)ABvr`X|TsRf2
z)Q8v6Xk>IW+?<F~5D}yYEl2__kTw;;+ICkFjA^p5&xLsSkx+Q}@GV59+utqdtdI)j
ziYx?)@XH@D+C|hWfs9!~N;~I%iPDqU01?2Abl5d?KtCnM!#D-H@LpW_)N|pR%o4Fy
z_zu=%96YFSE%%2S?wiH&b9HSH&vw+a+M)M|C=3iz#m-KUH89GP_<wc53_M}ji$}@Z
z5TFS$oSL)!Ke3<Q_lrvEb(Pd5+u;5Dl4wy{b|b3xvt9=D6SCwK&hx&RJWM+P43)n0
zwH6uphe9`mhLMc-G<p)t(FC6B)B$qPb8&f&w5Y9y%2=TS+62-rA<zgutP?3~`mVMI
zkTANEYc&YN0UzS-<a)SH0zq~3NdV8R3ka|lY`3A89ECJ~PiW-m`(r$q8Pno3>jkKI
zH)?4jf+Fh@@rAfH69-OGA|uH3MMghPO#FzwB<Vh4rGV*!vhE`ZfQ$%ajW7`X%m*hB
zpp=X`^DSIBUH9(+t0*=Vh@Cq~?2PC=wQ=)%$D;-}Z9LerXu)zLXwV8nu!#aj8wXo=
zQ$EDLBt1$YBLz?wkf}DU`%k_JT}Oz$5UVw<YJ(TqF+aRbvk!_qNkVXdathDof)6gS
zohRPHh#VOVEPTlb{~MtM19Uega|rjFRN<f-A3#8+ih2(&0jh5eXHeoT+PtbZ@+*e9
z*21*%F)(#f!~+!A#}=tT=?pSfPhFzMkP`=?+@Xk1tB5+xd66@rOuXcG>sa|ZAVhj8
z=E;nP>EBKcM@21FpLV+ewG2V0FJE44J~cjAjKY6QN`k%h#3Ua+>VmLw7I&l9KOFMo
zsZvXGMcj}&TkUtrri#Q3t|Pk|w=D^YI&p6uHI|_-C?{vo4!BXoywPhdUqPA2`SXtu
z6$u;erbVRlt3~tJe99NPOmQz{o>GuzMyef)t-nb`<U!kfnYn!Zy#T|)BS`O+Q*Gq2
zaD)mW)qv!J_zBnC2UH#SfEtMNiDq2Cbm6u?l>vxG#9Jci-PN^3r2#$Jp$NLw6m-V3
z{udZ|_>TXZ-Li!`C~5yf!xG)Fp#tzl4cbvi?7Il6vjtXeo`it^i`H*McH%WB=h_dc
z$vcwPg?JIyOj655dwsu!6bEFKtaowNbOkR!tX;j3m<#+f9y^@miAQ><6NZQYSE;$I
zLs!sw{s>I|3MNrF6EJ3s1_*+Tbl{tO52t1VMbBW6Za3!95KabRHjS^v85*O8`&LC}
zJ(5&~-~^ahGKv9C!fOkMBO)-!5dVb|w*uT$1FgrisDRrHr6`??@~~l{TtT6Ylh9E~
zWb<a-y?X=a_(8Qmb=PV{7`fa6z(r8qxs+uR2RfPA7z$6;dkCCTc7t>R>(^g&tf54%
z9~Zz{Q(i_K5J9Ua5q@aMfe0Frd-F{e78RX@?`G&dV(d>+H5@F-aQX9GJa=_bU6HA%
zXip@<U6~|}L@z}KT_DU|TRUAoI~78@hQ#{?OZNyEa{z^l!svZR!GQP&HYrT(YN3|O
zHDafX%m0_?CRly{Pi<MTcVBNTUk<e{RsUgC_vKHjv4=1koAOEjCntQ1vqD+9t3vqN
z?B=YjPf~L{!JF5@M$O$)2+=+`W{$Wm7EieJ^HGz7sL1(~i8#m4eqUM^{_C&uFGxx%
zxZ&qRPa4-<+;ur5;?y>({zbnf3;h-n?((eNz>Ht-=6BiWOt^PmU2tJp0e^P=`OIb4
zH;CL`!TCUvL&`<YqJUwyOx2D@r)zGUy+0FoUHtxD>iJj!D_t3@jy|V5Ma<6^vFogH
zQn#+G<{wcNd@Z|g+Z*;+v&)6P=RV~5@VLAdy7Rdy+j4As=l1CGz-Qab6_-%Ys`XEQ
zQ5^GXraW4_;M~jLTcFPS&IyUul8?K3PB%Iy^4U!kE6Vfe+*d&Q5ex$fH5(fnU}ODH
z1_qj5N{9fEl^MtxZU?R%gcU97w~3BY`U}M?00zSb8P<ADKC{zl;P@a}*@aZpwot)m
zRNn`EA$!#+?2q97{{HbUb*0<d85GLiZFCdM(AWs7t*u2PXg|hLZGn$EE4TD%9{U5k
zcI}!%T@V})fckV>%Qt^%3++u53g?;SZEGebC(Y5WeVuye&K(*~tC;|g)0u7lZsi^=
zY$j9ft7stE?UP3Vf1NN4Iay;$OtHKNVScNmI5QYh0^!$kpAmJR6-qze958!(;~LvU
zk(v1AA~O{g^jX)#26Ah{^XI+I*;L0(*DV)N>X*f^R9087z!Ayi4YE(am9;gOeK|Gn
zdq!Q(Sw|HW!!xfIkUmkNsLzjgGH8XVH&H7qS$TPRSy@2~YeG|Lu#i2P9#b2l4&IU%
zrj+_iRa`Vkx6&T@{=FE5gE>lr%I0QXL<?^F1Dcwe5UZ~f5b*x~EV>A@!IxP^+N-g5
z%%h0CncuZ*?_F6bH1zlZO$Wir8p)d>D=PE=AXSBMBoGj2cZ{AiYyx&MSo<PR3U~`A
zS{3?|l9Cq0WAXZ{<-ywNFJ7p7cuWS^z}SiQExNNjnxBlzTc5p|LP_-H1R3}Dr2taV
zaw5{WN_N4&Up4>yspp<LJFC8r;uUY?H3Zr9Cl}!Op076g`n4<R3O?Vt?M$a^+tq!!
zt1F=HMe0aaYU+rrK%7=-iU{TAsS`2ho+cZ6iOI9nzesiBrF268dE<CzS$YrUs{TAl
zt?=;hzx`NN7EoM#jmpKXNu`U4g8KCE;tK(0;%2qU0sj6Usb-Z7J1N~EB0S3V*an=G
z_dnPApFeeR$&VIFHRqj$=h7VWq$$p&OC`lz%ElOCm={$P)NOC&_LW_$OmE)lIdJ<q
zE<Kokr_-N5FGfISo}7Wt>?&JZ+sfKn_RdJh{gLk7k2rM~=@mw?7z^Q#q#J&;u(EoL
zO%&;LFfcHf0o}3(Twj8O(G1IzY#Y`wPjq%ez+{l*18*%8emOL+0S^<knw={-8ip&W
ztHfp8&XiwFOA#TThw1Ksf_ZUiPq{p8m1bBYTDs)J06Q~77)~=Qx?G16n{}WSn=#q&
zrNeUJ^P9}e8Hp;5c5s7Aq~}l@b=vHa_4M>K1}wgBqkj1Cp|Z7AsXG_`@M-9lxpm65
zZa>_yV>GSPPrss=nL{bxz~OOrwj9n<K8yosN@x{jzoNaFT~NRbrA=j3mBgi*niT91
z_d~}MZg&C&VVxY~_x5T?53o)M(O8E!rhe2p!Xne)ObV@oX!zTVgC~(4|LD=lYLD#f
zY?`LY#7EZI!`M2NgATI;2tmb|jVpY_#&~pO#BJ_nt6P<x4tS}YWZ9j@q0KZZ{%CU>
zrPwb0!}*|~`#rk%k;toJp46$<JOy;>_?w%X)3BRf_j};DHpj;#)B+2wb-LGL75JZ>
zP?eOtk;i_8IVkbqJV`Nc@3iVjWX{DX=*=OC5C;c56A-!?>i9H}7tjN!`sPJm5?9yC
z%Bu9qk%Z+Jb`}QnvNAH?$m2U1Sk*5whR?GpUTfJ+o4p<$9@uFinJ%;->0mc6R4IGS
zRIk$S>g)FpcVC*3eML;|O8|D8@1Dy36cMt`!s3LrjYw49)UmtyqSLy!D>1i%huhNE
z{o&j%&b{Smkg58)5NLtRn5x-~E_K5rQ`bL8g0imTd3OAXMHR}&>tyJa?*i<3`^28X
zR~Ev^uw%5P6lZ8CHoY$9^z~!z+cQ!b(yentFnEYzQ;B^GC7!=?IxLlEDvy?H6E*jd
zxE>2v#yV|GJc%Pud0kke*Y1e8+qj1mU7eTsXd4p`&;25(Hmpl4YekJD9oJ2si~YM_
z(DRliRc0;s`Yf2Wq9Z_rr=XDARy~WiOF8Q6o*OpU(3hkef}P|l=n}MX$p?eetD2ad
zDArdgOnY+#`y}!AzFcIN9V~lL5bH59v4`8z_zii+jeX~pvmCZ9;BznkWV%s>g*n_M
z#?__4lwNt?R(hs;!j=a=S&c2d(8()z-jZUCjy`!PCdiNS`NrRVR&ELGP)-n^3Uqv!
z7>!4{`))Sul3oz*$fEq1qk&pU)jtxvEq#n;_0?sml0>)<)Aa+s3?GAB_V&e=eEj*W
zcjvUs$2mLq*Dx5#_@-^_ym>^ue%@K}a4%jw+^BBthpQL!z~ZUexWS~Xny~d!j~%D@
zA2H!#X-RB7D(IQ+#jCz4F<0TNXh5G}29ZWttDUepAM!kvwp+E?)nv$c8F>|4j?L3I
zeHXuV3OW<3CA+Y@;M_IFoxG0|1gGC~%t+}X^}Q)GAx^i_<z7a?!@JvFItQAzIW(gP
zzZB2UrgiM@UDm1xyx!B){IYKYU)V*cyubg!p$#fI6Ezo2d<Kh_&YnK~+aJ-)_nFR{
z>ei1piB8DB<El`*@|oRT-h(p^d*0_cyErM-I**UdS~~CjQBZc%OH<RSxpAX>XCyu)
z55LVkHq))Mr&~r1^qra-S-r}?7nv?qYklTbX>V6@CP1-v?9F=e<}0TR4oa|?mRig-
zdya<Jqzx2Z^Vt49H&gVG-C-m0B=xiOY)SJ<&ay}ib2Nm=RKGXx;%BLjh@I`%PZinH
zXmxVqh0aSHEG~+26IoYddrBr0e^@YZCAi-{OSyPE+4#+thk<z_7F`o*c?T4+4Qi_1
z)0gh)3@x*g_x|jDt(e9g@2FvUxOIIrkbv%!U+66Ee+%EaLUdzs^1&0TyCe(A{|$zk
zt!UBN(1Z56R7JtPuiadYw~NZkuGz#0N)A(m4zz8#5YNzV?iD>HPS3RW@?d0lU*J4S
zBF-iJ<MjN<WTst|GZ5Q|Z9Fx9{%X<rR;D_23l}OIgD-sVOrkX9?6pRU7f|$KKu=Fa
z=!b*{PxZwH`BKZeF87JcF_`U<cW&?9pfY;*jgd`>KlcDdNdGkb2nL4mA`^1)>^>NM
z8O==j9*5N?#dSLI%Zvsubjq2zj1=AA2)rJxVO1oqQCziUt#GlN@0+kj%CxJ`nCl?$
zQJI$UV(DU%E@giDWl5aD=3<O?k8nPzHCf!C6VFigIpy4P7W16=sG5fLYM*zVrgu^)
zv@ssOZII`AHrXIeAj)&-&3Fl#;%<$NE$k%EdM3hT-Dq&sx>{K{m}`T*Zn0dq_7-n5
z6~)inl<gYQH)49wQ=~rjIpA)kp}m|5K!c_<JFG+_2}*6p4qI*;%^)tTB5nG0e8q0+
zxrs#w0!uG8F?0dpT2HDqH}r+u(U!4@gkd`?Yh_&>hsk6Eb!!T<wc++HQdwC7`;~H?
zoJ@D^`dZ%Cr`LKpo;~{l$3TL9#>6Go4AB>K699z*T7TSa<@BD$xIdt#asmQoz%w}Y
z7HqdqewLA)C6Mw`sQs2D^9*1=70=NmR(%tr8cPRids-X5klycK$OQ9P#E-vy^#Ud_
zB}D)#qK>Ivl~9*E>M`0KFg?jKdux?RVCFzxqSE#3+F&N>7VT8)K{mmmHTV$V7+i*K
zkSRZEo9&Ta{=QYRT4&WKr>ZTJR$Au78`re|yx=38mXWML3Xx<&(JKRMGl}Gd-KCX^
zN~87GMVtc(mKonKvHrM<^#hxU67aI>txlf21Y?TgebUV<@WIq;kMHp6>J4-&3ha79
z3mL2Lo-fw2gDW?lGxgEoRbtw6OJFQ-EVm9_P8E^@yy4WJ?`&bUxNCMQ+K+pY<@<ol
zzHL>z*b3YjJGi@f6{=@WC%8JDZ@V<qu}xWBWolSy)^Q3pNI|QN-HwenK$5c+pK*8W
zrx%rLx3=s!8(b^ba8>oCL-u(ZJoV2sp&dAg-kUfBBO{G}rj0)?HB}S_@0-f-`c-nb
z_mtZD&Wxd)A8*W`ZoH}J<vvwpGNmdHDKr4C0~SW?%J||jp6==g7rkVvNOiAC5{PqJ
zP9*?92}V%!LnL8QtVgFCHOTz-p{r!r)kdibh<G%n24-jMJkin%e39E{TUk-DjDMoy
z;!}|&G3|k_T6wX~Wl8!G;kN1NH-3FKT7IvzW9bO?BUu^tFykyHA63o7#6%wCOnV&O
z;iJR<P!W+)tt8n)&Szh?y|eJq!>A))Ur;i3DJAR;a?SB`zvn7R3wuXVK&uSJQO&m=
zw*4||_HXaF@zmSJ)O0o4B`cvj@<E}r1+2?usF%3y%fa6G;f4+(4HT3hG{d*G;C|_)
zdna)2nvR*65MW?r`R=zoZv6+h;%xJM>?$grH15^_5GP8|hip$wN=e-Pfs!E-K_@%6
z&h60+EW|vgkL7?nq>datYK|dWpcSPXq?xILh&dQAW>@!Q4T^;AEe^5UcpvkxxPD?Y
z&1n1e1@J0NL#J-M-!-3`$Z2R`t2}o4N|D-#B^@l+AAkm@{C=6X^KKt(jM*?dHi;&c
zdiI03PfC8i5}8FjmG$!Fb`)#sxi4_n(Wv4Z@a&WmJ|tbeVmc=ASA*hg2cabd)F-7Q
zhQ%oe2$abWzhC6RLdjVB;?BZogn>n~u1>D5B|{zWXqmKN?$OD=tJH*Y9vpdrVOHgj
z_Ua8LpU&Q5Haqh=^}rqWYtFXL7Mp3!R@WCG<c+B}NmMT$K#dlNQWeD_cYE%D*TDGY
z3q0NK$-+2)c;%v-Ik!AaSHW?yA*blqZuxgunq{te65XR$M6$XPl&F7S0*<zjdiG;P
zo>QUEA$4_iryNhkF3j&XM=!1<;dWFM-_K7qF{vJPiCepyQE#)Fh0#%7&K@_{&7gJb
zp!!q8kh9*FTsgc$9Eo<!jWV`RO;07Zi@uVIX^5^$jhfYypYE-)@3Tn_a<l!~P^IK0
zoA+|^j(x+WhsTfdunFO2z!;5hW|$O;q@I3}KixM7Nf+;ECvck;<TfIiN6mVfeB8<W
z*~z~q!@4;x37iu<^e&_Yx_0bkmz0(+0Yb%(L&h`Gx%Mzw+bT$?H>3Ah=q0qI_`#-`
z<%aZ$)7w8FFBnnq7=IJ}Z2zs5a*EH*hnAS>Ao#aJxh7~l09Sa$?`a{t!<`TI-{k^r
zC<NWNVoU}$bo?NdX8P(EgJ&o9fi)`z+5<c5{rXM}-BH=K>-G6+7AcZHOKb}tOo@E+
z3y4*cWP|XJ7cjPiCUsmMu|XUY*)CIAGEQ|qiy!zQ^U--O(rTfG8GIH!ot*4}sYLO)
zxn6}CyF3zG&<qaS#1s?vmUH0kt<45mPA5Gl8(V|PwgIzm6$ovPAPbm>wk)Xj4?lQB
ztxS0~KaBOLBDY`p^TOWLNiC672ksa&wHw;%+rU991aX(k4)3uwWpT_<v1|*k&C{6J
z*fpnLc7&7G)25?hpLb&=b!EN~M1qsLB5EL4qP^(4K)b2U)WS8gc6c?o+sRCIl4=9U
zG!0Ah{SyJ-OUH8jdY3vk8sr{#9<lw=d&Z4#Yc-dNZ&?J6MX0_@l8i?a+t?lH>FLwJ
zskAzE>V4+c)dG7Su59o$U@ev_J%s#ZL(ctW2WK{$rPBh_i_4>%`0hQmcoRz4+80e=
zu{u1M;+iY``m*n_U`px+8n5m}Qbbp$;2jXv*X&pP@wttP{)<De+(_xM)!e^7tE@~%
zcuL>;)MlggCiwgZWWU}UuMF{)oSHxz_gJJ``z6jp^Ori#cT+{kmR+&<zT>?UA5bQ3
z@Y=%uc>lZjrj&DepuEvw$QL4=)5at&Bg2soW=1>od}UdiSI3Q<J<;06$TA-3cnU#U
zA-y=2<>lRT#=uEP93Biy^;~y4fBO4)eYrQgozG<BGm#}id-e4Td+S*bKDd#0Xw;!y
z(5tMzMe&QY_Q8StU*WNPT^RcVMWxQv7wP;<0ERvc;)m^kK)Fq3h9bx-sC?*q!`<gt
zmXsPio<Hll*Dmf$i_Znl-x4qMzc?`!x%*oWL=+|X*R5l!V)oV!2aIEvVG2m$Ir{sl
zW6jcd$Lj<6zx?)&?g;NV@zir#09gC4d{y|SC@RN3=aOO0tUV6o>4eS>>zcLrWC%ib
zh7fL~VPs+*ERs;CZ}rC6rquJ)bm!-|u`Zbw6_J)%8IFM%#p+3i4C_!F{#E;=i{x5=
zlx2e7MgyZg<V8-$7C~k%U->tx1C@WgjaS0h&$n?0vBDmAzXb4>IiPhC9g4tn!*M92
zBH|trOxyRleLBQ%khyzTuBd*J(90frV*OW$mLI_;>_M_9!KhD8R@T{x)Y;qi$Hmed
zDm|Qd>;zT~C|$#!U16W75p<QPuye1v(;j<~@3Ng&&(T4<+3$At5XTl{k3~2irXbt#
zo6s3}SCgpGKW~KG3^#D8`O)IE_w3rme;Ithz`DE4S63CwC5VpZ89u)KSwz_cybOYk
zf&A!Z=s8$KsOZIL67ll?C{=F<snw;ngH9Loq}+7}Y@3*EV@2gH1zqXfOEuQtU7>H}
ztk;BSlm^sp4!OiY`L|h{Z;D9`Y98lNtu35<7l6Nv4dSj5!0iZwo`Mh91Gyc^n&%d2
zv*qg+Yg_Eg$=`G=N3cBIKK@8Gn8`UwmrOR%Harz-;TLNo4=${@N{?48(rzx0(QAAZ
zA0A#KO2_38)8PFm@a^1#2f3v+-*w-iFm2b`m#-UK`Y#?-UZATv_spaH!9)J;dTi>4
z<r29|7nN93r6m9lT-v-P$&QrW0zY4kBF0ejl+PnMy^UMW2B<;zdoAv~*!=>#Y{vzx
z2Sn!`7K2-WmL+}-vp-5qnbm7)MN)`Hbf0B<!e(3oI~Pj#{#^c|$4sr`z#o*GAB{;B
zG<*vRp)gq_MJerCu+^ODOy7&PccFOTVGDUc;+T!^yIU*7?6rpBBqcXG!@23|two3F
z4a@HNUh*>p>d-sBbY*Gj`D8i~OPzwk9U?N9mX+?;)I9n9IIudlWUYj6S*;F23-<DS
z{)H2-FtLH|tWMdl)=W1Ycli4AB;K0d`17H;*L!lF-r<Cm&)bUS*|F0bS@mvf3gaD>
z)%=T})+A|aZas2j&#Un;bHnXpTBGx=N^iyY9S~Qf-}_|l4XWieWma4E9-#BRv`%?f
zSzYuz3Y(^{e`K1{0*Vk2g6;8X3n+Ux-&wfk;@p$W8U73hQmprUh<v{1^kH1Y>cHD_
zwqb?<8t`NM-RFz16AuMz|Efce80JcQJW&e3jqmNYbiG7C>bf_Y2Mbgj8aCA^uVRid
z>kVcU@~b-Z<~FC*+&^OY08kdy@v<l@;EG*NljHMs2Qs(P+*SUnri;4cLad*-B;odv
zD$VSV++2Hn_H)vN2Pa?kyRCNhcTpc3Rb9%=*TlO;g(_tvvzig55b78tKYrAxvqJ6k
z{2Qi!7|=ZrZI-|rJ9JoO$1}M&zgsH#xuCm7@b&Q3RVJHL>HYgTDe9)KlJU!8)-VQd
z3yzQ|o}AW-2Hl4TgjE8}8-T@rtTi9}<n};`vViB$pIh13Tp_t0r>;hBLc{T57j=on
zxW<gQ4evyPM{?6<a#~q_NQsFp1JmF#rdzGPedmrjw9W*DgQMw8z$Bp%8svL>sp;v6
z20sr>ezXx)SX}y&iLfHoI^G~ot5QEexhJcb^S!4AJAwwYEUTlzha$ZQ3~OEwjGJBf
zP;vlV*iIO+J0~VaT&w5(umSh}+|D)ab~8d|#esg*6Ym4YC#dhZY17^bN`D7&`qX)W
zd~+ZY`DGf2-F0BRTZHfCeKc|0;tw#7i>D`Q9nV!ZTAz2U2g5}|2Qveu%l_9z>kEN>
zqD=I1mIt}wVzd5c_IQKMYAS`8!WZ}Op=GG><9&!XqshiDjkfW2=cBfLaEy~(R6%i0
z0M1z$I7AT7uUCs6NEi=ZD`r$0e>GnHP*JK|DJec)P}bP^3DowaXJFs4q%4+Xk=@6q
zr{hH>MA!xQE7m*4O&I)2r*-j9xm_Jj8ImQV@u}7{3|Rp4>{&1m33~ZP<htxJPR@+t
zrIO0LpmSSeNR}dKyj(~i>@mQ&5LJU27#(S6+I}?vlgqPhiAex@)<w`+(bB8-UqKZv
z-1GE}_`A{sID{zQwYIYh`TqSoD-|QBi`$gImfHp?fL#4?I$Z@#I05~WZqm;!Rd5d&
z|4jusG+aJXQ&N(J=lP~p7Bk;GV`L<FfD^Q)+~A6_mA(zwf;Z13>BF2)I&K6!%TwKS
zzJu5zi(WWdHaJrfP5@eIJ96Y>3caq>9og<E;4YbeM*gkeRy+vK%L{n*%I}Ws`K~Mh
zWqP(0t$+>ef`3a!Y*){>zPgY1Jg~{N8&}Lfv6}Hz_iEA(xKaLq85k*LLo|Mi2CSCi
z9odB>u70em)*d6z&|S3Ufek_Ud^d?zo4F@wwrE}Xx{$4<1{av3yt2Y9%Z0$WddgS6
zHUM^|=+fc&7s*R4NWYoFOG5YGo@B1qmgo4#f?Hf(qrojl@lC66O*u_^bR0{(I=d01
zC}QjrWi?y=p(}T~+9Pi;$gU9lBXa=M_8^k&bGSfJ{xWhm>1eGozu9j^CUjfMB&&~8
zy<!HV0db$1sWuOH4-a!N^9C_0UYeYG;4eA(ucW8k>gwfr$soJ4^`-1FhJW3lD<^po
z+2P2F)Kv;^Sw`JbeMC_XCvQD(<S#9{`22jILWKkJgDNC%w}+c$v0TewFj{BB77>dm
zy7c&#q3d|ezd@5lko%F5kwWYcTHF<p%V45z-Vf6Y*0GgWiZ-xcVP#DN1*Gfi*EcZ@
z4Nrg8eNihZ7Q1{ERsO!AP^f@nU@q+;K!N35=R=W>W!ozo8@b_Lz-4bTSFts*IA9t}
zn%iLeqSX9a(qq_vd?=Z+nmhaqj~SvCc{*6KAz-YiySln=Mi0MG0i4LM-`=;9ZXc4~
z^=7DbIdzv%3M|-SSfC`VCczEUBP7x0UPXGJn+=|<h6ex}u${{R0i*~5<<1P^TYG4h
zN+84I!Em(FOt@wX`K{xLxBBZAla`b$1chMvizbJ677+{-<>G(Rx%wZq=Kg8#_qW}8
z3@6Ho^5C0W?~i083HMy4#JKPMkVT2fa&lI6a#HDS)OT;D>bvu@YHzU6Uq#VfB2iJW
zFiD?(Y5A`2eZodChPC>%OwRRz5u>lWpfbH+diuTJe{^zDliU`ZQ(I?{1!t>@f;Iox
zrsIfP!g%K9&K;%<zJh>RDU`por~h(B|2KM)ZWmHg0_vX$KG?g_tFQZQmO$c%>f|LD
zyToo}gl;YC<o7V+vC%MNb(Iy1>uXe<$vb~TmcioRxdy*WdA$rt)TiU*TRjB6OreO{
zoYctP@3kBe!nn4cv$9TR=!m)bSJ|b+qsSOI`-&^9f9o-ATedY6Bbq3wS{oL==edg*
z^8c^N`fu*!pIW!NX@B?qDW=x73n=kD|Frp<;Cdn^cF*HKHD61EDNK4`jUtOH_`D3i
z%pf<FB82#+m-tK~F3tXPMse{+XUb~1d6K%1cCLvt2)rpJ92Xlql3|d3ZW-n2gC&f*
zX>+&L`tMS$|Hf^cd#dt<-`dK*x01Cr4nD(7zSRe?aAjo$un2Se$)OUkVa3y=D{slY
z)Y|dK0$tA!CpzE74KF47WHU$_oDeft1rw{Om|tY&1i84?jcrJ+`}2vLRtQ`!3OM{s
zfQxI7$rV-Zg~KYCFz35ic*XoL@)G}@C8}`w!yV}o;6(8m8B$<cy4H;iKmjI03zUd^
z{CG=4LqmI#$+R_t-aN|H1?yS){!JKX9V2G{Y2MQlfr;}etA$U*un?f#%GNdr&TWP0
zR-#Y2G7V^QG7Rrx_AIQp*flT^0>qd(|Mcn8!80>6uluthW}IVm5O@JpQ=w-n6v|eM
zMI{`Dkawjybi4(|$JR~|KM+V|F)AhYb-(a$jbDQ2vIEGdzh4_U`y+bxl0=z|V9!?T
zUo68Qp>(k@+NswwFd$tU_QvOR$6(~sW65$^{;yv3e0UCO#v(cml^wdTUL{{i5Q2DQ
zcN1WmLDAVEQG|IQj{yPWfSZT42lzQS0;=}QZGkYgK7Pg#zzGMECYDV{Pcfg}-!Yt=
zpf=-kSg|}&K8mfTbb0VDKvf&a`3CA#8kllj{r$lx=z=mk!NTU58AbwWhWV5~*&%7?
zY0E%(3IZ*$1a<-DI5SgTez|P`pBy;t2;sYNfR%AY6p2w9F#MZHp(0%xPTE8Nciadb
zy4O{F`BeYh=SA*?bM>{R+}oPUO6_Y6LuHPeqAgXWhGyRlojrtWkf2tJ#6Wo-$t#d8
zDeOQtZBmpC{)NnB?=u<>WLMPbodIJrgaGs9%a`#grRge-WU-?qvLe$!_iPDbP}$JH
z1#bXvqZd`%YoBpB0OtjE=GWZ4cawQ@8elH-<oNJ~6>Jueou&c(>FVnX0z;n1S#12h
zr6L>587kKvig+AL`-tH)Gj$Nqo&pXGGQzg@s8p{$8*vDr#$gD+hc9a6IZ+i|me6o2
zY&cm!APQ-7CIr>(XsDFO?CHLh*53u>P`p;#W`5VAr&w&6kW2~+v_oVV>@)Guhif#C
zA4KTE>VpT0JX(gM-fR{<*ufJq^S1K^$-V)5RMQ-$9YVrq$tR1R8F@(4OLf;{(9<ao
zbXK>_5e`WUJ|d`m;3{yKO~#N5AmQjS9?NnAQqP>-38wqtV2F?W44?`3r|;hBZZEl?
zKNLlPwT%xWmt&yo`+K#Ancki~q%=_(Vuq0rgN#1k7_S@gar6Yga?|HC=gm;#&bD`>
zX1A;)cocJXOMU(Jc4u&qK$_ukX$9ukK@Mfyi$zJ-Jmt=OQl6@hp1x)LdG7{u-op~L
z+P3q%?A<WuJY+nV(1He$gBL@rJ3(alG&Gjvd!s;@Tg&G?-ZnlIk<WwjAP6Z%2^OSh
zwbc2u75{)ON8!!fph#)QkaJjwo!IO0=$2gVGvYhhzFz5v#%ALj88Czfub31f?KeZZ
zPg3|dA`y{WYlH3%rpOn|T9>V6YiEldFZ=|a1ta+%zPsG`n*-iM31dXX&TojegvE{U
zmjr?Nmi(-YYbq50&nx0K;D@Nt{|QgM&N_yg!TZ~Rna>A+X)!WN!?Tyz16XQXf5Amf
z9r4H?^YEJ;<ewG4v{Gl_k6UKWG&29?%tQ}y#R>qIU5*c<;4|?6)&{Dc9iHF(_}^z{
zoyr}`9QlpNB<^WA@sq<8GCLkZLt1$WRVYDJyUiY0OLoW$mxlha+#*$Am9*&JW<#-i
zK~ynJ&&Xh!wL|w092K^Qkq&$C00F?-Nw5(VP&4$Yy3?`(S8Dy%6uuZ-I+<<jS$E#r
zkA#)<V$=;A0rgGw56qnQ>^a(cxfNg{>HLsJlS<*YQS3uSScT+d_C9^rPJrAvC*P->
zJ)i2{EAP4x>GV-!@cWB=1>)KelIe|vQSIo_qa$Numb#qE9UMK%zWz@YfXcq+KWPjr
zh>Kpc<ZGil;XyERR_a)9JsUuLVvVp5##QDglKsq|ou_{**-&zul@-FIGm)4<f!^c|
za&1W^K)xk;JZJQ=T~JF)%XnRhB73SwbI;ZUuGK+l%)$G7{q4Dq%(~v|A-Sz<4vp6y
zG%^)@EQXcy0;AKInV+7g7pgmKd3Pn%#K)`2scPThh<Rei0-dYji7WI9cyHS{dLwLd
zF1@8$D|{Cr*ceMLWoBkRVFVNB>cs@B`X4jK{%gq$h4Qc1i~rN~@V}da{O1Gw=L-Bk
eU4d`2bXWKSuLW?Y<$;e((NNv5lCsnEm;VEvwHcWJ

diff --git a/examples_skyline/fair_prep_results.py b/examples_skyline/fair_prep_results.py
deleted file mode 100644
index e2f9565..0000000
--- a/examples_skyline/fair_prep_results.py
+++ /dev/null
@@ -1,92 +0,0 @@
-import os
-import sys
-import shutil
-import warnings
-warnings.simplefilter('ignore')
-
-from helper import extract_info
-
-sys.path.append(os.getcwd())
-
-from fp.traindata_samplers import CompleteData
-from fp.missingvalue_handlers import CompleteCaseAnalysis
-from fp.dataset_experiments import GermanCreditDatasetSexExperiment
-from fp.scalers import NamedStandardScaler, NamedMinMaxScaler
-from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          
-from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing
-from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing
-
-import numpy as np
-import matplotlib.pyplot as plt
-
-#creating list of parameters that we will alter to observe variations
-seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]
-learners = [NonTunedLogisticRegression(), LogisticRegression()]
-processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing())]
-
-def calculate_metrics(seed, learner, pre_processor, post_processor):
-    '''
-    Experiment function to run the experiments
-    '''
-    exp = GermanCreditDatasetSexExperiment(
-        fixed_random_seed=seed,
-        train_data_sampler=CompleteData(),
-        missing_value_handler=CompleteCaseAnalysis(),
-        numeric_attribute_scaler=NamedStandardScaler(),
-        learners=[learner],
-        pre_processors=[pre_processor],
-        post_processors=[post_processor])
-    exp.run()
-
-def run_exp(seeds, learners, processors):
-    '''
-    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.
-    '''
-    accuracy, disp_imp, fnr, fpr = [], [], [], []
-    for processor in processors:
-        for learner in learners:
-            learner_acc, learner_di, learner_fnr, learner_fpr = [], [], [], []
-            for seed in seeds:    
-                calculate_metrics(seed, learner, pre_processor=processor[0], post_processor=processor[1])
-                extract_info(learner_acc, learner_di, learner_fnr, learner_fpr)
-            accuracy.append(learner_acc)
-            disp_imp.append(learner_di)
-            fnr.append(learner_fnr)
-            fpr.append(learner_fpr)
-    
-    return accuracy, disp_imp, fnr, fpr
-
-accuracy, disp_imp, fnr, fpr  = run_exp(seeds, learners, processors)
-print(accuracy)
-print(disp_imp)
-print(fnr)
-print(fpr)
-
-def plotter(x, y, x_ticks, x_label, main_title):
-    '''
-    Function to plot various comparison plots.
-    '''
-    title_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']
-    label_list = [('NonTunedLogistic', 'TunedLogistic'), ('NonTunedDecisionTree', 'TunedDecisionTree')] 
-    fig, axs = plt.subplots(6, 2, figsize=((10,20)))
-    axs = axs.flatten()
-    for i in range(0, len(y), 2):
-        loc = i//2
-        axs[loc].scatter(x[i], y[i], c='b', marker='o')
-        axs[loc].scatter(x[i+1], y[i+1], c='r', marker='o')
-        axs[loc].set_xticks(x_ticks)
-        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))
-        axs[loc].set_title(title_list[i//4])
-        axs[loc].grid(True)
-        axs[loc].set_xlabel(x_label)
-        axs[loc].set_ylabel('Accuracy')
-        axs[loc].legend(label_list[int(i%4/2)])
-    fig.suptitle(main_title)
-    plt.subplots_adjust(wspace=0.3, hspace=0.43)
-    fig.savefig('examples/' + main_title + '.png')
-    plt.show()
-    
-plotter(x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di')
-plotter(x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr')
-plotter(x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr')
-	
diff --git a/examples_skyline/helper.py b/examples_skyline/helper.py
deleted file mode 100644
index 05af226..0000000
--- a/examples_skyline/helper.py
+++ /dev/null
@@ -1,53 +0,0 @@
-import os 
-import sys
-import shutil
-import pandas as pd
-import numpy as np
-
-path = 'logs'
-def extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal):
-    try:
-        for folder_name, sub_folders, file_names in os.walk(path):
-            for sub_folder in sub_folders:
-                file_list = os.listdir(os.path.join(path, sub_folder))
-                file_list.sort()
-                for file in file_list:
-                    file_path = os.path.join(path,sub_folder, file)
-                    df = pd.read_csv(str(file_path), header=None, names = ['split_type', 'label', 'parameter', 'value'])
-                    test_data = df['split_type'] == 'test'
-                    
-                    flags = file.split("__")
-                    pre_processor = flags[1]
-                    post_processor = flags[2]
- 
-                    if(len(flags) == 8 and flags[-1] == "OPTIMAL.csv"):
-                        learner_optimal.append(True)
-                    else:
-                        learner_optimal.append(False)
-                        
-                    if(post_processor == "no_post_processing"):
-                        learner_label.append(pre_processor)
-                    else:
-                        learner_label.append(post_processor[:17])
-
-                    di = (df['parameter'] == 'disparate_impact') & test_data
-                    di_value = df.loc[di, :]
-
-                    learner_di.append(di_value.iloc[0, -1])
-                    
-                    acc = (df['parameter'] == 'accuracy') & test_data
-                    acc_value = df.loc[acc, :]
-                    learner_acc.append(acc_value.iloc[0, -1])
-
-                    fnr = (df['parameter'] == 'generalized_false_negative_rate') & test_data
-                    fnr_value = df.loc[fnr, :]
-                    learner_fnr.append(fnr_value.iloc[0, -1])
-
-                    fpr = (df['parameter'] == 'generalized_false_positive_rate') & test_data
-                    fpr_value = df.loc[fpr, :]
-                    learner_fpr.append(fpr_value.iloc[0, -1])
-
-                shutil.rmtree(path + '/' + sub_folder)
-    except:
-        pass
-
diff --git a/examples_skyline/missing_data_results.py b/examples_skyline/missing_data_results.py
deleted file mode 100644
index 29186ce..0000000
--- a/examples_skyline/missing_data_results.py
+++ /dev/null
@@ -1,99 +0,0 @@
-import os
-import sys
-import shutil
-import warnings
-warnings.simplefilter('ignore')
-
-from helper import extract_info
-
-sys.path.append(os.getcwd())
-
-from fp.traindata_samplers import CompleteData
-from fp.missingvalue_handlers import CompleteCaseAnalysis, ModeImputer, DataWigSimpleImputer
-from fp.dataset_experiments import AdultDatasetWhiteExperiment
-from fp.scalers import NamedStandardScaler, NamedMinMaxScaler
-from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          
-from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing
-from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing
-
-import numpy as np
-import itertools
-import matplotlib.pyplot as plt
-
-#creating list of parameters that we will alter to observe variations
-seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]
-learners = [NonTunedLogisticRegression(), LogisticRegression(), NonTunedDecisionTree(), DecisionTree()]
-processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()),
-              (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]
-impute_column_list = ['workclass', 'occupation', 'native-country']
-label_column = 'income-per-year'
-datawig_imputer = DataWigSimpleImputer(impute_column_list, label_column,out='out')
-missing_value_imputers = [CompleteCaseAnalysis(),  ModeImputer(impute_column_list), datawig_imputer]
-
-
-def calculate_metrics(seed, learner, missing_value_imputer,pre_processor,post_processor):
-    '''
-    Experiment function to run the experiments
-    '''
-    exp = AdultDatasetWhiteExperiment( 
-        fixed_random_seed=seed,
-        train_data_sampler=CompleteData(),
-        missing_value_handler=missing_value_imputer,
-        numeric_attribute_scaler=NamedStandardScaler(),
-        learners=[learner],
-        pre_processors=[pre_processor],
-        post_processors=[post_processor])
-    exp.run()
-
-def run_exp(seeds, learners, processors):
-    '''
-    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.
-    '''
-    accuracy, disp_imp, fnr, fpr = [], [], [], []
-    for learner in learners:
-        for processor in processors:
-            for imputer in missing_value_imputers:
-                imputer_acc, imputer_di, imputer_fnr, imputer_fpr = [], [], [], []
-                for seed in seeds:    
-                    calculate_metrics(seed, learner, imputer, pre_processor=processor[0], post_processor=processor[1])
-                    extract_info(imputer_acc, imputer_di, imputer_fnr, imputer_fpr)
-                accuracy.append(imputer_acc)
-                disp_imp.append(imputer_di)
-                fnr.append(imputer_fnr)
-                fpr.append(imputer_fpr)
-    return accuracy, disp_imp, fnr, fpr
-
-accuracy, disp_imp, fnr, fpr  = run_exp(seeds, learners, processors)
-
-def plotter(x, y, x_ticks, x_label, main_title):
-    '''
-    Function to plot various comparison plots.
-    '''
-    learner_list = ['NonTunedLogistic','TunedLogistic', 'NonTunedDecisionTree', 'TunedDecisionTree']
-    processor_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']
-    title_list = list(itertools.product(learner_list,processor_list))
-    label_list = [('CompleteCase',  'ModeImputer', 'datawig_simple')]
-    fig, axs = plt.subplots(len(learner_list), len(processor_list), figsize=((10,20)))
-    axs = axs.flatten()
-    for i in range(0, len(y), 3):
-        loc = i//3
-        axs[loc].scatter(x[i], y[i], c='b', marker='o')
-        axs[loc].scatter(x[i+1], y[i+1], c='r', marker='o')
-        axs[loc].scatter(x[i+2], y[i+2], c='g', marker='o')
-        axs[loc].set_xticks(x_ticks)
-        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))
-        axs[loc].set_title(title_list[i//3],fontsize=8)
-
-        axs[loc].grid(True)
-        axs[loc].set_xlabel(x_label)
-        axs[loc].set_ylabel('Accuracy')
-        axs[loc].legend(label_list[int(i%3/2)])
-    fig.suptitle(main_title)
-    plt.subplots_adjust(wspace=0.3, hspace=0.43)
-    fig.savefig('examples/' + main_title + '.png')
-    plt.show()
-
-plotter(x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='missing_data_accuracy_vs_di')
-plotter(x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='missing_data_accuracy_vs_fnr')
-plotter(x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='missing_data_accuracy_vs_fpr')
-
diff --git a/examples_skyline/results_play_skyline_formula.ipynb b/examples_skyline/results_play_skyline_formula.ipynb
deleted file mode 100644
index e6ed2c6..0000000
--- a/examples_skyline/results_play_skyline_formula.ipynb
+++ /dev/null
@@ -1,1778 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Import error: No module named 'tensorflow'\n"
-     ]
-    }
-   ],
-   "source": [
-    "#load relevant libraries\n",
-    "import os\n",
-    "import sys\n",
-    "import pathlib\n",
-    "import shutil\n",
-    "import warnings\n",
-    "warnings.simplefilter('ignore')\n",
-    "from helper import extract_info\n",
-    "\n",
-    "os.chdir('..')\n",
-    "#parent_dir = os.path.dirname(os.getcwd())\n",
-    "#sys.path.append(parent_dir)\n",
-    "\n",
-    "from fp.traindata_samplers import CompleteData\n",
-    "from fp.missingvalue_handlers import CompleteCaseAnalysis\n",
-    "from fp.dataset_experiments import GermanCreditDatasetSexExperiment\n",
-    "from fp.scalers import NamedStandardScaler, NamedMinMaxScaler\n",
-    "from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          \n",
-    "from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing\n",
-    "from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing\n",
-    "\n",
-    "import numpy as np\n",
-    "import matplotlib.pyplot as plt"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "#parameters in this cell can be adjusted for experimentation\n",
-    "\n",
-    "seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]\n",
-    "#seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe]\n",
-    "learners = [NonTunedLogisticRegression(), LogisticRegression()] \n",
-    "#tuned and non tuned DecisionTree() can also be used.\n",
-    "\n",
-    "#pairs of pre and post processors\n",
-    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()), (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]\n",
-    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing())]\n",
-    "pre_processors = [NoPreProcessing(), DIRemover(1.0)]\n",
-    "post_processors = [NoPostProcessing(), RejectOptionPostProcessing()]\n",
-    "\n",
-    "skyline_formula = {'accuracy': 0.6, 'selection_rate': 0.2, 'false_discovery_rate': 0.2}\n",
-    "\n",
-    "# TODO: add more test cases for above parameters\n",
-    "#update these for the purpose of plotting as per your experiment\n",
-    "#title_list = ['NoPreProcessing', 'DIRemover(1.0)', 'DIRemover(0.5)', 'Reweighing', 'Reject Option', 'Caliberated Equal Odds']\n",
-    "#add pair wise tuples for each pair of learners.\n",
-    "label_list = [('NonTunedLogistic', 'TunedLogistic')]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def calculate_metrics(seed, learner, pre_processor, post_processor, val_strategy):\n",
-    "    '''\n",
-    "    Experiment function to run the experiments\n",
-    "    '''\n",
-    "    exp = GermanCreditDatasetSexExperiment(\n",
-    "        fixed_random_seed=seed,\n",
-    "        train_data_sampler=CompleteData(),\n",
-    "        missing_value_handler=CompleteCaseAnalysis(),\n",
-    "        numeric_attribute_scaler=NamedStandardScaler(),\n",
-    "        learners=learner,\n",
-    "        pre_processors=pre_processor,\n",
-    "        post_processors=post_processor,\n",
-    "        optimal_validation_strategy=val_strategy)\n",
-    "    exp.run()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  50 tasks      | elapsed:    4.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    4.2s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.0s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.0s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n"
-     ]
-    }
-   ],
-   "source": [
-    "def run_exp(seeds, learners, pre_processors, post_processors, optimal_validation_strategies):\n",
-    "    '''\n",
-    "    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.\n",
-    "    '''\n",
-    "    total_experiments = len(pre_processors) * len(post_processors) * len(learners)\n",
-    "    #pd.set_option(\"display.max_rows\", None, \"display.max_columns\", None)\n",
-    "    accuracy, disp_imp, fnr, fpr, label, optimal = [], [], [], [], [], []\n",
-    "    for j in range(total_experiments):\n",
-    "        accuracy.append(list())\n",
-    "        disp_imp.append(list())\n",
-    "        fnr.append(list())\n",
-    "        fpr.append(list())\n",
-    "        label.append(list())\n",
-    "        optimal.append(list())\n",
-    "    for seed in seeds:  \n",
-    "        learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal = [], [], [], [], [], []\n",
-    "        calculate_metrics(seed, learners, pre_processors, post_processors, optimal_validation_strategies)\n",
-    "        extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal)\n",
-    "        for i in range(len(learner_acc)):\n",
-    "            accuracy[i].append(learner_acc[i])\n",
-    "            disp_imp[i].append(learner_di[i])\n",
-    "            fnr[i].append(learner_fnr[i])\n",
-    "            fpr[i].append(learner_fpr[i])\n",
-    "            #label[i].append(learner_label[i])\n",
-    "            optimal[i].append(learner_optimal[i])\n",
-    "    label = learner_label\n",
-    "    \n",
-    "    return accuracy, disp_imp, fnr, fpr, label, optimal\n",
-    "    #return learner_acc, learner_di, learner_fnr, learner_fpr, label, optimal\n",
-    "\n",
-    "s_accuracy, s_disp_imp, s_fnr, s_fpr, s_label, s_optimal  = run_exp(seeds, learners, pre_processors, post_processors, skyline_formula)\n",
-    "\n",
-    "#Converting string types to floating values for plotting\n",
-    "\n",
-    "accuracy, disp_imp, fnr, fpr = [],[],[],[]\n",
-    "for item in s_accuracy:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    accuracy.append(temp_list)\n",
-    "\n",
-    "for item in s_disp_imp:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    disp_imp.append(temp_list)\n",
-    "\n",
-    "for item in s_fnr:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    fnr.append(temp_list)\n",
-    "\n",
-    "for item in s_fpr:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    fpr.append(temp_list)\n",
-    "\n",
-    "title_list = s_label"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def plotter(s_optimal, title_list, label_list, x, y, x_ticks, x_label, main_title):\n",
-    "    '''\n",
-    "    Function to plot various comparison plots.\n",
-    "    '''\n",
-    "    #update label list and title list as per the experiment being performed.\n",
-    "    fig, axs = plt.subplots(2, 2, figsize=((10,20)))\n",
-    "    axs = axs.flatten()\n",
-    "    for i in range(0, len(y), 2):\n",
-    "        loc = i//2\n",
-    "        for k in range(len(x[i])):\n",
-    "            if s_optimal[i][k]:\n",
-    "                axs[loc].scatter(x[i][k], y[i][k], c='y', marker='o')\n",
-    "            else:\n",
-    "                axs[loc].scatter(x[i][k], y[i][k], c='b', marker='o')\n",
-    "            if s_optimal[i+1][k]:\n",
-    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='g', marker='o')\n",
-    "            else:\n",
-    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='r', marker='o')\n",
-    "        axs[loc].set_xticks(x_ticks)\n",
-    "        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))\n",
-    "        axs[loc].set_title(title_list[i])\n",
-    "        axs[loc].grid(True)\n",
-    "        axs[loc].set_xlabel(x_label)\n",
-    "        axs[loc].set_ylabel('Accuracy')\n",
-    "        axs[loc].legend(label_list[int(i%(len(label_list)*2)/2)])\n",
-    "    fig.suptitle(main_title)\n",
-    "    plt.subplots_adjust(wspace=0.3, hspace=0.43)\n",
-    "    fig.savefig('examples_skyline/' + main_title + '.png')\n",
-    "    plt.show()\n",
-    "\n",
-    "plotter(s_optimal, title_list, label_list, x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di-skyline_formula')\n",
-    "plotter(s_optimal, title_list, label_list, x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr-skyline_formula')\n",
-    "plotter(s_optimal, title_list, label_list, x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr-skyline_formula')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.7.3"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 1
-}
diff --git a/examples_skyline/results_play_skyline_order.ipynb b/examples_skyline/results_play_skyline_order.ipynb
deleted file mode 100644
index 1ab6950..0000000
--- a/examples_skyline/results_play_skyline_order.ipynb
+++ /dev/null
@@ -1,1767 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Import error: No module named 'tensorflow'\n"
-     ]
-    }
-   ],
-   "source": [
-    "#load relevant libraries\n",
-    "import os\n",
-    "import sys\n",
-    "import pathlib\n",
-    "import shutil\n",
-    "import warnings\n",
-    "warnings.simplefilter('ignore')\n",
-    "from helper import extract_info\n",
-    "\n",
-    "os.chdir('..')\n",
-    "#parent_dir = os.path.dirname(os.getcwd())\n",
-    "#sys.path.append(parent_dir)\n",
-    "\n",
-    "from fp.traindata_samplers import CompleteData\n",
-    "from fp.missingvalue_handlers import CompleteCaseAnalysis\n",
-    "from fp.dataset_experiments import GermanCreditDatasetSexExperiment\n",
-    "from fp.scalers import NamedStandardScaler, NamedMinMaxScaler\n",
-    "from fp.learners import NonTunedLogisticRegression, LogisticRegression, DecisionTree, NonTunedLogisticRegression, NonTunedDecisionTree          \n",
-    "from fp.post_processors import NoPostProcessing, RejectOptionPostProcessing, EqualOddsPostProcessing, CalibratedEqualOddsPostProcessing\n",
-    "from fp.pre_processors import NoPreProcessing, DIRemover, Reweighing\n",
-    "\n",
-    "import pandas as pd\n",
-    "import numpy as np\n",
-    "import matplotlib.pyplot as plt"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "#parameters in this cell can be adjusted for experimentation\n",
-    "\n",
-    "seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe, 0xdeadbeef, 0xbeefcafe, 0xcafebeef, 50, 583, 5278, 100000, 0xefac,0xfeeb, 0xdaed, 0xefacdaed, 0xfeebdead]\n",
-    "#seeds = [0xbeef, 0xcafe, 0xdead, 0xdeadcafe]\n",
-    "learners = [NonTunedLogisticRegression(), LogisticRegression()] \n",
-    "#tuned and non tuned DecisionTree() can also be used.\n",
-    "\n",
-    "#processors = [(NoPreProcessing(), NoPostProcessing()), (DIRemover(1.0), NoPostProcessing()), (DIRemover(0.5), NoPostProcessing()), (Reweighing(), NoPostProcessing()), (NoPreProcessing(), RejectOptionPostProcessing()), (NoPreProcessing(), CalibratedEqualOddsPostProcessing())]\n",
-    "skyline_order = ['accuracy', 'selection_rate', 'false_discovery_rate']\n",
-    "pre_processors = [NoPreProcessing(), DIRemover(1.0)]\n",
-    "post_processors = [NoPostProcessing(), RejectOptionPostProcessing()]\n",
-    "#add pair wise tuples for each pair of learners.\n",
-    "label_list = [('NonTunedLogistic', 'TunedLogistic')]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def calculate_metrics(seed, learner, pre_processor, post_processor, val_strategy):\n",
-    "    '''\n",
-    "    Experiment function to run the experiments\n",
-    "    '''\n",
-    "    exp = GermanCreditDatasetSexExperiment(\n",
-    "        fixed_random_seed=seed,\n",
-    "        train_data_sampler=CompleteData(),\n",
-    "        missing_value_handler=CompleteCaseAnalysis(),\n",
-    "        numeric_attribute_scaler=NamedStandardScaler(),\n",
-    "        learners=learner,\n",
-    "        pre_processors=pre_processor,\n",
-    "        post_processors=post_processor,\n",
-    "        optimal_validation_strategy=val_strategy)\n",
-    "    exp.run()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {
-    "scrolled": true
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    2.0s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.6s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.6s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.6s finished\n",
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from training data\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.5s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  53 out of  60 | elapsed:    0.4s remaining:    0.1s\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.4s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "Fitting 5 folds for each of 12 candidates, totalling 60 fits\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
-      "[Parallel(n_jobs=-1)]: Done  60 out of  60 | elapsed:    0.5s finished\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n",
-      "complete_case removed 0 instances from validation data\n",
-      "Injecting zero columns for features not present set()\n"
-     ]
-    }
-   ],
-   "source": [
-    "def run_exp(seeds, learners, pre_processors, post_processors, optimal_validation_strategies):\n",
-    "    '''\n",
-    "    This is the main driver function that calls the calculate_metrics to give metrices on combinations of various learners, pre and post processing techniques.\n",
-    "    '''\n",
-    "    total_experiments = len(pre_processors) * len(post_processors) * len(learners)\n",
-    "    #pd.set_option(\"display.max_rows\", None, \"display.max_columns\", None)\n",
-    "    accuracy, disp_imp, fnr, fpr, label, optimal = [], [], [], [], [], []\n",
-    "    for j in range(total_experiments):\n",
-    "        accuracy.append(list())\n",
-    "        disp_imp.append(list())\n",
-    "        fnr.append(list())\n",
-    "        fpr.append(list())\n",
-    "        label.append(list())\n",
-    "        optimal.append(list())\n",
-    "    for seed in seeds:  \n",
-    "        learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal = [], [], [], [], [], []\n",
-    "        calculate_metrics(seed, learners, pre_processors, post_processors, optimal_validation_strategies)\n",
-    "        extract_info(learner_acc, learner_di, learner_fnr, learner_fpr, learner_label, learner_optimal)\n",
-    "        for i in range(len(learner_acc)):\n",
-    "            accuracy[i].append(learner_acc[i])\n",
-    "            disp_imp[i].append(learner_di[i])\n",
-    "            fnr[i].append(learner_fnr[i])\n",
-    "            fpr[i].append(learner_fpr[i])\n",
-    "            #label[i].append(learner_label[i])\n",
-    "            optimal[i].append(learner_optimal[i])\n",
-    "    label = learner_label\n",
-    "    \n",
-    "    return accuracy, disp_imp, fnr, fpr, label, optimal\n",
-    "    #return learner_acc, learner_di, learner_fnr, learner_fpr, label, optimal\n",
-    "\n",
-    "s_accuracy, s_disp_imp, s_fnr, s_fpr, s_label, s_optimal  = run_exp(seeds, learners, pre_processors, post_processors, skyline_order)\n",
-    "\n",
-    "#Converting string types to floating values for plotting\n",
-    "\n",
-    "accuracy, disp_imp, fnr, fpr = [],[],[],[]\n",
-    "for item in s_accuracy:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    accuracy.append(temp_list)\n",
-    "\n",
-    "for item in s_disp_imp:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    disp_imp.append(temp_list)\n",
-    "\n",
-    "for item in s_fnr:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    fnr.append(temp_list)\n",
-    "\n",
-    "for item in s_fpr:\n",
-    "    temp_list = list()\n",
-    "    for i in item:\n",
-    "        temp_list.append(float(i))\n",
-    "    fpr.append(temp_list)\n",
-    "\n",
-    "title_list = s_label"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def plotter(s_optimal, title_list, label_list, x, y, x_ticks, x_label, main_title):\n",
-    "    '''\n",
-    "    Function to plot various comparison plots.\n",
-    "    '''\n",
-    "    #update label list and title list as per the experiment being performed.\n",
-    "    fig, axs = plt.subplots(len(pre_processors), len(post_processors), figsize=((10,20)))\n",
-    "    axs = axs.flatten()\n",
-    "    for i in range(0, len(y), 2):\n",
-    "        loc = i//2\n",
-    "        for k in range(len(x[i])):\n",
-    "            if s_optimal[i][k]:\n",
-    "                axs[loc].scatter(x[i][k], y[i][k], c='y', marker='o')\n",
-    "            else:\n",
-    "                axs[loc].scatter(x[i][k], y[i][k], c='b', marker='o')\n",
-    "            if s_optimal[i+1][k]:\n",
-    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='g', marker='o')\n",
-    "            else:\n",
-    "                axs[loc].scatter(x[i+1][k], y[i+1][k], c='r', marker='o')\n",
-    "        axs[loc].set_xticks(x_ticks)\n",
-    "        axs[loc].set_yticks(np.arange(0.5, 1, 0.1))\n",
-    "        axs[loc].set_title(title_list[i])\n",
-    "        axs[loc].grid(True)\n",
-    "        axs[loc].set_xlabel(x_label)\n",
-    "        axs[loc].set_ylabel('Accuracy')\n",
-    "        axs[loc].legend(label_list[int(i%(len(label_list)*2)/2)])\n",
-    "    fig.suptitle(main_title)\n",
-    "    plt.subplots_adjust(wspace=0.3, hspace=0.43)\n",
-    "    fig.savefig('examples_skyline/' + main_title + '.png')\n",
-    "    plt.show()\n",
-    "\n",
-    "plotter(s_optimal, title_list, label_list, x=disp_imp, y=accuracy, x_ticks=[0.5, 1, 1.5], x_label='DI', main_title='accuracy_vs_di-skyline-order')\n",
-    "plotter(s_optimal, title_list, label_list, x=fnr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FNR', main_title='accuracy_vs_fnr-skyline-order')\n",
-    "plotter(s_optimal, title_list, label_list, x=fpr, y=accuracy, x_ticks=[-0.4, 0, 0.4], x_label='FPR', main_title='accuracy_vs_fpr-skyline-order')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.7.3"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 1
-}
diff --git a/fp/dataset_experiments.py b/fp/dataset_experiments.py
index 91e0229..08b07aa 100644
--- a/fp/dataset_experiments.py
+++ b/fp/dataset_experiments.py
@@ -6,7 +6,7 @@
 class AdultDatasetWhiteMaleExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -27,16 +27,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}, {1.0: 'Male', 0.0: 'Female'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'adultwhitemale', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultwhitemale')
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -45,7 +41,7 @@ def load_raw_data(self):
 class AdultDatasetMaleExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -66,16 +62,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'Male', 0.0: 'Female'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'adultmale', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultmale')
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -84,7 +76,7 @@ def load_raw_data(self):
 class AdultDatasetWhiteExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -105,16 +97,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
             'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'adultwhite', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'adultwhite')
 
     def load_raw_data(self):
         return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
@@ -124,7 +112,7 @@ def load_raw_data(self):
 class PropublicaDatasetWhiteExperiment(BinaryClassificationExperiment):
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
         label_name = 'two_year_recid'
@@ -144,16 +132,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'Caucasian', 0.0: 'Non-white'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'propublicawhite', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'propublicawhite')
 
     def load_raw_data(self):
         """The custom pre-processing function is adapted from
@@ -179,7 +163,7 @@ class GermanCreditDatasetSexExperiment(BinaryClassificationExperiment):
 
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -202,16 +186,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'male', 0.0: 'female'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'germancreditsex', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'germancreditsex')
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/german.csv', na_values='?', sep=',')
@@ -273,7 +253,7 @@ class RicciRaceExperiment(BinaryClassificationExperiment):
     '''
 
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -294,16 +274,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
             'label_maps': [{1.0: 1, 0.0: 0}],
             'protected_attribute_maps': [{1.0: 'W', 0.0: 'NW'}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'riccirace', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+        
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names,attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'riccirace')
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/ricci.txt', na_values='?', sep=',')
@@ -329,7 +305,7 @@ class GiveMeSomeCreditExperiment(BinaryClassificationExperiment):
     Fairness intervention for the Age attribute (priviledge for age>=25) while predicting if a person will experience 90 days past due delinquency or worse. 
     '''
     def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors, optimal_validation_strategy):
+                 learners, pre_processors, post_processors):
 
         test_set_ratio = 0.2
         validation_set_ratio = 0.1
@@ -351,16 +327,12 @@ def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler,
         dataset_metadata = {
             'label_maps': [{1.0: 1, 0.0: 0}]
         }
-        if optimal_validation_strategy:
-            super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                             numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                             train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners,
-                             pre_processors,
-                             post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                             unprivileged_groups, dataset_metadata, 'givecredit', optimal_validation_strategy)
-        else:
-            print("Need to specify the strategy that is used to select the optimal results!")
-            exit()
+
+        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
+                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
+                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
+                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
+                         unprivileged_groups, dataset_metadata, 'givecredit')
 
     def load_raw_data(self):
         df = pd.read_csv('datasets/raw/givemesomecredit.csv', na_values='?', sep=',',index_col=False)
diff --git a/fp/dataset_experiments_old.py b/fp/dataset_experiments_old.py
deleted file mode 100644
index 08b07aa..0000000
--- a/fp/dataset_experiments_old.py
+++ /dev/null
@@ -1,341 +0,0 @@
-from fp.experiments import BinaryClassificationExperiment
-import pandas as pd
-import numpy as np
-
-
-class AdultDatasetWhiteMaleExperiment(BinaryClassificationExperiment):
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'income-per-year'
-        positive_label = '>50K'
-        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
-        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
-                                       'native-country']
-        attributes_to_drop_names = ['fnlwgt']
-
-        protected_attribute_names = ['race', 'sex']
-        privileged_classes = [['White'], ['Male']]
-
-        privileged_groups = [{'race': 1, 'sex': 1}]
-        unprivileged_groups = [{'race': 1, 'sex': 0}, {'sex': 0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
-            'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}, {1.0: 'Male', 0.0: 'Female'}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultwhitemale')
-
-    def load_raw_data(self):
-        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
-
-
-class AdultDatasetMaleExperiment(BinaryClassificationExperiment):
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'income-per-year'
-        positive_label = '>50K'
-        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
-        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
-                                       'native-country']
-        attributes_to_drop_names = ['fnlwgt', 'race']
-
-        protected_attribute_names = ['sex']
-        privileged_classes = [['Male']]
-
-        privileged_groups = [{'sex': 1}]
-        unprivileged_groups = [{'sex': 0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
-            'protected_attribute_maps': [{1.0: 'Male', 0.0: 'Female'}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultmale')
-
-    def load_raw_data(self):
-        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
-
-
-class AdultDatasetWhiteExperiment(BinaryClassificationExperiment):
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'income-per-year'
-        positive_label = '>50K'
-        numeric_attribute_names = ['capital-gain', 'capital-loss', 'age', 'hours-per-week']
-        categorical_attribute_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship',
-                                       'native-country']
-        attributes_to_drop_names = ['fnlwgt', 'sex']
-
-        protected_attribute_names = ['race']
-        privileged_classes = [['White']]
-
-        privileged_groups = [{'race': 1}]
-        unprivileged_groups = [{'race': 0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: '>50K', 0.0: '<=50K'}],
-            'protected_attribute_maps': [{1.0: 'White', 0.0: 'Non-white'}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'adultwhite')
-
-    def load_raw_data(self):
-        return pd.read_csv('datasets/raw/adult.csv', na_values='?', sep=',')
-
-
-
-class PropublicaDatasetWhiteExperiment(BinaryClassificationExperiment):
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'two_year_recid'
-        positive_label = 1
-        numeric_attribute_names = ['age', 'decile_score', 'priors_count', 'days_b_screening_arrest', 'decile_score',
-                                   'is_recid']
-        categorical_attribute_names = ['c_charge_degree', 'age_cat', 'score_text']
-        attributes_to_drop_names = ['sex', 'c_jail_in', 'c_jail_out']
-
-        protected_attribute_names = ['race']
-        privileged_classes = [['Caucasian']]
-
-        privileged_groups = [{'race': 1}]
-        unprivileged_groups = [{'race': 0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: 1, 0.0: 0}],
-            'protected_attribute_maps': [{1.0: 'Caucasian', 0.0: 'Non-white'}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'propublicawhite')
-
-    def load_raw_data(self):
-        """The custom pre-processing function is adapted from
-        https://github.com/IBM/AIF360/blob/master/aif360/algorithms/preprocessing/optim_preproc_helpers/data_preproc_functions.py
-        https://github.com/fair-preprocessing/nips2017/blob/master/compas/code/Generate_Compas_Data.ipynb
-        """
-        df = pd.read_csv('datasets/raw/propublica-recidivism.csv', na_values='?', sep=',')
-        df = df[['age', 'c_charge_degree', 'race', 'age_cat', 'score_text',
-                 'sex', 'priors_count', 'days_b_screening_arrest', 'decile_score',
-                 'is_recid', 'two_year_recid', 'c_jail_in', 'c_jail_out']]
-        ix = df['days_b_screening_arrest'] <= 100
-        ix = (df['days_b_screening_arrest'] >= -100) & ix
-        ix = (df['is_recid'] != -1) & ix
-        ix = (df['c_charge_degree'] != "O") & ix
-        ix = (df['score_text'] != 'N/A') & ix
-        df = df.loc[ix, :]
-        df['length_of_stay'] = (pd.to_datetime(df['c_jail_out']) - pd.to_datetime(df['c_jail_in'])).apply(
-            lambda x: x.days)
-        return df
-
-
-class GermanCreditDatasetSexExperiment(BinaryClassificationExperiment):
-
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'credit'
-        positive_label = 1
-        numeric_attribute_names = ['month', 'credit_amount', 'residence_since', 'age', 'number_of_credits',
-                                   'people_liable_for']
-        categorical_attribute_names = ['credit_history', 'savings', 'employment']
-        attributes_to_drop_names = ['personal_status', 'status', 'purpose', 'investment_as_income_percentage',
-                                    'other_debtors', 'property', 'installment_plans', 'housing', 'skill_level',
-                                    'telephone', 'foreign_worker']
-
-        protected_attribute_names = ['sex']
-        privileged_classes = [[1.0]]
-
-        privileged_groups = [{'sex': 1.0}]
-        unprivileged_groups = [{'sex': 0.0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: 1, 0.0: 0}],
-            'protected_attribute_maps': [{1.0: 'male', 0.0: 'female'}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'germancreditsex')
-
-    def load_raw_data(self):
-        df = pd.read_csv('datasets/raw/german.csv', na_values='?', sep=',')
-
-        def group_credit_hist(x):
-            if x in ['A30', 'A31', 'A32']:
-                return 'None/Paid'
-            elif x == 'A33':
-                return 'Delay'
-            elif x == 'A34':
-                return 'Other'
-            else:
-                return 'NA'
-
-        def group_employ(x):
-            if x == 'A71':
-                return 'Unemployed'
-            elif x in ['A72', 'A73']:
-                return '1-4 years'
-            elif x in ['A74', 'A75']:
-                return '4+ years'
-            else:
-                return 'NA'
-
-        def group_savings(x):
-            if x in ['A61', 'A62']:
-                return '<500'
-            elif x in ['A63', 'A64']:
-                return '500+'
-            elif x == 'A65':
-                return 'Unknown/None'
-            else:
-                return 'NA'
-
-        def group_status(x):
-            if x in ['A11', 'A12']:
-                return '<200'
-            elif x in ['A13']:
-                return '200+'
-            elif x == 'A14':
-                return 'None'
-            else:
-                return 'NA'
-
-        status_map = {'A91': 1.0, 'A93': 1.0, 'A94': 1.0, 'A92': 0.0, 'A95': 0.0}
-        df['sex'] = df['personal_status'].replace(status_map)
-        #         group credit history, savings, and employment
-        df['credit_history'] = df['credit_history'].apply(lambda x: group_credit_hist(x))
-        df['savings'] = df['savings'].apply(lambda x: group_savings(x))
-        df['employment'] = df['employment'].apply(lambda x: group_employ(x))
-        df['age'] = df['age'].apply(lambda x: np.float(x >= 25))
-        df['status'] = df['status'].apply(lambda x: group_status(x))
-        return df
-   
-class RicciRaceExperiment(BinaryClassificationExperiment):
-    '''
-    Check for fairness based on race (white vs minority i.e Black and Hispanic) while predicting if a candidate will pass i.e obtain total 
-    marks greater than or equal to 70.0  
-    '''
-
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        
-        label_name = 'combine'
-        positive_label = 1
-        numeric_attribute_names = ['oral', 'written']
-        categorical_attribute_names = ['position']
-        attributes_to_drop_names = []
-                
-        protected_attribute_names = ['race']
-        privileged_classes = [[1.0]]
-        
-        privileged_groups = [{'race': 1.0}]
-        unprivileged_groups = [{'race': 0.0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: 1, 0.0: 0}],
-            'protected_attribute_maps': [{1.0: 'W', 0.0: 'NW'}]
-        }
-        
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names,attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'riccirace')
-
-    def load_raw_data(self):
-        df = pd.read_csv('datasets/raw/ricci.txt', na_values='?', sep=',')
-        df.columns = map(str.lower, df.columns)
-
-        def group_race_minority(x):
-            if x in ['B', 'H', 'B']:
-                return 'NW'
-            else:
-                return 'W'
-
-        post_map = {'Captain': 0.0, 'Lieutenant': 1.0}
-        df['position'] = df['position'].replace(post_map)
-        
-        #group minorities i.e Black and Hispanic are combined to 'NW'(non white)
-        df['race'] = df['race'].apply(lambda x: group_race_minority(x))
-        df['combine'] = df['combine'].apply(lambda x: int(x >= 70))
-
-        return df
-
-class GiveMeSomeCreditExperiment(BinaryClassificationExperiment):
-    '''
-    Fairness intervention for the Age attribute (priviledge for age>=25) while predicting if a person will experience 90 days past due delinquency or worse. 
-    '''
-    def __init__(self, fixed_random_seed, train_data_sampler, missing_value_handler, numeric_attribute_scaler,
-                 learners, pre_processors, post_processors):
-
-        test_set_ratio = 0.2
-        validation_set_ratio = 0.1
-        label_name = 'SeriousDlqin2yrs'
-        positive_label = 1
-
-        numeric_attribute_names = ['RevolvingUtilizationOfUnsecuredLines','age','NumberOfTime30-59DaysPastDueNotWorse',
-                                  'DebtRatio','MonthlyIncome','NumberOfOpenCreditLinesAndLoans','NumberOfTimes90DaysLate',
-                                  'NumberRealEstateLoansOrLines','NumberOfTime60-89DaysPastDueNotWorse','NumberOfDependents']
-        categorical_attribute_names = []
-        attributes_to_drop_names = []
-
-        protected_attribute_names = ['age']
-        privileged_classes = [[1.0]]
-
-        privileged_groups = [{'age': 1}]
-        unprivileged_groups = [{'age': 0}]
-
-        dataset_metadata = {
-            'label_maps': [{1.0: 1, 0.0: 0}]
-        }
-
-        super().__init__(fixed_random_seed, test_set_ratio, validation_set_ratio, label_name, positive_label,
-                         numeric_attribute_names, categorical_attribute_names, attributes_to_drop_names,
-                         train_data_sampler, missing_value_handler, numeric_attribute_scaler, learners, pre_processors,
-                         post_processors, protected_attribute_names, privileged_classes, privileged_groups,
-                         unprivileged_groups, dataset_metadata, 'givecredit')
-
-    def load_raw_data(self):
-        df = pd.read_csv('datasets/raw/givemesomecredit.csv', na_values='?', sep=',',index_col=False)
-        df['age'] = df['age'].apply(lambda x: int(x >= 25))
-        return df
-        
diff --git a/fp/experiments.py b/fp/experiments.py
index bbc5d5e..cb5f33c 100644
--- a/fp/experiments.py
+++ b/fp/experiments.py
@@ -10,7 +10,6 @@
 from sklearn.base import clone
 from sklearn.metrics import roc_auc_score
 from sklearn.model_selection import train_test_split
-from fp.utils import filter_optimal_results_skyline_order, filter_optimal_results_skyline_formula
 
 
 class BinaryClassificationExperiment:
@@ -36,8 +35,7 @@ def __init__(self,
                  privileged_groups,
                  unprivileged_groups,
                  dataset_metadata,
-                 dataset_name,
-                 optimal_validation_strategy):
+                 dataset_name):
 
         self.fixed_random_seed = fixed_random_seed
         self.test_set_ratio = test_set_ratio
@@ -61,21 +59,20 @@ def __init__(self,
         self.dataset_name = dataset_name
         self.log_path = 'logs/'
         self.exec_timestamp = self.generate_timestamp()
-        self.optimal_validation_strategy = optimal_validation_strategy
 
 
 
     # --- Helper Methods Begin ------------------------------------------------
 
 
-    def unique_file_name(self, pre_processor, post_processor, learner):
-        return '{0}__{1}__{2}__{3}__{4}__{5}__{6}'.format(self.dataset_name,
-                                                   pre_processor.name(),
-                                                   post_processor.name(),
+    def unique_file_name(self, learner, pre_processor, post_processor):
+        return '{}__{}__{}__{}__{}__{}__{}'.format(self.dataset_name,
                                                    learner.name(),
                                                    self.missing_value_handler.name(),
                                                    self.train_data_sampler.name(),
-                                                   self.numeric_attribute_scaler.name())
+                                                   self.numeric_attribute_scaler.name(),
+                                                   pre_processor.name(),
+                                                   post_processor.name())
 
 
     def generate_file_path(self, file_name=''):
@@ -134,6 +131,7 @@ def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processo
 
         feature_names_in_train_but_not_in_current = set(train_feature_names).difference(
             set(current_feature_names))
+
         print("Injecting zero columns for features not present", feature_names_in_train_but_not_in_current)
 
         validation_data_df, _ = adjusted_annotated_data.convert_to_dataframe()
@@ -149,12 +147,14 @@ def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processo
         if learner.needs_annotated_data_for_prediction():
             adjusted_annotated__data_with_predictions = model.predict(adjusted_annotated_data)
         else:
-            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features).reshape(-1,1)
+            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features)
+
             try:
                 class_probs = model.predict_proba(adjusted_annotated_data.features)
                 adjusted_annotated__data_with_predictions.scores = class_probs[:, 0]
             except AttributeError:
                 print("WARNING: MODEL CANNOT ASSIGN CLASS PROBABILITIES")
+
         return adjusted_annotated_data, adjusted_annotated__data_with_predictions
 
 
@@ -235,11 +235,10 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
         post_processor : fairprep pre-processor abstraction from 
             aif360.algorithms.post_processors
         """
-
+        
         adjusted_annotated_train_data = self.preprocess_data(pre_processor, annotated_train_data)
-
         model = self.learn_classifier(learner, adjusted_annotated_train_data, self.fixed_random_seed)
-        
+
         adjusted_annotated_train_data_with_predictions = adjusted_annotated_train_data.copy()
 
         if learner.needs_annotated_data_for_prediction():
@@ -261,12 +260,13 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
             adjusted_annotated_test_data_with_predictions)
 
         results_file_name = '../{}{}-{}.csv'.format(
-            self.generate_file_path(), self.unique_file_name(pre_processor, post_processor, learner), self.fixed_random_seed)
+            self.generate_file_path(), self.unique_file_name(pre_processor, learner, post_processor), self.fixed_random_seed)
         results_file_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), results_file_name)
         
         results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
         if not os.path.exists(results_dir_name):
             os.makedirs(results_dir_name)
+        
         results_file = []
         
         results_file = self.log_metrics(results_file, model, adjusted_annotated_validation_data,
@@ -280,70 +280,45 @@ def run_single_exp(self, annotated_train_data, validation_data, test_data, scale
         results_file.to_csv(results_file_path, index=False)
 
 
-
     def filter_optimal_results(self):
         """Identifies the experiment(s) with the highest accuracy as optimal 
             result. Keeps the test metrics just for the experiment(s) with the
             optimal result.
         """
-
         results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
         results_dir = os.listdir(Path(results_dir_name))
+        accuracies = dict()
+        max_accuracy = 0
 
-        ##### SKYLINE FORMULA IMPLEMENTATION
-
-        privileged_metric_names = ['num_true_positives', 'num_false_positives', 'num_false_negatives',
-                                   'num_true_negatives', 'num_generalized_true_positives',
-                                   'num_generalized_false_positives', 'num_generalized_false_negatives',
-                                   'num_generalized_true_negatives', 'true_positive_rate', 'false_positive_rate',
-                                   'false_negative_rate', 'true_negative_rate', 'generalized_true_positive_rate',
-                                   'generalized_false_positive_rate', 'generalized_false_negative_rate',
-                                   'generalized_true_negative_rate', 'positive_predictive_value',
-                                   'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value',
-                                   'accuracy', 'error_rate', 'num_pred_positives', 'num_pred_negatives',
-                                   'selection_rate']
-         
-        dictionary = {}
-        filenames = list()
+        # Fetching the accuracy from the row('val', 'None', 'accuracy') of all the experiment results
         for result_filename in results_dir:
             file_path = os.path.join(results_dir_name, result_filename)
             result_df = pd.read_csv(file_path)
             result_df.fillna(value='', inplace=True)
-            for privileged_metric in privileged_metric_names:
-                if privileged_metric not in dictionary:
-                    dictionary[privileged_metric] = list()
-                p_metric = (result_df.loc[(result_df['Split'] == 'val') & 
+            accuracy = (result_df.loc[(result_df['Split'] == 'val') & 
                                       (result_df['PrivilegedStatus'] == '') & 
-                                      (result_df['MetricName'] == privileged_metric), 'MetricValue'].values[0])
-                dictionary[privileged_metric].append(p_metric)
-            filenames.append(result_filename)
-
-        privileged_metric_values = pd.DataFrame(dictionary)
-        privileged_metric_values['filenames'] = filenames
-
-        if isinstance(self.optimal_validation_strategy, dict):
-            skyline_result = filter_optimal_results_skyline_formula(privileged_metric_values, self.optimal_validation_strategy)
-        else:
-            skyline_result = filter_optimal_results_skyline_order(privileged_metric_values, self.optimal_validation_strategy)
+                                      (result_df['MetricName'] == 'accuracy'), 'MetricValue'].values[0])
+            accuracies[result_filename] = accuracy
+            if accuracy > max_accuracy:
+                max_accuracy = accuracy
 
+        # List of non optimal and optimal filenames and accuracy
         non_optimal_filenames = list()
         optimal_filenames = list()
-        filenames_list = privileged_metric_values['filenames'].tolist()
-        for file_name in filenames_list:
-            if file_name == skyline_result[-1]:
-                optimal_filenames.append(file_name)
+        for filename, accuracy in accuracies.items():
+            if accuracy != max_accuracy:
+                non_optimal_filenames.append(filename)
             else:
-                non_optimal_filenames.append(file_name)
+                optimal_filenames.append(filename)
 
         # Removing the test results from the non optimal experiment results      
-        '''
         for file_name in non_optimal_filenames:
-            file_path = os.path.join(results_dir_name, file_name)
+            file_path = os.path.join(results_dir_name, result_filename)
             result_df = pd.read_csv(file_path)
             result_df = result_df[(result_df['Split'] != 'test')]
             os.remove(file_path)
             result_df.to_csv(file_path, index=False, header=False)
-        '''
+
         # Renaming the optimal experiment results file (or files if tie) 
         for file_name in optimal_filenames:
             file_path = os.path.join(results_dir_name, file_name)
@@ -366,7 +341,6 @@ def run(self):
                                                                     self.validation_set_ratio,
                                                                     random_state=self.fixed_random_seed)
 
-        #Just returns complete data
         train_data = self.train_data_sampler.sample(all_train_data)
 
         second_split_ratio = self.test_set_ratio / (self.test_set_ratio + self.validation_set_ratio)
@@ -374,8 +348,6 @@ def run(self):
         validation_data, test_data = train_test_split(test_and_validation_data, test_size=second_split_ratio,
                                                       random_state=self.fixed_random_seed)
 
-
-        #fit just passes and hadle_missing does drop_na operation
         self.missing_value_handler.fit(train_data)
         filtered_train_data = self.missing_value_handler.handle_missing(train_data)
 
@@ -402,8 +374,10 @@ def run(self):
             features_to_drop=self.attributes_to_drop_names,
             metadata=self.dataset_metadata
         )
+
         for pre_processor in self.pre_processors:
             for learner in self.learners:
                 for post_processor in self.post_processors:
-                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, pre_processor, learner, post_processor)
+                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, 
+                                        pre_processor, learner, post_processor)
         self.filter_optimal_results()
diff --git a/fp/experiments_old.py b/fp/experiments_old.py
deleted file mode 100644
index cb5f33c..0000000
--- a/fp/experiments_old.py
+++ /dev/null
@@ -1,383 +0,0 @@
-import os
-import numpy as np
-import pandas as pd
-
-from time import time
-from pathlib import Path
-from datetime import datetime
-from aif360.datasets import StandardDataset
-from aif360.metrics import ClassificationMetric
-from sklearn.base import clone
-from sklearn.metrics import roc_auc_score
-from sklearn.model_selection import train_test_split
-
-
-class BinaryClassificationExperiment:
-
-
-    def __init__(self,
-                 fixed_random_seed,
-                 test_set_ratio,
-                 validation_set_ratio,
-                 label_name,
-                 positive_label,
-                 numeric_attribute_names,
-                 categorical_attribute_names,
-                 attributes_to_drop_names,
-                 train_data_sampler,
-                 missing_value_handler,
-                 numeric_attribute_scaler,
-                 learners,
-                 pre_processors,
-                 post_processors,
-                 protected_attribute_names,
-                 privileged_classes,
-                 privileged_groups,
-                 unprivileged_groups,
-                 dataset_metadata,
-                 dataset_name):
-
-        self.fixed_random_seed = fixed_random_seed
-        self.test_set_ratio = test_set_ratio
-        self.validation_set_ratio = validation_set_ratio
-        self.label_name = label_name
-        self.positive_label = positive_label
-        self.numeric_attribute_names = numeric_attribute_names
-        self.categorical_attribute_names = categorical_attribute_names
-        self.attributes_to_drop_names = attributes_to_drop_names
-        self.train_data_sampler = train_data_sampler
-        self.missing_value_handler = missing_value_handler
-        self.numeric_attribute_scaler = numeric_attribute_scaler
-        self.learners = learners
-        self.pre_processors = pre_processors
-        self.post_processors = post_processors
-        self.protected_attribute_names = protected_attribute_names
-        self.privileged_classes = privileged_classes
-        self.privileged_groups = privileged_groups
-        self.unprivileged_groups = unprivileged_groups
-        self.dataset_metadata = dataset_metadata
-        self.dataset_name = dataset_name
-        self.log_path = 'logs/'
-        self.exec_timestamp = self.generate_timestamp()
-
-
-
-    # --- Helper Methods Begin ------------------------------------------------
-
-
-    def unique_file_name(self, learner, pre_processor, post_processor):
-        return '{}__{}__{}__{}__{}__{}__{}'.format(self.dataset_name,
-                                                   learner.name(),
-                                                   self.missing_value_handler.name(),
-                                                   self.train_data_sampler.name(),
-                                                   self.numeric_attribute_scaler.name(),
-                                                   pre_processor.name(),
-                                                   post_processor.name())
-
-
-    def generate_file_path(self, file_name=''):
-        dir_name = '{}_{}/'.format(self.exec_timestamp, self.dataset_name)
-        return self.log_path + dir_name + file_name
-
-
-    def generate_timestamp(self):
-        return datetime.fromtimestamp(time()).strftime('%Y-%m-%d_%H-%M-%S-%f')[:-3]
-
-
-    def load_raw_data(self):
-        raise NotImplementedError
-
-
-    def learn_classifier(self, learner, annotated_train_data, fixed_random_seed):
-        return learner.fit_model(annotated_train_data, fixed_random_seed)
-
-
-    def preprocess_data(self, pre_processor, annotated_dataset):
-        return pre_processor.pre_process(annotated_dataset, self.privileged_groups, self.unprivileged_groups)
-
-
-    def post_process_predictions(self, post_processor, validation_dataset, validation_dataset_with_predictions,
-                                 testset_with_predictions):
-        return post_processor.post_process(validation_dataset, validation_dataset_with_predictions,
-                                                testset_with_predictions, self.fixed_random_seed,
-                                                self.privileged_groups, self.unprivileged_groups)
-
-
-    def apply_model(self, data, scalers, adjusted_annotated_train_data, pre_processor, learner, model):
-        filtered_data = self.missing_value_handler.handle_missing(data)
-        print(self.missing_value_handler.name(), 'removed', len(data) - len(filtered_data),
-              'instances from validation data')
-
-        for numerical_attribute, scaler in scalers.items():
-            numerical_attribute_data = np.array(filtered_data[numerical_attribute]).reshape(-1, 1)
-            scaled_numerical_attribute_data = scaler.transform(numerical_attribute_data)
-            filtered_data.loc[:, numerical_attribute] = scaled_numerical_attribute_data
-
-        annotated_data = StandardDataset(
-            df=filtered_data,
-            label_name=self.label_name,
-            favorable_classes=[self.positive_label],
-            protected_attribute_names=self.protected_attribute_names,
-            privileged_classes=self.privileged_classes,
-            categorical_features=self.categorical_attribute_names,
-            features_to_drop=self.attributes_to_drop_names,
-            metadata=self.dataset_metadata
-        )
-
-        adjusted_annotated_data = self.preprocess_data(pre_processor, annotated_data)
-
-        train_feature_names = adjusted_annotated_train_data.feature_names
-        current_feature_names = adjusted_annotated_data.feature_names
-
-        feature_names_in_train_but_not_in_current = set(train_feature_names).difference(
-            set(current_feature_names))
-
-        print("Injecting zero columns for features not present", feature_names_in_train_but_not_in_current)
-
-        validation_data_df, _ = adjusted_annotated_data.convert_to_dataframe()
-
-        for feature_name in feature_names_in_train_but_not_in_current:
-            validation_data_df.loc[:, feature_name] = 0.0
-
-        adjusted_annotated_data.feature_names = train_feature_names
-        adjusted_annotated_data.features = validation_data_df[train_feature_names].values.copy()
-
-        adjusted_annotated__data_with_predictions = adjusted_annotated_data.copy()
-
-        if learner.needs_annotated_data_for_prediction():
-            adjusted_annotated__data_with_predictions = model.predict(adjusted_annotated_data)
-        else:
-            adjusted_annotated__data_with_predictions.labels = model.predict(adjusted_annotated_data.features)
-
-            try:
-                class_probs = model.predict_proba(adjusted_annotated_data.features)
-                adjusted_annotated__data_with_predictions.scores = class_probs[:, 0]
-            except AttributeError:
-                print("WARNING: MODEL CANNOT ASSIGN CLASS PROBABILITIES")
-
-        return adjusted_annotated_data, adjusted_annotated__data_with_predictions
-
-
-    def log_metrics(self, results_file, model, annotated_data, annotated_data_with_predictions, prefix):
-        metric = ClassificationMetric(annotated_data, annotated_data_with_predictions,
-                                      unprivileged_groups=self.unprivileged_groups,
-                                      privileged_groups=self.privileged_groups)
-
-        privileged_metric_names = ['num_true_positives', 'num_false_positives', 'num_false_negatives',
-                                   'num_true_negatives', 'num_generalized_true_positives',
-                                   'num_generalized_false_positives', 'num_generalized_false_negatives',
-                                   'num_generalized_true_negatives', 'true_positive_rate', 'false_positive_rate',
-                                   'false_negative_rate', 'true_negative_rate', 'generalized_true_positive_rate',
-                                   'generalized_false_positive_rate', 'generalized_false_negative_rate',
-                                   'generalized_true_negative_rate', 'positive_predictive_value',
-                                   'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value',
-                                   'accuracy', 'error_rate', 'num_pred_positives', 'num_pred_negatives',
-                                   'selection_rate']
-
-        for maybe_privileged in [None, True, False]:
-            for metric_name in privileged_metric_names:
-                metric_function = getattr(metric, metric_name)
-                metric_value = metric_function(privileged=maybe_privileged)
-                results_file.append([prefix, maybe_privileged, metric_name, metric_value])
-
-        if hasattr(model, 'predict_proba'):
-            auc = roc_auc_score(annotated_data.labels, model.predict_proba(annotated_data.features)[:, 1])
-        else:
-            auc = None
-
-        results_file.append([prefix, '', 'roc_auc', auc])
-
-        global_metric_names = ['true_positive_rate_difference', 'false_positive_rate_difference',
-                               'false_negative_rate_difference', 'false_omission_rate_difference',
-                               'false_discovery_rate_difference', 'false_positive_rate_ratio',
-                               'false_negative_rate_ratio', 'false_omission_rate_ratio',
-                               'false_discovery_rate_ratio', 'average_odds_difference', 'average_abs_odds_difference',
-                               'error_rate_difference', 'error_rate_ratio', 'disparate_impact',
-                               'statistical_parity_difference', 'generalized_entropy_index',
-                               'between_all_groups_generalized_entropy_index',
-                               'between_group_generalized_entropy_index', 'theil_index', 'coefficient_of_variation',
-                               'between_group_theil_index', 'between_group_coefficient_of_variation',
-                               'between_all_groups_theil_index', 'between_all_groups_coefficient_of_variation']
-
-        for metric_name in global_metric_names:
-            metric_function = getattr(metric, metric_name)
-            metric_value = metric_function()
-            results_file.append([prefix, '', metric_name, metric_value])
-
-        return results_file
-
-
-    # --- Helper Methods End --------------------------------------------------
-
-    
-    def run_single_exp(self, annotated_train_data, validation_data, test_data, scalers, pre_processor,
-                       learner, post_processor):
-        """Executes a single instance of experiment out of all the possible 
-        experiments from the given parameters.
-        
-        Parameters:
-        -----------
-        annotated_train_data : annotated aif360.datasets.StandardDataset of 
-            train data
-
-        validation_data : pandas dataframe of validation data
-
-        test_data : pandas dataframe of test data
-
-        scalers : dictionary with (key='feature name', value='type of scaler')
-
-        pre_processor : fairprep pre-processor abstraction from 
-            aif360.algorithms.pre_processors
-    
-        learner : fairprep learner abstraction from sci-kit learn or 
-            aif360.algorithms.inprocessing
-    
-        post_processor : fairprep pre-processor abstraction from 
-            aif360.algorithms.post_processors
-        """
-        
-        adjusted_annotated_train_data = self.preprocess_data(pre_processor, annotated_train_data)
-        model = self.learn_classifier(learner, adjusted_annotated_train_data, self.fixed_random_seed)
-
-        adjusted_annotated_train_data_with_predictions = adjusted_annotated_train_data.copy()
-
-        if learner.needs_annotated_data_for_prediction():
-            adjusted_annotated_train_data_with_predictions = model.predict(
-                adjusted_annotated_train_data_with_predictions)
-        else:
-            adjusted_annotated_train_data_with_predictions.labels = model.predict(
-                adjusted_annotated_train_data_with_predictions.features)
-
-        adjusted_annotated_validation_data, adjusted_annotated_validation_data_with_predictions = \
-            self.apply_model(validation_data, scalers, adjusted_annotated_train_data, pre_processor, learner, model)
-
-        adjusted_annotated_test_data, adjusted_annotated_test_data_with_predictions = \
-            self.apply_model(test_data, scalers, adjusted_annotated_train_data, pre_processor, learner, model)
-
-        adjusted_annotated_test_data_with_predictions = self.post_process_predictions(post_processor,
-            adjusted_annotated_validation_data,
-            adjusted_annotated_validation_data_with_predictions,
-            adjusted_annotated_test_data_with_predictions)
-
-        results_file_name = '../{}{}-{}.csv'.format(
-            self.generate_file_path(), self.unique_file_name(pre_processor, learner, post_processor), self.fixed_random_seed)
-        results_file_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), results_file_name)
-        
-        results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
-        if not os.path.exists(results_dir_name):
-            os.makedirs(results_dir_name)
-        
-        results_file = []
-        
-        results_file = self.log_metrics(results_file, model, adjusted_annotated_validation_data,
-                                        adjusted_annotated_validation_data_with_predictions, 'val')
-        results_file = self.log_metrics(results_file, model, adjusted_annotated_train_data,
-                                        adjusted_annotated_train_data_with_predictions, 'train')
-        results_file = self.log_metrics(results_file, model, adjusted_annotated_test_data, 
-                                        adjusted_annotated_test_data_with_predictions, 'test')
-        
-        results_file = pd.DataFrame(results_file, columns=['Split', 'PrivilegedStatus', 'MetricName', 'MetricValue'])
-        results_file.to_csv(results_file_path, index=False)
-
-
-    def filter_optimal_results(self):
-        """Identifies the experiment(s) with the highest accuracy as optimal 
-            result. Keeps the test metrics just for the experiment(s) with the
-            optimal result.
-        """
-        results_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../{}'.format(self.generate_file_path()))
-        results_dir = os.listdir(Path(results_dir_name))
-        accuracies = dict()
-        max_accuracy = 0
-
-        # Fetching the accuracy from the row('val', 'None', 'accuracy') of all the experiment results
-        for result_filename in results_dir:
-            file_path = os.path.join(results_dir_name, result_filename)
-            result_df = pd.read_csv(file_path)
-            result_df.fillna(value='', inplace=True)
-            accuracy = (result_df.loc[(result_df['Split'] == 'val') & 
-                                      (result_df['PrivilegedStatus'] == '') & 
-                                      (result_df['MetricName'] == 'accuracy'), 'MetricValue'].values[0])
-            accuracies[result_filename] = accuracy
-            if accuracy > max_accuracy:
-                max_accuracy = accuracy
-
-        # List of non optimal and optimal filenames and accuracy
-        non_optimal_filenames = list()
-        optimal_filenames = list()
-        for filename, accuracy in accuracies.items():
-            if accuracy != max_accuracy:
-                non_optimal_filenames.append(filename)
-            else:
-                optimal_filenames.append(filename)
-
-        # Removing the test results from the non optimal experiment results      
-        for file_name in non_optimal_filenames:
-            file_path = os.path.join(results_dir_name, result_filename)
-            result_df = pd.read_csv(file_path)
-            result_df = result_df[(result_df['Split'] != 'test')]
-            os.remove(file_path)
-            result_df.to_csv(file_path, index=False, header=False)
-
-        # Renaming the optimal experiment results file (or files if tie) 
-        for file_name in optimal_filenames:
-            file_path = os.path.join(results_dir_name, file_name)
-            optimal_file_name = '{}{}'.format(file_name[:-4], '__OPTIMAL.csv')
-            optimal_file_path = os.path.join(results_dir_name, optimal_file_name)
-            os.rename(file_path, optimal_file_path)
-
-
-    def run(self):
-        """Executes all the possible experiments from the combination of  given
-            learners, pre-processors and post-processors.
-            
-            No. of experiments = (#learners * #preprocessors * #postprocessors)
-        """
-        np.random.seed(self.fixed_random_seed)
-
-        data = self.load_raw_data()
-
-        all_train_data, test_and_validation_data = train_test_split(data, test_size=self.test_set_ratio +
-                                                                    self.validation_set_ratio,
-                                                                    random_state=self.fixed_random_seed)
-
-        train_data = self.train_data_sampler.sample(all_train_data)
-
-        second_split_ratio = self.test_set_ratio / (self.test_set_ratio + self.validation_set_ratio)
-
-        validation_data, test_data = train_test_split(test_and_validation_data, test_size=second_split_ratio,
-                                                      random_state=self.fixed_random_seed)
-
-        self.missing_value_handler.fit(train_data)
-        filtered_train_data = self.missing_value_handler.handle_missing(train_data)
-
-        print(self.missing_value_handler.name(), 'removed', len(train_data) - len(filtered_train_data),
-              'instances from training data')
-
-        scalers = {}
-
-        for numerical_attribute in self.numeric_attribute_names:
-            numerical_attribute_data = np.array(filtered_train_data[numerical_attribute]).reshape(-1, 1)
-            scaler = clone(self.numeric_attribute_scaler).fit(numerical_attribute_data)
-            scaled_numerical_attribute_data = scaler.transform(numerical_attribute_data)
-
-            filtered_train_data.loc[:, numerical_attribute] = scaled_numerical_attribute_data
-            scalers[numerical_attribute] = scaler
-
-        annotated_train_data = StandardDataset(
-            df=filtered_train_data,
-            label_name=self.label_name,
-            favorable_classes=[self.positive_label],
-            protected_attribute_names=self.protected_attribute_names,
-            privileged_classes=self.privileged_classes,
-            categorical_features=self.categorical_attribute_names,
-            features_to_drop=self.attributes_to_drop_names,
-            metadata=self.dataset_metadata
-        )
-
-        for pre_processor in self.pre_processors:
-            for learner in self.learners:
-                for post_processor in self.post_processors:
-                    self.run_single_exp(annotated_train_data, validation_data, test_data, scalers, 
-                                        pre_processor, learner, post_processor)
-        self.filter_optimal_results()

From 2752ac234b7c109342a985788f978299202dae08 Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Wed, 15 Jul 2020 10:22:43 -0400
Subject: [PATCH 5/7] clean up wrong files

---
 fp/utils.py | 43 -------------------------------------------
 1 file changed, 43 deletions(-)
 delete mode 100644 fp/utils.py

diff --git a/fp/utils.py b/fp/utils.py
deleted file mode 100644
index 6d9113e..0000000
--- a/fp/utils.py
+++ /dev/null
@@ -1,43 +0,0 @@
-import pandas as pd
-
-def filter_optimal_results_skyline_order(_df, _order_list):
-    _df['selection_rate'] = abs(1 - _df['selection_rate'])
-    higher_is_better = ['num_true_positives', 'num_true_negatives', 'num_generalized_true_positives',
-                        'num_generalized_true_negatives', 'true_positive_rate', 'true_negative_rate',
-                        'generalized_true_positive_rate', 'generalized_true_negative_rate', 'positive_predictive_value',
-                        'accuracy', 'num_pred_positives']
-    lower_is_better = ['selection_rate', 'num_false_positives', 'num_false_negatives',
-                       'num_generalized_false_positives', 'num_generalized_false_negatives', 'false_positive_rate',
-                       'false_negative_rate', 'generalized_false_positive_rate', 'generalized_false_negative_rate',
-                       'false_discovery_rate', 'false_omission_rate', 'negative_predictive_value', 'error_rate',
-                       'num_pred_negatives']
-    order = []
-    for item in _order_list:
-        if item in higher_is_better:
-            order.append(False)
-        else:
-            order.append(True)
-    _df = _df.sort_values(_order_list, ascending=order)
-
-    return _df.values[0]
-
-
-def filter_optimal_results_skyline_formula(_df, _formula):
-    df = pd.DataFrame()
-    for key in _formula:
-        df["norm_" + key] = (_df[key] - _df[key].min()) / (_df[key].max() - _df[key].min())
-
-    df_temp = list(_formula.values())
-    keys = list(_formula.keys())
-    for col in range(len(keys)):
-        keys[col] = "norm_" + keys[col]
-
-    # Multiplying with the multiplier to perform sorting operation
-    df['norm_avg'] = df[keys].multiply(df_temp).sum(axis=1)
-    frames = [_df, df]
-    df_fin = pd.concat(frames, axis=1)
-
-    df_fin = df_fin.sort_values(by='norm_avg', ascending=False)
-    cols = [c for c in df_fin.columns if c[:4] != 'norm']
-    df_fin = df_fin[cols]
-    return df_fin.values[0]
\ No newline at end of file

From dceaab8324116f04500993737c940bf78b448deb Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Thu, 30 Jul 2020 18:21:06 -0400
Subject: [PATCH 6/7] update for inputs and flexible order

---
 pipeline/fairprep.py                          | 666 +++++++-----------
 .../RS_NS-test.csv                            | 201 ++++++
 .../RS_NS-train.csv                           | 501 +++++++++++++
 .../RS_NS-val.csv                             | 301 ++++++++
 .../RS_NS_NI-test.csv                         | 201 ++++++
 .../RS_NS_NI-train.csv                        | 501 +++++++++++++
 .../RS_NS_NI-val.csv                          | 301 ++++++++
 .../RS_NS_NI_MS-test.csv                      | 201 ++++++
 .../RS_NS_NI_MS-train.csv                     | 501 +++++++++++++
 .../RS_NS_NI_MS-val.csv                       | 301 ++++++++
 .../RS_NS_NI_MS_NB-test.csv                   | 201 ++++++
 .../RS_NS_NI_MS_NB-train.csv                  | 501 +++++++++++++
 .../RS_NS_NI_MS_NB-val.csv                    | 301 ++++++++
 .../RS_NS_NI_MS_NB_HE-test.csv                | 201 ++++++
 .../RS_NS_NI_MS_NB_HE-train.csv               | 501 +++++++++++++
 .../RS_NS_NI_MS_NB_HE-val.csv                 | 301 ++++++++
 .../RS_NS_NI_MS_NB_HE_ME-test.csv             | 201 ++++++
 .../RS_NS_NI_MS_NB_HE_ME-train.csv            | 501 +++++++++++++
 .../RS_NS_NI_MS_NB_HE_ME-val.csv              | 301 ++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW-test.csv          | 201 ++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW-train.csv         | 501 +++++++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW-val.csv           | 301 ++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR-test.csv      | 201 ++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR-train.csv     | 501 +++++++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR-val.csv       | 301 ++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-test.csv   | 201 ++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-train.csv  | 501 +++++++++++++
 .../RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-val.csv    | 301 ++++++++
 .../german_AIF_test.json                      |   1 +
 pipeline/model/classifiers.py                 | 239 ++++++-
 pipeline/model/fair_classifiers.py            | 172 ++++-
 pipeline/model/inprocessor.py                 |  74 --
 pipeline/postprocess/fair_postprocessors.py   | 203 +++++-
 pipeline/postprocess/postprocessor.py         |  59 --
 pipeline/preprocess/categorizers.py           | 132 +++-
 pipeline/preprocess/encoders.py               | 170 +++--
 pipeline/preprocess/fair_preprocessors.py     | 166 ++++-
 pipeline/preprocess/filters.py                | 113 +--
 pipeline/preprocess/imputers.py               | 173 ++++-
 pipeline/preprocess/preprocessor.py           |  71 --
 pipeline/preprocess/samplers.py               | 132 +++-
 pipeline/preprocess/scalers.py                | 109 ++-
 pipeline/preprocess/splitters.py              | 180 +++--
 pipeline/step.py                              |  78 +-
 44 files changed, 10693 insertions(+), 1072 deletions(-)
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-test.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-train.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-val.csv
 create mode 100644 pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/german_AIF_test.json
 delete mode 100644 pipeline/model/inprocessor.py
 delete mode 100644 pipeline/postprocess/postprocessor.py
 delete mode 100644 pipeline/preprocess/preprocessor.py

diff --git a/pipeline/fairprep.py b/pipeline/fairprep.py
index 5cc8321..a4dcc61 100644
--- a/pipeline/fairprep.py
+++ b/pipeline/fairprep.py
@@ -2,9 +2,18 @@
     Class to run the pipeline
 
 """
-import pandas as pd
-import numpy as np
+
+
 import os
+import numpy as np
+import pandas as pd
+
+from time import time
+from pathlib import Path
+from datetime import datetime
+from itertools import islice
+
+
 import warnings
 warnings.filterwarnings("ignore")
 from pipeline.preprocess.splitters import *
@@ -18,41 +27,24 @@
 from pipeline.model.fair_classifiers import *
 from pipeline.postprocess.fair_postprocessors import *
 
-# for integrity check of user inputs
-SUPPORT_STEPS = {"Splitter": "Splitter", "Sampler": "Sampler", "Imputer": "Imputer", "Scaler": "Scaler",
-                 "Categorizer": "Categorizer", "Encoder": "Encoder", "SensitiveEncoder": "SensitiveAttEncoder",
-                 "FairPreprocessor": "AIF_", "model": "SK_OPT_", "FairPostprocessor": "AIF_Postprocessing"}
-
-ALL_STEPS = ["RandomSplitter", "BalanceTargetSplitter",
-             "RandomSampler", "BalancePopulationSampler",
-             "DropNAImputer", "ModeImputer", "DatawigImputer",
-             "SK_StandardScaler", "SK_MinMaxScaler",
-             "SK_Discretizer", "SK_Binarizer",
-             "SK_OrdinalEncoder", "SK_OneHotEncoder",
-             "CustomCateAttsEncoder",
-             "AIF_Reweighing", "AIF_DIRemover",
-             "SK_LogisticRegression", "SK_DecisionTree", "OPT_LogisticRegression", "OPT_DecisionTree", "AIF_AdversarialDebiasing",
-             "AIF_EqOddsPostprocessing", "AIF_CalibratedEqOddsPostprocessing"]
+import json
+
 PRINT_SPLIT = "\n===================================================\n"
+# for integrity check of user inputs
+PREREQUISITE_STEPS = {"Splitter": [], "Sampler": [], "Imputer": ["Splitter"], "Scaler": ["Imputer"],
+                    "Categorizer": ["Imputer"], "Encoder": ["Imputer"], "SpecialEncoder": ["Imputer"],
+                    "FairPreprocessor": ["Splitter", "Imputer", "Scaler", "Categorizer", "Encoder", "SpecialEncoder"],
+                    "Model": ["Splitter", "Imputer", "Scaler", "Categorizer", "Encoder", "SensitiveAttEncoder"],
+                    "FairPostprocessor": ["Splitter", "Imputer", "Scaler", "Categorizer", "Encoder", "SpecialEncoder"]}
 
-def init_input_steps(step_tuple, input_df):
-    step_str = step_tuple[0] + '(input_df, '
-    for pi in step_tuple[1:]:
-        if isinstance(pi, str):
-            step_str += "'"+pi + "', "
-        else:
-            step_str += str(pi) + ", "
-    end_idx = step_str.rfind(", ")
-    step_str = step_str[0:end_idx] + step_str[end_idx:].replace(", ", ")")
-    print(step_str)
-    return eval(step_str)
 
 
-class FairnessLabelPipeline():
-    def __init__(self, data_file_name, target_col, target_positive_values, sensitive_atts=[], protected_values={}, sep_flag=None, na_mark=None):
+class FairPipeline():
+    def __init__(self, data_file_name, target_col, target_positive_values, sensitive_atts, value_mapping, sep_flag=None, na_mark=None, verbose=True):
         """
 
         :param data_file_name: str, file name that stores the data.
+        :param *fixed_random_seed: int, the fixed random seed that will be used through the pipeline.
         :param target_col: str, the name of the target variable in above data.
         :param target_positive_values: list of str, each str is the value of the target variable in above data that represents the positive outcome.
         :param sensitive_atts: list, stores the user specified sensitive attributes in above data. Optional.
@@ -60,129 +52,176 @@ def __init__(self, data_file_name, target_col, target_positive_values, sensitive
                                 Key is the name in sensitive_atts, value is a list of str, representing the values of the attribute.
                                  Order should mapping to the order in the sensitive_atts.
         """
-        # print(os.path.realpath(data_file_name))
-        if not os.path.exists(os.path.realpath(data_file_name)):
+
+        # TODO: move integrity check to a separate function
+        self.data_file_name = data_file_name
+        self.target_col = target_col
+        self.target_positive_values = target_positive_values
+        self.sensitive_atts = sensitive_atts
+        self.value_mapping = value_mapping
+        self.sep_flag = sep_flag
+        self.na_mark = na_mark
+        self.verbose = verbose
+
+        self.data_name = None
+        self.raw_data = None
+
+        self.attributes = None
+        self.metadata_desp = None
+        self.num_atts = None
+        self.cate_atts = None
+        self.pred_target = "pred_" + target_col
+
+        self.log_path = 'logs/'
+        self.exec_timestamp = self.generate_timestamp()
+        self.log_dir_name = None
+
+    def validate_input_parameters(self):
+        if not os.path.exists(os.path.realpath(self.data_file_name)):
             print("The data you specified doesn't exist!")
             raise ValueError
-        if ".csv" not in data_file_name:
+        if ".csv" not in self.data_file_name:
             print("The data you specified is not valid! Only support .csv file.")
             raise ValueError
-        data_name = data_file_name.replace(".csv", "")
+        data_name = self.data_file_name.replace(".csv", "")
         self.data_name = data_name[data_name.rfind("/")+1:]
-        if sep_flag:
-            df = pd.read_csv(data_file_name, sep=sep_flag)
+
+        if self.sep_flag:
+            df = pd.read_csv(self.data_file_name, sep=self.sep_flag)
         else: # default ',' separated data
-            df = pd.read_csv(data_file_name)
+            df = pd.read_csv(self.data_file_name)
+
         if not df.shape[0]:
             print("Uploaded data is empty!")
             raise ValueError
 
+        # np.random.seed(fixed_random_seed)
+        # self.fixed_random_seed = fixed_random_seed
+
         # integrity check for target_col
-        if target_col is None or target_positive_values is None:
+        if self.target_col is None or self.target_positive_values is None:
             print("Need to specify target_col and target_positive_value!")
             raise ValueError
 
-        if target_col not in df.columns:
+        if self.target_col not in df.columns:
             print("Need to specify a valid target attribute to be predicted!")
             raise ValueError
-        target_values = df[target_col].unique()
+        target_values = df[self.target_col].unique()
         if len(target_values) != 2:
             print("Only support binary target feature now!")
             raise ValueError
-        if len(set(target_positive_values).intersection(target_values)) == 0:
+        if len(set(self.target_positive_values).intersection(target_values)) == 0:
             print("Need to specify a valid target positive value!")
             raise ValueError
-        self.target_col = target_col
-        self.target_positive_values = target_positive_values
 
-        # integrity check for sensitive_atts and protected_values
-        input_indicator = sum([len(x)== 0 for x in [sensitive_atts, protected_values]])
-        if input_indicator == 0: # both are specified
-            if len(sensitive_atts) != len(protected_values):
-                print("Different size of input sensitive attributes and protected values!")
-                raise ValueError
-            if sum([len(set(protected_values[x]).difference(df[x].unique())) > 0 for x in protected_values]) > 0:
-                print("Some specified protected values do not appear in the column specified in sensitive_atts!")
-                raise ValueError
-        elif input_indicator == 1: # one of parameter is empty
-            print("Need to specify both sensitive_atts and protected_values!")
-            raise ValueError
-        else: # both are empty
-            # TODO: add auto-generation for the below two variables: sensitive_atts and protected_values
-            # for adult only
-            sensitive_atts = ["sex", "race"]
-            protected_values = {"sex": ["Female"], "race": ["Black", "Asian-Pac-Islander", "Amer-Indian-Eskimo", "Other"]}
+        # TODO: update the below code
+        # # integrity check for sensitive_atts and protected_values
+        # input_indicator = sum([len(x)== 0 for x in [self.sensitive_atts, self.protected_values]])
+        # if input_indicator == 0: # both are specified
+        #     if len(self.sensitive_atts) != len(self.protected_values):
+        #         print("Different size of input sensitive attributes and protected values!")
+        #         raise ValueError
+        #     if sum([len(set(self.protected_values[x]).difference(df[x].unique())) > 0 for x in self.protected_values]) > 0:
+        #         print("Some specified protected values do not appear in the column specified in sensitive_atts!")
+        #         raise ValueError
+        # elif input_indicator == 1: # one of parameter is empty
+        #     print("Need to specify both sensitive_atts and protected_values!")
+        #     raise ValueError
+        # else: # both are empty
+        #     print("Need to specify both sensitive_atts and protected_values!")
+        #     raise ValueError
+
+        self.raw_data = df
+        self.attributes = df.columns
+        return self
+
+    def get_meta_information(self):
+        # infer the numerical and categorical attributes first
+        if self.raw_data.describe().shape[0] == 8:# DataFrame.describe() usually returns 8 rows.
+            num_atts = set(self.raw_data.describe().columns)
+        else:# DataFrame.describe() returns less than 8 rows when there is no numerical attribute.
+            num_atts = set()
 
-        self.sensitive_atts = sensitive_atts
-        # self.protected_values = protected_values
-        self.zero_mapping = {target_col: [x for x in target_values if x not in target_positive_values]}
-        self.zero_mapping.update(protected_values)
-        if na_mark:
-            self.na_mark = na_mark
-        else:
-            self.na_mark = None
+        cate_atts = set(self.attributes).difference(num_atts)
 
-        # refer numerical and categorical attributes first
-        # DataFrame.describe() usually returns 8 rows.
-        if df.describe().shape[0] == 8:
-            num_atts = set(df.describe().columns)
-        # DataFrame.describe() returns less than 8 rows when there is no numerical attribute.
-        else:
-            num_atts = set()
 
-        cate_atts = set(df.columns).difference(num_atts)
+        self.metadata_desp = {"size": self.raw_data.shape[0], "features": self.raw_data.shape[1],
+                              "categorical features": list(cate_atts), "numerical features": list(num_atts)}
+        # record the domain of the attributes
+        feature_domain = {}
+        for attri in self.attributes:
+            if attri in cate_atts:
+                all_values = self.raw_data[attri].unique()
+                feature_domain[attri] = (len(all_values), list(all_values))
+            else:
+                feature_domain[attri] = (min(self.raw_data[attri]), max(self.raw_data[attri]))
+        self.metadata_desp.update({"domain": feature_domain})
+
+        # remove the sensitive and target attributes from the inferred list since these attributes are treated separately in the pipeline
         if self.target_col in cate_atts:
             cate_atts.remove(self.target_col)
         if self.target_col in num_atts:
             num_atts.remove(self.target_col)
-        for si in sensitive_atts:
+        for si in self.sensitive_atts:
             cate_atts.remove(si)
+
         self.num_atts = list(num_atts)
         self.cate_atts = list(cate_atts)
-        self.df = df
-
-        # record the sensitive attributes and target variable value mapping
-        sensi_target_value_mapping = {}
-        for atti in [self.target_col] + self.sensitive_atts:
-            atti_values = list(df[atti].unique())
-            pro_values = self.zero_mapping[atti]
-            if len(pro_values) > 1:
-                pro_value_str = pro_values[0] + "&more"
-            else:
-                pro_value_str = pro_values[0]
-            other_values = list(set(atti_values).difference(pro_values))
-            if len(other_values) > 1:
-                other_value_str = other_values[0] + "&more"
-            else:
-                other_value_str = other_values[0]
-            sensi_target_value_mapping[atti] = {0: pro_value_str, 1: other_value_str}
-        # print(sensi_target_value_mapping)
-        self.sensi_target_value_mapping = sensi_target_value_mapping
-        self.pipeline_id = self.data_name[:2]
-
-    def init_necessary_steps(self, step_flag, apply_df, input_weights=[]):
-        # initialize the necessary steps
-        if step_flag == "Imputer":
-            return DropNAImputer(apply_df, na_mark=self.na_mark)
-        # elif step_flag == "Scaler":
-        #     return SK_StandardScaler(apply_df, list(self.num_atts))
-        elif step_flag == "Encoder":
-            return SK_OneHotEncoder(apply_df, list(self.cate_atts))
-        elif step_flag == "SensitiveEncoder":
-            return CustomCateAttsEncoder(apply_df, self.sensitive_atts+[self.target_col], self.zero_mapping)
-        else:
-            return OPT_LogisticRegression(apply_df, self.target_col, instance_weights=input_weights)
 
-    def print_necessary_steps(self):
-        # for printout and efficiency
-        return {"Imputer": ("DropNAImputer", "?"),
-                # "Scaler": ("SK_StandardScaler", list(self.num_atts)),
-                "Encoder": ("SK_OneHotEncoder", list(self.cate_atts)),
-                "SensitiveEncoder": ("CustomCateAttsEncoder", self.sensitive_atts+[self.target_col], self.zero_mapping),
-                "model": ("OPT_LogisticRegression", self.target_col)}
+        log_dir_name = os.path.join(os.path.dirname(os.path.realpath(__file__)), self.generate_file_path())
+        if not os.path.exists(log_dir_name):
+            os.makedirs(log_dir_name)
+
+        self.log_dir_name = log_dir_name
+        print("**"*10, log_dir_name)
+        # print(self.metadata_desp)
+
+        with open(log_dir_name+"/"+self.data_name+'.json', 'w') as outfile:
+            json.dump(dict(self.metadata_desp), outfile)
+
+        return self
+
+    def iter_steps(self, steps, input_df):
+        """
+        Generator function to iterate steps.
+        :param steps: list of objects that represent the steps user want to perform on the input data.
+                      Supported steps are listed in STEPS.md.
+        :return:  the pandas dataframes that are returned by applying a step on the input data.
+        """
+        for idx, stepi in enumerate(islice(steps, 0, len(steps))):
+            # if the current step is the encoder, feed inputdata's metadata so that the encoded data can have the domain of the raw dataset
+            stepi.fit(input_df)
+            return_df = stepi.apply(input_df)
+            if len(return_df) == 2: # special heck for the step that return weights
+                return_df = return_df[0]
+            yield idx, stepi.fitted_step, return_df
+
+    def validate_input_steps(self, steps):
+
+        if not steps or not isinstance(steps, list):
+            print("Require non-empty list of steps as input!")
+            raise ValueError
+
+        # TODO: integrity check for steps using pre-requisite order
+
+        return self
 
+    def fill_zero_to_dummy_data(self, input_step, df):
+        raw_df = pd.get_dummies(self.raw_data, columns=self.cate_atts, prefix_sep='=')
+        if input_step.step_name() in ["FairPreprocessor", "Model", "FairInprocessor", "FairPostprocessor"]: # these steps return non-encoded data
+            encoded_df = pd.get_dummies(df, columns=self.cate_atts, prefix_sep='=')
+        else:
+            encoded_df = df
+        # print("**" * 10, raw_df.shape, encoded_df.shape)
+        if raw_df.shape[1] != encoded_df.shape[1]:
+            for coli in set(raw_df.columns):
+                if "=" in coli and coli[:coli.find("=")] in self.cate_atts and coli not in encoded_df.columns:
+                    encoded_df.loc[:, coli] = 0
+
+        # print("**" * 10, raw_df.shape, encoded_df.shape)
+        return encoded_df
 
-    def run_pipeline(self, steps, return_test=True, output_interdata=False):
+    def run_pipeline(self, steps, save_interdata=False):
         """
 
         :param df: pandas dataframe, the data on which steps are applied.
@@ -191,316 +230,117 @@ def run_pipeline(self, steps, return_test=True, output_interdata=False):
         :return:  two pandas dataframes: before_data and after_data. after_data is the data after applied the input steps.
 
         """
+        self.validate_input_parameters()
+        self.get_meta_information()
 
-        if not steps:
-            print("Require list of steps as input!")
-            raise ValueError
+        self.validate_input_steps(steps)
 
-        if len(steps) < len(SUPPORT_STEPS):
-            print("Missing some input steps! Required steps are listed in the order below.\n"+" ".join(SUPPORT_STEPS.keys()))
-            raise ValueError
-        if sum([len(set(x[0]).intersection(list(SUPPORT_STEPS.values())[idx]))==0 for idx, x in enumerate(steps) if x is not None]) > 0:
-            print("Some input steps are not supported!")
-            raise ValueError
-        if sum([x[0] not in ALL_STEPS for x in steps if x is not None]) > 0:
-            print("Some input steps don't include enough parameters!")
-            raise ValueError
-        if sum([len(x) <=1 for x in steps if x is not None]) > 0:
-            print("Some input steps don't include enough parameters!")
-            raise ValueError
+        # first run the splitter
+        print(self.print_log_message(steps, 0))
+        train_df, val_df, test_df = steps[0].apply(self.raw_data)
+        print(PRINT_SPLIT)
 
-        if steps[-1] is not None and len(steps[0][1])==2: # run fair-postprocessor, then require validation set
-            print("FairPostprocessor requires a validation set! Specify through split_ratio in Splitter!")
-            raise ValueError
+        # for step_idx, train_fitted_step, train_df in self.iter_steps(steps, train_df):
+        for step_idx, stepi in enumerate(steps):
+            if step_idx == 0:
+                continue
 
-        self.pipeline_id = "_".join([self.pipeline_id]+[str(x[0]) for x in steps if x is not None])
+            print(self.print_log_message(steps, step_idx))
 
-        support_step_names = list(SUPPORT_STEPS.keys())
+            # run the same step on validation data, if multiple methods are specified at one step, the one with optimal performance on validation set is selected
+            if isinstance(steps[step_idx], list): # multiple methods specified at one step
+                # TODO: add the selection on validation data
+                pass
 
-        # split the data into separated datasets for train, [validation], and test
-        if steps[0] is None: # default splitter
-            if steps[-1] is not None: # train, validation and test data
-                cur_splitter = BalanceTargetSplitter(self.df, [0.5, 0.3, 0.2], self.target_col)
-            else: # train and test data
-                cur_splitter = BalanceTargetSplitter(self.df, [0.7, 0.3], self.target_col)
-        else:
-            cur_splitter = init_input_steps(steps[0], self.df)
-        after_data = cur_splitter.apply(self.df)
-        after_data = list(after_data)
-        if output_interdata:
-            self.save_inter_data(after_data, cur_splitter.get_name())
-        print("Done "+support_step_names[0]+PRINT_SPLIT)
-        # record the before data to output
-        before_data = [x for x in after_data]
-
-
-        # run sampler on train
-        if steps[1] is not None:
-            for idx_df, cur_df in enumerate(after_data):
-                cur_sampler = init_input_steps(steps[1], cur_df)
-                after_data[idx_df] = cur_sampler.apply(cur_df)
-            if output_interdata:
-                self.save_inter_data(after_data, cur_sampler.get_name(), steps[:1])
-            print("Done "+support_step_names[1]+PRINT_SPLIT)
-
-
-        # run the preprocess steps: "Imputer", "Scaler", "Categorizer" that fit on train and apply on others
-        for idx, step_i in enumerate(steps[2:5]):
-            idx = idx + 2
-            step_i_key = support_step_names[idx]
-            # fit on train data
-            if step_i is None:
-                if step_i_key in list(self.print_necessary_steps().keys()):  # add default operation for necessary steps
-                    step_i = self.init_necessary_steps(step_i_key, after_data[0])
-                else:  # skip the step
-                    continue
-            else:  # user-specified step
-                step_i = init_input_steps(steps[idx], after_data[0])
-            # apply on train, validation and test data
-            for idx_df, cur_df in enumerate(after_data):
-                after_data[idx_df] = step_i.apply(cur_df)
-
-            if output_interdata:
-                self.save_inter_data(after_data, step_i.get_name(), steps[:idx])
-            print("Done " + support_step_names[idx] + PRINT_SPLIT)
-
-        # run the preprocess steps: "Encoder"
-        # fit and apply on the same data
-        after_data, encoder_name = self.run_encoder(steps[5], after_data)
-        if output_interdata:
-            self.save_inter_data(after_data, encoder_name, steps[:5])
-        print("Done " + support_step_names[5] + PRINT_SPLIT)
-
-        # run the preprocess steps: "SensitiveAttEncoder"
-        # fit and apply on the same data
-        if steps[6] is None:
-            for idx_df, cur_df in enumerate(after_data):
-                cur_sensi_encoder = self.init_necessary_steps("SensitiveEncoder", cur_df)
-                after_data[idx_df] = cur_sensi_encoder.apply(cur_df)
-        else: # user-specified sensitive encoder
-            for idx_df, cur_df in enumerate(after_data):
-                cur_sensi_encoder = init_input_steps(steps[6], cur_df)
-                after_data[idx_df] = cur_sensi_encoder.apply(cur_df)
-        if output_interdata:
-            self.save_inter_data(after_data, cur_sensi_encoder.get_name(), steps[:6])
-        print("Done " + support_step_names[6] + PRINT_SPLIT)
-
-        # run the preprocess steps: "FairPreprocessor"
-        # fit and apply on the same data
-        if steps[7] is not None:
-            weights = [[0 for _ in range(x.shape[0])] for x in after_data]
-            for idx_df, cur_df in enumerate(after_data):
-                cur_fair_preprossor = init_input_steps(steps[7], cur_df)
-                if "AIF_Reweighing" in cur_fair_preprossor.get_name():  # special heck for methods updating sample weight
-                    after_data[idx_df], weights[idx_df] = cur_fair_preprossor.apply(cur_df)
-                else:
-                    after_data[idx_df] = cur_fair_preprossor.apply(cur_df)
-            if output_interdata:
-                self.save_inter_data(after_data, cur_fair_preprossor.get_name(), steps[:7])
-            print("Done " + support_step_names[7] + PRINT_SPLIT)
-
-            # after fair-preprocess, rerun encoder
-            after_data, encoder_name = self.run_encoder(steps[5], after_data)
-            if output_interdata:
-                self.save_inter_data(after_data, encoder_name+"_prep", steps[:5])
-            print("Done " + support_step_names[5] + " for fair preprocessor "+PRINT_SPLIT)
-
-        # run model step
-        # fit on train data
-        if steps[8] is None:
-            if weights:
-                cur_model = self.init_necessary_steps("model", after_data[0], input_weights=weights[0])
-            else:
-                cur_model = self.init_necessary_steps("model", after_data[0])
-        else: # TODO: add the support for weight in user-specified models
-            cur_model = init_input_steps(steps[8], after_data[0])
-
-        # predict on train, validation and test data
-        for idx_df, cur_df in enumerate(after_data):
-            after_data[idx_df] = cur_model.apply(cur_df)
-        if output_interdata:
-            self.save_inter_data(after_data, cur_model.get_name(), steps[:8])
-        print("Done " + support_step_names[8] + PRINT_SPLIT)
-
-        # run fair postprocess step
-        if steps[9] is not None:
-            # encode first
-            after_data, encoder_name = self.run_encoder(steps[5], after_data)
-            if output_interdata:
-                self.save_inter_data(after_data, encoder_name+"_post", steps[:5])
-            print("Done " + support_step_names[5] + " for fair post processor " + PRINT_SPLIT)
-            # fit on validation data
-            cur_postprocessor = init_input_steps(steps[9], after_data[1])
-            # predict on validation and test data
-            for idx_df, cur_df in enumerate(after_data[1:]):
-                after_data[idx_df+1] = cur_postprocessor.apply(cur_df)
-            if output_interdata:
-                self.save_inter_data(after_data, cur_postprocessor.get_name(), steps[:9])
-            print("Done " + support_step_names[9] + PRINT_SPLIT)
-
-        # transfer back to original values for encoded sensitive and target columns
-        for idx, df_i in enumerate(after_data):
-            for atti in [self.target_col]+ self.sensitive_atts:
-                df_i[atti] = df_i[atti].apply(lambda x: self.sensi_target_value_mapping[atti][x])
-            if "pred_" +self.target_col in df_i.columns:
-                df_i["pred_" + self.target_col] = df_i["pred_" + self.target_col].apply(lambda x: int(x>=0.5))
-                df_i["pred_" +self.target_col] = df_i["pred_" +self.target_col].apply(lambda x: self.sensi_target_value_mapping[self.target_col][x])
-
-        if return_test: # only return the before and after of test data
-            return before_data[-1], after_data[-1]
-        else: # return all before and after data
-            return before_data, after_data
-
-    def run_encoder(self, encode_step_tuple, data_list):
-        # run the preprocess steps: "Encoder"
-        # fit and apply on the same data
-        if len(self.cate_atts) > 0:
-            for idx_df, cur_df in enumerate(data_list):
-                if encode_step_tuple is None:  # default encoder
-                    cur_encoder = self.init_necessary_steps(list(SUPPORT_STEPS.keys())[5], cur_df)
-                else:
-
-                    # check for user specified encoder that cover partial categorical atts
-                    non_encoded_cate = set(self.cate_atts).difference(encode_step_tuple[1])
-                    if non_encoded_cate:
-                        cur_encoder = init_input_steps((encode_step_tuple[0], self.cate_atts), cur_df)
-                    else:
-                        cur_encoder = init_input_steps(encode_step_tuple, cur_df)
-                data_list[idx_df] = cur_encoder.apply(cur_df)
-
-            # check for different dimensions after encoding for validation and test set
-            if len(data_list) > 2:
-                for idx_df, cur_df in enumerate(data_list[1:]):
-                    if cur_df.shape[1] != data_list[0].shape[1]:
-                        for feature_i in set(data_list[0].columns).difference(cur_df.columns):
-                            cur_df[feature_i] = 0.0
-                        data_list[idx_df] = cur_df.copy()
-            else: # check the dimensions for train and test set
-                if data_list[0].shape[1] != data_list[1].shape[1]:
-                    diff_features_1 = set(data_list[0].columns).difference(data_list[1].columns)
-                    diff_features_2 = set(data_list[1].columns).difference(data_list[0].columns)
-                    add_df = data_list[1].copy()
-                    for feature_i in diff_features_1.union(diff_features_2):
-                        if feature_i not in add_df.columns:
-                            add_df[feature_i] = 0.0
-                    data_list[1] = add_df.copy()
-
-                    add_df = data_list[0].copy()
-                    for feature_i in diff_features_1.union(diff_features_2):
-                        if feature_i not in add_df.columns:
-                            add_df[feature_i] = 0.0
-                    data_list[0] = add_df.copy()
-
-            return data_list, cur_encoder.get_name()
-        else:
-            return data_list, "None"
+            else: # single method specified at a step
+                stepi.fit(train_df)
+                train_df = stepi.apply(train_df)
 
-    def save_inter_data(self, input_dfs, step_name, pre_steps=[], path="data/inter_data/"):
-        if len(input_dfs) == 2:
-            suffix_names = ["train", "test"]
-        else:
-            suffix_names = ["train", "validation", "test"]
-
-        for idx, df_i in enumerate(input_dfs):
-            output_df_i = df_i.copy()
-            df_path = os.path.realpath(path) + "/" + self.pipeline_id + "/" + suffix_names[idx] + "/"
-            if not os.path.exists(df_path):
-                os.makedirs(df_path)
-            if pre_steps:
-                pre_step_names = [x[0] for x in pre_steps if x is not None]+[step_name]
-            else:
-                pre_step_names = [step_name]
+                if stepi.fit_only_on_train(): # the steps that fit only on train
+                    val_df = stepi.apply(val_df)
+                    test_df = stepi.apply(test_df)
+
+                else: # the steps that treat train, val, test independently through the same rule
+                    stepi.fit(val_df)
+                    val_df = stepi.apply(val_df)
+
+                    stepi.fit(test_df)
+                    test_df = stepi.apply(test_df)
 
-            for atti in [self.target_col]+ self.sensitive_atts:
-                if not isinstance(output_df_i[atti].values[0], str):
-                    output_df_i[atti] = output_df_i[atti].apply(lambda x: self.sensi_target_value_mapping[atti][x])
-            if "pred_" +self.target_col in output_df_i.columns:
-                output_df_i["pred_" + self.target_col] = output_df_i["pred_" + self.target_col].apply(lambda x: int(x >= 0.5))
-                output_df_i["pred_" +self.target_col] = output_df_i["pred_" +self.target_col].apply(lambda x: self.sensi_target_value_mapping[self.target_col][x])
+            # autofill encoded columns after the steps that might affect the dimension of the encoded data. Only for categorical attributes that are not sensitive and target column
+            if stepi.input_encoded_data() != stepi.output_encoded_data():
+                train_df = self.fill_zero_to_dummy_data(stepi, train_df)
+                val_df = self.fill_zero_to_dummy_data(stepi, val_df)
+                test_df = self.fill_zero_to_dummy_data(stepi, test_df)
 
-            output_name = df_path + "__".join([self.data_name, "after"]+[x[:x.find("@")] for x in pre_step_names]) + ".csv"
-            print("!!!!!!!", suffix_names[idx], output_df_i.shape, "!!!!!!!")
-            output_df_i.to_csv(output_name, index=False)
-            print("Current "+suffix_names[idx]+" data after "+" ".join([x[:x.find("@")] for x in pre_step_names])+" \n Stored in ", output_name)
-            print()
+            if save_interdata: # save intermediate data on the disc
+                for file_i, df_i in zip(["train", "val", "test"], [train_df, val_df, test_df]):
+                    self.save_inter_data(file_i, df_i, steps[:step_idx+1])
 
+            print(PRINT_SPLIT)
+
+        # # TODO: move to other location
+        # # transfer back to original values for encoded sensitive and target columns
+        # for dfi in [train_df, val_df, test_df]:
+        #     for atti in [self.target_col] + self.sensitive_atts:
+        #         df_i[atti] = df_i[atti].apply(lambda x: self.sensi_target_value_mapping[atti][x])
+        #
+        #     if self.pred_target in df_i.columns:
+        #         # TODO: check whether to keep the line for score prediction
+        #         df_i[self.pred_target] = df_i[self.pred_target].apply(lambda x: int(x >= 0.5)) # for the model that returns a score instead of labels
+        #         df_i[self.pred_target] = df_i[self.pred_target].apply(lambda x: self.sensi_target_value_mapping[self.target_col][x])
+
+        return train_df, val_df, test_df
+
+    def print_log_message(self, steps, step_idx):
+        if not self.verbose:
+            return None
+        return '(step %d of %d) running %s' % (step_idx + 1, len(steps), steps[step_idx].abbr_name()) + PRINT_SPLIT
+
+    def get_executed_steps_name(self, executed_steps):
+        return "_".join([x.abbr_name() for x in executed_steps])
+
+    def generate_file_path(self, file_name=''):
+        dir_name = '{}__{}/'.format(self.exec_timestamp, self.data_name)
+        return self.log_path + dir_name + file_name
+
+    def generate_timestamp(self):
+        return datetime.fromtimestamp(time()).strftime('%Y-%m-%d_%H-%M-%S-%f')[:-3]
+
+
+    def save_inter_data(self, file_name, df: pd.DataFrame, executed_steps):
+        data_file_name = '{}-{}.csv'.format(self.get_executed_steps_name(executed_steps), file_name)
+        data_file_path = os.path.join(self.log_dir_name, data_file_name)
+
+        df.to_csv(data_file_path, index=False)
+        if self.verbose:
+            print ("Data is saved to ", data_file_path)
 
     
 
 if __name__ == '__main__':
-    # input_steps = [("BalanceTargetSplitter", [0.5, 0.3, 0.2], "income-per-year"),
-    #                ("RandomSampler", 10000),  # sampler
-    #                ("DropNAImputer", "?"),
-    #                ("SK_StandardScaler", ["fnlwgt", "age"]),
-    #                ("SK_Discretizer", ["fnlwgt", "age"], [2, 3]),
-    #                ("SK_OneHotEncoder", ["workclass"]),  # encoder
-    #                ("CustomCateAttsEncoder", ["sex", "race", "income-per-year"], {"sex": ["Female"], "race": ["Black"], "income-per-year": ["<=50K"]}),
-    #                ("AIF_DIRemover", "income-per-year", "sex", 0.8),  # fair-preprocessor
-    #                ("AIF_AdversarialDebiasing", "income-per-year", "sex"), # test Adversial learning
-    #                ("AIF_CalibratedEqOddsPostprocessing", "income-per-year", "sex")  # fair-post-postprocessor
-    #                ]
-    # cur_pip = FairnessLabelPipeline(data_file, y_col, y_posi, sensitive_atts=sensi_atts, protected_values=sensi_pro_valus, na_mark="?")
-    # before_test, after_test = cur_pip.run_pipeline(input_steps, return_test=True, output_interdata=True)
-
-    # data_file = "../data/adult.csv"
-    # y_col = "income-per-year"
-    # y_posi = [">50K"]
-    # na_symbol = "?"
-    # sensi_atts = ["sex", "race"]
-    # sensi_pro_valus = {"sex": ["Female"], "race": ["Black", "Asian-Pac-Islander", "Amer-Indian-Eskimo", "Other"]}
-
-    # data_file = "../data/mylsn_cleaned_2.csv"
-    # y_col = "status"
-    # y_posi = ["Ac"]
-    # na_symbol = "N/A"
-    # sensi_atts = ["sex", "race"]
-    # sensi_pro_valus = {"sex": ["female"], "race": ["black", "hispanic", "native-american", "asian"]}
-
-    #
-    #  input_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),  # splitter
-    #               None,  # sampler
-    #               ("DropNAImputer", "?"),  # imputer
-    #               None,  # scaler
-    #               ("SK_Discretizer", ["age"], [3]),  # categorizer
-    #               None,  # encoder
-    #               None,  # sensitive att and target encoder
-    #               None,  # fair-preprocessor
-    #               ("OPT_LogisticRegression", y_col),  # model
-    #               None  # fair-post-postprocessor
-    #               ]
-
-    # debias_focus_att = "race"
-    # input_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),
-    #               ("RandomSampler", 5000),  # sampler
-    #               None,  # imputer
-    #               None,  # scaler
-    #               None,
-    #               None,  # encoder
-    #               None,
-    #               None, #("AIF_DIRemover", y_col, debias_focus_att, 1.0),  # fair-preprocessor
-    #               ("SK_LogisticRegression", y_col),  # model
-    #               None  # fair-post-postprocessor
-    #               ]
-
-    data_file = "../data/german_AIF.csv"
+
+    data_file = "../data/german_AIF_test.csv"
     y_col = "credit"
     y_posi = ["good"]
     sensi_atts = ["age", "sex"]
-    sensi_pro_valus = {"age": ["young"], "sex": ["female"]}
-    debias_focus_att = "age"
-    fair_steps = [("BalanceTargetSplitter", [0.7, 0.3], y_col),
-                  None,  # sampler
-                  None,  # ("ModeImputer", [], ["workclass"], "?"), # imputer
-                  None,  # scaler
-                  None,  # categorizer
-                  None,  # encoder
-                  None,
-                  ("AIF_Reweighing", y_col, debias_focus_att),  # fair-preprocessor
-                  None,  # ("OPT_LogisticRegression", y_col), # model
-                  None  # fair-post-postprocessor
-                  ]
-
-    cur_pip = FairnessLabelPipeline(data_file, y_col, y_posi, sensitive_atts=sensi_atts,
-                                    protected_values=sensi_pro_valus)
-    before_test, after_test = cur_pip.run_pipeline(fair_steps, return_test=True, output_interdata=True)
+
+    value_mapping = {"female": 0, "male": 1, "good": 1, "bad": 0, "young": 0, "old": 1}
+
+    debias_focus_att = "sex"
+    global_seed = 0
+
+    numerical_atts = ["month", "credit_amount"]
+    categorical_atts = ["status", "employment", "housing"]
+
+    pipeline = [RandomSplitter([0.5, 0.3, 0.2], global_seed), NoSampler(), NoImputer(), SK_MinMaxScaler(numerical_atts),
+                  NoBinarizer(),  OneHotEncoder(categorical_atts), MappingEncoder([y_col] + sensi_atts, value_mapping),
+                  AIF_Reweighing(y_col, debias_focus_att), OPT_LogisticRegression(y_col, global_seed), NoFairPostprocessor()]
+
+
+
+    cur_pip = FairPipeline(data_file, y_col, y_posi, sensi_atts, value_mapping)
+
+
+    train_now, val_now, test_now = cur_pip.run_pipeline(pipeline, save_interdata=True)
 
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-test.csv
new file mode 100644
index 0000000..37c539e
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-test.csv
@@ -0,0 +1,201 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A13,36,5848,A73,A152,male,young,good
+A11,36,2302,A73,A151,male,old,bad
+A12,12,1103,A74,A152,male,old,good
+A14,18,1530,A73,A152,male,old,bad
+A11,15,1721,A72,A152,male,old,good
+A14,18,2775,A74,A152,male,old,bad
+A13,12,1881,A73,A151,female,old,good
+A14,15,3029,A74,A152,male,old,good
+A11,12,1657,A73,A152,male,old,good
+A12,21,2353,A73,A152,male,old,good
+A11,18,2659,A73,A152,male,old,good
+A14,12,1592,A74,A152,female,old,good
+A13,24,1925,A73,A152,male,old,good
+A14,36,11054,A73,A152,male,old,good
+A12,24,1967,A75,A152,female,young,good
+A12,36,2671,A73,A153,female,old,bad
+A14,24,3430,A75,A152,male,old,good
+A13,36,4473,A75,A152,male,old,good
+A14,12,4675,A72,A151,female,young,good
+A14,24,999,A75,A152,male,young,good
+A12,48,10961,A74,A152,male,old,bad
+A12,12,1318,A75,A152,male,old,good
+A11,6,609,A74,A152,female,old,good
+A14,21,2993,A73,A152,male,old,good
+A14,60,15653,A74,A152,male,young,good
+A14,15,3368,A75,A151,male,young,good
+A11,48,4788,A74,A152,male,old,good
+A11,12,2246,A75,A152,male,old,bad
+A11,24,1747,A72,A152,male,young,good
+A12,48,5096,A73,A152,female,old,bad
+A12,18,884,A75,A152,male,old,bad
+A14,18,6070,A75,A152,male,old,good
+A11,24,1382,A74,A152,male,old,good
+A12,6,753,A73,A152,female,old,good
+A14,9,2697,A73,A152,male,old,good
+A12,6,590,A72,A152,male,old,good
+A12,30,1919,A72,A152,male,old,bad
+A11,21,1647,A73,A152,male,old,bad
+A12,18,5866,A73,A152,male,old,good
+A14,24,5943,A72,A152,female,old,bad
+A11,12,727,A72,A152,male,old,bad
+A11,24,2812,A75,A151,female,old,good
+A11,48,4308,A72,A151,female,young,bad
+A11,30,6187,A74,A151,male,young,good
+A12,18,1295,A72,A152,female,old,good
+A14,36,5711,A75,A152,male,old,good
+A14,48,6110,A73,A153,male,old,good
+A12,30,8386,A74,A152,male,old,bad
+A13,18,1961,A75,A152,female,young,good
+A12,12,1158,A73,A152,male,old,good
+A12,12,1410,A73,A152,male,old,good
+A14,33,7253,A74,A152,male,old,good
+A13,6,683,A72,A152,female,old,good
+A14,12,1413,A74,A152,male,old,good
+A11,36,5371,A73,A152,male,old,good
+A12,30,4280,A73,A151,female,old,bad
+A11,24,1358,A75,A152,male,old,bad
+A13,10,781,A75,A153,male,old,good
+A14,4,3380,A74,A152,female,old,good
+A12,9,1082,A75,A152,male,old,good
+A12,36,2337,A75,A152,male,old,good
+A14,7,846,A75,A153,male,old,good
+A12,48,8487,A74,A152,female,young,good
+A13,15,1905,A75,A151,male,old,good
+A12,12,1804,A72,A152,male,old,good
+A11,12,652,A75,A151,female,young,good
+A12,12,1331,A72,A152,male,young,bad
+A14,36,3079,A73,A152,male,old,good
+A13,15,2687,A74,A151,male,old,good
+A13,12,1330,A72,A152,male,old,good
+A11,12,1168,A73,A152,male,old,good
+A11,36,3446,A75,A152,male,old,bad
+A12,9,1199,A74,A152,female,old,good
+A11,12,741,A71,A152,female,young,bad
+A12,18,7374,A71,A152,male,old,good
+A14,18,2169,A73,A152,male,old,bad
+A14,6,1338,A73,A152,male,old,good
+A11,24,6419,A75,A153,female,old,good
+A11,15,975,A73,A152,male,young,good
+A14,24,2684,A73,A152,male,old,good
+A14,24,4526,A73,A152,male,old,good
+A12,12,951,A72,A151,female,old,bad
+A12,54,15945,A72,A151,male,old,bad
+A14,15,3556,A73,A152,male,old,good
+A14,4,1455,A74,A152,male,old,good
+A12,24,1935,A75,A152,male,old,bad
+A12,12,3124,A72,A152,male,old,good
+A14,6,1543,A73,A152,male,old,good
+A11,24,2964,A75,A153,male,old,good
+A11,18,1442,A74,A153,male,old,bad
+A14,60,13756,A75,A153,male,old,good
+A12,20,6148,A75,A152,male,old,good
+A14,24,6313,A75,A152,male,old,good
+A11,12,7865,A75,A153,male,old,bad
+A13,12,1297,A73,A151,male,young,good
+A11,18,2462,A73,A152,male,young,bad
+A14,18,6761,A73,A151,male,old,bad
+A14,36,9572,A72,A152,male,old,bad
+A14,36,7409,A75,A152,male,old,good
+A11,36,6887,A73,A152,male,old,bad
+A14,24,3621,A75,A152,male,old,bad
+A11,6,1352,A71,A151,female,young,good
+A13,12,1480,A71,A153,male,old,good
+A14,18,1028,A73,A152,female,old,good
+A11,12,902,A74,A151,male,young,bad
+A11,10,2315,A75,A152,male,old,good
+A14,9,1980,A72,A151,female,young,bad
+A12,24,1553,A74,A151,female,young,good
+A13,10,1240,A75,A153,female,old,bad
+A12,12,1223,A75,A151,male,old,bad
+A11,21,2606,A72,A151,female,old,good
+A11,12,1409,A75,A152,male,old,good
+A14,18,1568,A73,A151,female,young,good
+A14,5,3448,A74,A152,male,old,good
+A12,18,3872,A71,A152,female,old,good
+A12,12,2366,A74,A152,male,old,good
+A14,36,7678,A74,A152,female,old,good
+A12,36,9398,A72,A151,male,old,bad
+A14,36,9566,A73,A152,female,old,good
+A11,24,1938,A72,A152,male,old,bad
+A11,12,2149,A73,A153,male,old,bad
+A14,36,6614,A75,A152,male,old,good
+A14,12,2859,A71,A152,male,old,good
+A12,20,2629,A73,A152,male,old,good
+A12,18,6204,A73,A152,male,old,good
+A11,6,3676,A73,A151,male,old,good
+A12,15,2326,A73,A152,male,old,good
+A11,6,3384,A73,A151,male,old,bad
+A12,48,3060,A74,A152,male,old,bad
+A13,6,1299,A73,A152,male,old,good
+A14,10,1418,A73,A151,male,old,good
+A14,12,996,A74,A152,female,young,good
+A14,20,3485,A72,A152,male,old,good
+A14,18,3378,A73,A152,male,old,good
+A14,9,1224,A73,A152,male,old,good
+A13,9,745,A73,A152,female,old,bad
+A12,12,1860,A71,A152,male,old,good
+A11,20,4272,A75,A152,female,young,good
+A12,7,2415,A73,A152,male,old,good
+A14,24,3777,A73,A152,male,old,good
+A12,30,2181,A75,A152,male,old,good
+A14,18,4594,A72,A152,male,old,good
+A11,12,759,A74,A152,male,old,bad
+A14,30,4530,A74,A151,female,old,good
+A14,10,1924,A73,A152,male,old,good
+A14,12,3447,A73,A152,female,old,good
+A12,60,14027,A74,A152,male,old,bad
+A13,18,2864,A73,A152,male,old,bad
+A14,12,1255,A75,A152,male,old,good
+A11,42,4370,A74,A152,male,old,bad
+A14,12,926,A71,A152,female,old,good
+A12,15,802,A75,A152,male,old,bad
+A12,12,1092,A73,A152,female,old,good
+A14,21,1591,A74,A152,male,old,good
+A14,36,4454,A73,A152,female,old,good
+A14,12,682,A74,A152,female,old,good
+A12,18,6361,A75,A152,male,old,good
+A14,12,1258,A72,A151,female,young,good
+A12,27,2520,A73,A152,male,young,bad
+A11,6,338,A75,A152,male,old,good
+A12,20,7057,A74,A151,male,old,good
+A11,20,2235,A73,A151,male,old,bad
+A13,12,1424,A75,A152,female,old,good
+A12,24,11328,A73,A152,male,old,bad
+A14,24,1249,A72,A152,male,old,good
+A12,9,2118,A73,A152,male,old,good
+A11,24,1285,A74,A151,female,old,bad
+A14,48,3578,A75,A152,male,old,good
+A14,10,2848,A73,A152,male,old,good
+A11,18,3509,A74,A152,female,young,good
+A11,8,1164,A75,A153,male,old,good
+A12,21,3976,A74,A152,male,old,good
+A12,18,3612,A75,A152,female,old,good
+A12,72,5595,A73,A152,male,young,bad
+A11,12,709,A75,A152,male,old,bad
+A11,36,8229,A73,A152,male,old,bad
+A11,21,3414,A72,A152,male,old,bad
+A12,36,12612,A73,A153,male,old,bad
+A14,6,1750,A75,A152,male,old,good
+A14,12,3077,A73,A152,male,old,good
+A14,15,5324,A75,A153,female,old,good
+A13,4,1494,A72,A152,male,old,good
+A11,12,2578,A71,A153,female,old,good
+A12,7,2329,A72,A152,female,old,good
+A11,9,654,A73,A152,male,old,bad
+A14,36,8133,A73,A152,female,old,good
+A14,24,3863,A73,A153,male,old,good
+A12,12,7472,A71,A151,female,young,good
+A14,12,1555,A75,A153,male,old,bad
+A11,12,1262,A75,A152,male,old,good
+A12,9,790,A73,A152,female,old,good
+A12,30,5234,A71,A152,male,old,bad
+A11,30,2406,A74,A151,female,young,bad
+A12,12,2969,A72,A151,female,young,bad
+A14,21,12680,A75,A153,male,old,bad
+A11,12,1082,A73,A152,male,old,bad
+A12,27,3915,A73,A152,male,old,bad
+A14,9,3832,A75,A152,male,old,good
+A12,18,1928,A72,A152,male,old,bad
+A12,36,9857,A74,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-train.csv
new file mode 100644
index 0000000..1045c8d
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-train.csv
@@ -0,0 +1,501 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,36,3959,A71,A152,male,old,good
+A14,9,3577,A73,A151,male,old,good
+A14,18,2515,A73,A152,male,old,good
+A12,12,1995,A72,A152,male,old,good
+A14,60,10366,A75,A152,male,old,good
+A14,24,7393,A73,A152,male,old,good
+A13,12,409,A73,A151,female,old,good
+A14,9,2507,A75,A153,male,old,good
+A14,30,4811,A74,A151,female,young,good
+A12,48,6560,A74,A152,male,young,bad
+A12,36,9034,A72,A151,male,old,bad
+A11,24,3552,A74,A152,male,old,bad
+A12,20,6468,A71,A152,male,old,good
+A11,30,2522,A75,A152,male,old,good
+A12,36,2820,A72,A152,male,old,bad
+A14,6,783,A73,A152,male,old,good
+A11,21,3763,A74,A152,male,young,good
+A14,30,7485,A71,A152,female,old,bad
+A11,15,1403,A73,A151,female,old,good
+A12,11,4771,A74,A152,male,old,good
+A11,24,4020,A73,A152,male,old,good
+A12,24,1201,A72,A152,male,old,good
+A12,48,5381,A71,A153,male,old,good
+A14,24,5511,A73,A152,male,young,good
+A14,18,1098,A71,A152,female,old,good
+A12,48,6224,A75,A153,male,old,bad
+A14,11,1393,A72,A152,female,old,good
+A12,9,1919,A74,A151,male,old,good
+A14,15,5045,A75,A152,female,old,good
+A12,48,5951,A73,A152,female,young,bad
+A11,8,731,A75,A152,male,old,good
+A14,18,629,A75,A152,male,old,good
+A14,18,1808,A74,A152,female,young,bad
+A12,21,1188,A75,A152,female,old,bad
+A14,24,3235,A75,A152,male,old,good
+A12,24,11560,A73,A151,female,young,bad
+A12,9,1154,A75,A152,male,old,good
+A14,18,3850,A74,A152,male,old,good
+A11,24,2359,A71,A152,male,old,bad
+A14,24,1533,A72,A152,female,old,good
+A13,36,4210,A73,A152,male,old,bad
+A11,30,3857,A73,A152,male,old,good
+A14,12,1393,A75,A152,male,old,good
+A14,18,1800,A73,A152,male,young,good
+A13,12,2251,A73,A152,female,old,good
+A12,12,836,A72,A152,female,young,bad
+A11,18,1345,A73,A152,male,old,bad
+A12,36,4795,A72,A152,female,old,good
+A11,12,708,A73,A152,male,old,good
+A11,12,1289,A73,A152,male,young,good
+A14,18,1864,A73,A152,female,old,bad
+A12,48,9960,A72,A152,female,old,bad
+A11,18,5302,A75,A153,male,old,good
+A11,6,1872,A71,A153,male,old,good
+A11,24,2439,A72,A152,female,old,bad
+A14,9,1388,A73,A151,female,old,good
+A12,48,12169,A71,A153,male,old,good
+A14,18,2662,A74,A152,male,old,good
+A14,13,1409,A71,A152,female,old,good
+A11,15,3643,A75,A152,female,old,good
+A11,48,6758,A73,A152,female,old,bad
+A14,24,1851,A74,A152,male,old,good
+A11,27,3416,A73,A152,male,old,good
+A14,27,5117,A74,A152,male,old,good
+A14,10,1597,A73,A151,male,old,good
+A14,6,1382,A73,A152,female,old,good
+A14,10,1231,A75,A152,male,old,good
+A12,15,2631,A73,A151,female,old,bad
+A12,30,2150,A73,A152,female,young,bad
+A12,12,2930,A74,A152,female,old,good
+A14,6,6761,A74,A152,male,old,good
+A12,21,2745,A74,A152,male,old,good
+A11,24,2325,A74,A152,male,old,good
+A14,6,1204,A73,A151,male,old,good
+A11,18,1049,A72,A151,female,young,good
+A14,12,2292,A71,A152,male,old,bad
+A14,9,1449,A74,A152,female,old,good
+A11,18,7511,A75,A153,male,old,bad
+A12,9,918,A73,A152,female,old,bad
+A14,24,2346,A74,A152,male,old,good
+A11,24,1603,A75,A152,female,old,good
+A11,11,3905,A73,A151,male,old,good
+A14,18,1126,A72,A151,female,young,good
+A13,21,2923,A73,A152,female,old,good
+A12,9,3195,A73,A152,female,old,good
+A11,6,4716,A72,A152,male,old,good
+A12,6,14555,A71,A152,male,young,bad
+A11,36,5493,A75,A153,male,old,good
+A14,24,1585,A74,A152,male,old,good
+A11,12,3590,A73,A152,male,old,good
+A14,12,2759,A75,A152,male,old,good
+A11,12,2122,A73,A151,male,old,good
+A12,12,2002,A74,A151,male,old,good
+A14,24,1901,A73,A151,male,old,good
+A14,12,1493,A72,A152,female,old,good
+A12,6,1209,A71,A152,male,old,bad
+A12,48,3979,A74,A152,male,old,good
+A11,24,7721,A72,A152,female,old,good
+A11,18,3650,A72,A151,female,young,good
+A11,18,1880,A74,A152,male,old,good
+A12,18,1941,A73,A152,male,old,good
+A14,15,874,A72,A152,female,young,good
+A14,27,5190,A75,A152,male,old,good
+A14,36,10477,A75,A153,male,old,good
+A12,24,4351,A73,A152,female,old,good
+A14,24,1376,A74,A152,female,old,good
+A12,30,3496,A73,A152,male,old,good
+A11,40,5998,A73,A152,male,old,bad
+A11,28,4006,A73,A152,male,old,bad
+A14,36,3342,A75,A152,male,old,good
+A12,18,2622,A73,A152,male,old,good
+A12,36,2384,A72,A151,male,old,bad
+A11,24,1659,A72,A151,female,old,bad
+A14,24,3488,A74,A152,female,young,good
+A12,60,7418,A73,A152,male,old,good
+A14,18,2238,A73,A152,female,young,good
+A11,24,1333,A71,A153,male,old,bad
+A14,22,1283,A74,A151,female,young,good
+A14,6,3518,A73,A151,male,old,good
+A12,12,585,A73,A151,male,young,good
+A12,39,4933,A74,A152,male,young,bad
+A14,24,1278,A75,A152,male,old,good
+A14,12,1768,A73,A151,male,young,good
+A11,24,3349,A72,A153,male,old,bad
+A12,9,1391,A73,A152,male,old,good
+A14,18,3422,A75,A152,male,old,good
+A14,10,2146,A72,A151,female,young,good
+A14,15,1520,A75,A152,male,old,good
+A14,24,6842,A73,A152,male,old,good
+A14,18,2051,A72,A152,male,old,good
+A11,24,2996,A73,A152,male,young,bad
+A11,24,3660,A73,A152,female,old,good
+A11,24,1442,A74,A151,female,young,bad
+A11,9,1138,A73,A152,male,young,good
+A14,30,5771,A74,A152,female,young,good
+A11,18,2600,A73,A153,male,old,bad
+A13,24,3749,A72,A152,female,old,good
+A14,12,3059,A74,A152,male,old,good
+A11,6,662,A72,A152,male,old,good
+A11,18,1940,A72,A153,male,old,good
+A12,36,2323,A74,A151,male,young,good
+A12,48,12204,A73,A152,male,old,good
+A14,24,5507,A75,A153,male,old,good
+A14,18,433,A71,A151,female,young,bad
+A14,6,4611,A72,A152,female,old,bad
+A12,12,2762,A75,A152,female,young,bad
+A14,30,4272,A73,A152,male,old,good
+A12,39,11760,A74,A151,male,old,good
+A12,36,2225,A75,A153,male,old,bad
+A13,12,1474,A72,A152,female,old,good
+A14,12,1402,A74,A151,female,old,good
+A12,18,1913,A72,A152,male,old,good
+A14,14,802,A73,A152,male,old,good
+A13,15,1271,A73,A153,male,old,bad
+A13,24,1258,A73,A152,female,old,good
+A14,12,3565,A72,A152,male,old,good
+A11,12,3386,A75,A153,male,old,bad
+A13,12,2247,A73,A152,female,old,good
+A14,24,5804,A73,A152,male,old,good
+A14,15,960,A74,A152,female,old,good
+A12,18,1056,A75,A152,male,old,bad
+A13,42,6289,A72,A152,male,old,good
+A14,9,2753,A75,A152,male,old,good
+A14,10,2210,A73,A151,male,young,bad
+A11,48,10297,A74,A153,male,old,bad
+A12,36,14318,A75,A153,male,old,bad
+A12,30,4249,A71,A152,male,old,bad
+A14,24,929,A74,A152,male,old,good
+A12,9,959,A73,A152,female,old,bad
+A14,21,2782,A74,A152,female,old,good
+A12,48,8358,A72,A152,female,old,good
+A14,9,2301,A72,A151,female,young,good
+A12,48,3566,A74,A152,male,old,good
+A11,18,3966,A75,A151,female,old,bad
+A14,6,1237,A73,A152,female,old,good
+A11,18,976,A72,A152,female,young,bad
+A11,24,915,A75,A152,female,old,bad
+A13,24,947,A74,A153,male,old,bad
+A12,24,2825,A74,A152,male,old,good
+A14,21,2288,A72,A152,female,young,good
+A14,10,2901,A72,A151,female,old,good
+A12,45,4576,A71,A152,male,old,good
+A14,48,7238,A75,A152,male,old,good
+A11,36,2746,A75,A152,male,old,bad
+A12,24,1743,A75,A152,male,old,good
+A11,9,1366,A72,A151,female,young,bad
+A14,18,1505,A73,A153,male,old,good
+A11,6,343,A72,A152,female,old,good
+A14,24,2223,A75,A152,male,old,good
+A14,12,1495,A75,A152,male,old,good
+A13,15,2327,A72,A152,female,young,bad
+A11,45,11816,A75,A151,male,old,bad
+A12,12,1484,A73,A152,male,young,bad
+A14,36,9055,A73,A153,male,old,good
+A14,12,1655,A75,A152,male,old,good
+A14,18,1453,A72,A152,female,old,good
+A11,12,1228,A73,A152,female,young,bad
+A12,6,932,A74,A152,female,old,good
+A12,36,8086,A75,A152,male,old,bad
+A14,6,1346,A75,A153,male,old,good
+A12,36,2862,A75,A153,male,old,good
+A11,24,3123,A72,A152,female,old,bad
+A12,60,7408,A72,A152,female,young,bad
+A12,15,1264,A73,A151,male,young,bad
+A14,6,1554,A74,A151,female,young,good
+A12,21,3652,A74,A152,male,old,good
+A11,48,3931,A74,A153,male,old,bad
+A11,6,1361,A72,A152,male,old,good
+A11,6,1203,A75,A152,male,old,good
+A14,12,3527,A72,A152,male,old,good
+A14,36,7855,A73,A152,female,young,bad
+A11,21,3357,A72,A152,female,old,good
+A13,18,2100,A73,A152,male,old,bad
+A14,18,1169,A73,A152,male,old,good
+A14,36,8947,A74,A152,male,old,good
+A11,18,1936,A74,A151,male,young,good
+A12,24,3512,A74,A152,male,old,good
+A12,24,6967,A74,A151,male,old,good
+A14,30,2333,A75,A152,male,old,good
+A11,18,2039,A73,A151,female,young,bad
+A11,36,9271,A74,A152,male,young,bad
+A14,21,1572,A75,A152,female,old,good
+A12,6,454,A72,A152,male,young,good
+A14,36,6304,A75,A152,male,old,good
+A14,24,2197,A74,A152,male,old,good
+A14,48,7629,A75,A152,male,old,good
+A13,10,1275,A72,A152,female,young,good
+A11,48,7119,A73,A153,male,old,bad
+A12,24,12579,A75,A153,female,old,bad
+A12,18,2779,A73,A151,male,young,good
+A14,4,1503,A74,A152,male,old,good
+A12,8,1237,A73,A152,female,young,bad
+A14,48,3914,A73,A152,male,old,bad
+A14,24,3062,A75,A151,male,old,good
+A12,36,1953,A75,A153,male,old,bad
+A12,30,3832,A72,A152,male,young,good
+A14,6,1538,A72,A152,female,old,good
+A14,6,2108,A74,A151,male,old,good
+A11,6,428,A75,A152,female,old,good
+A12,36,4455,A73,A152,male,old,bad
+A13,21,2319,A72,A151,male,old,bad
+A14,12,1163,A73,A152,male,old,good
+A14,9,936,A75,A152,male,old,good
+A11,18,2473,A71,A152,male,young,bad
+A14,15,1569,A75,A152,male,old,good
+A12,12,3017,A72,A151,female,old,good
+A14,12,2331,A75,A152,male,old,good
+A14,12,2096,A74,A152,male,old,good
+A12,12,1037,A74,A152,male,old,good
+A14,6,1236,A73,A151,male,old,good
+A14,24,4151,A73,A152,male,old,good
+A14,48,3609,A73,A152,female,old,good
+A14,12,1412,A73,A152,female,old,good
+A11,11,3939,A73,A152,male,old,good
+A11,12,1526,A75,A153,male,old,good
+A14,22,2675,A75,A152,male,old,good
+A12,30,4221,A73,A152,female,old,good
+A11,24,2924,A73,A152,male,old,good
+A14,18,1817,A73,A152,female,old,good
+A13,10,1225,A73,A152,male,old,good
+A11,15,1845,A72,A151,female,old,good
+A14,6,250,A73,A152,female,old,good
+A14,6,660,A74,A151,male,young,good
+A13,6,709,A72,A152,male,old,good
+A11,36,15857,A71,A152,male,old,good
+A12,24,1837,A74,A153,female,old,bad
+A11,13,1797,A72,A152,male,old,good
+A11,6,448,A72,A152,female,young,bad
+A12,18,3213,A72,A151,male,young,good
+A11,12,2579,A72,A152,male,old,bad
+A11,6,1198,A75,A153,female,old,bad
+A14,21,5003,A73,A152,female,old,bad
+A12,24,2333,A72,A152,male,old,good
+A14,15,1262,A74,A152,male,old,good
+A12,12,1155,A75,A152,male,old,good
+A14,10,894,A74,A152,female,old,good
+A14,12,3331,A75,A152,male,old,good
+A12,18,1113,A73,A152,female,old,good
+A14,12,719,A75,A152,male,old,bad
+A12,24,2064,A71,A152,female,old,bad
+A13,30,3656,A75,A152,male,old,good
+A11,24,6615,A71,A153,male,old,good
+A14,18,3229,A71,A152,male,old,good
+A14,15,3343,A73,A153,male,old,good
+A11,30,3108,A72,A152,male,old,bad
+A12,9,458,A73,A152,male,young,good
+A14,15,1471,A73,A153,male,old,good
+A12,11,1322,A73,A152,female,old,good
+A11,9,1288,A75,A152,male,old,good
+A14,12,707,A73,A152,male,old,good
+A12,15,1308,A75,A152,male,old,good
+A14,9,3074,A73,A152,male,old,good
+A11,15,2511,A71,A151,female,young,good
+A14,6,1740,A75,A151,male,old,good
+A14,12,1123,A73,A151,female,old,bad
+A12,11,1577,A72,A152,female,young,good
+A11,45,1845,A73,A153,male,young,bad
+A14,24,2028,A74,A152,male,old,good
+A12,27,8318,A75,A153,female,old,bad
+A12,12,1295,A72,A151,female,young,bad
+A12,24,11938,A73,A152,male,old,bad
+A14,36,1819,A73,A153,male,old,bad
+A11,15,1478,A75,A152,male,old,good
+A14,24,1516,A73,A152,female,old,good
+A11,9,2136,A73,A152,male,young,good
+A14,15,3812,A72,A152,female,young,good
+A14,18,4165,A73,A152,male,old,bad
+A12,24,4057,A74,A152,male,old,bad
+A14,48,10127,A73,A153,male,old,bad
+A12,6,484,A74,A152,male,old,good
+A12,60,6288,A73,A153,male,old,bad
+A11,24,6872,A72,A152,male,old,bad
+A14,15,1532,A73,A152,female,old,good
+A14,9,2134,A73,A152,male,old,good
+A11,12,1107,A73,A151,male,young,good
+A11,20,2212,A74,A152,male,old,good
+A14,15,3186,A74,A151,female,young,good
+A11,47,10722,A72,A152,female,old,good
+A14,7,730,A75,A151,male,old,good
+A14,12,2073,A73,A152,female,old,good
+A11,21,571,A75,A152,male,old,good
+A13,15,2360,A73,A152,male,old,good
+A11,12,1344,A73,A152,male,old,good
+A11,42,7174,A74,A152,female,old,bad
+A13,42,4796,A75,A153,male,old,good
+A11,42,3965,A72,A152,male,old,bad
+A14,11,7228,A73,A152,male,old,good
+A12,24,1216,A72,A152,male,old,bad
+A14,12,2748,A75,A153,female,old,good
+A12,42,9283,A71,A153,male,old,good
+A14,27,4526,A72,A152,male,old,good
+A14,21,5248,A73,A152,male,old,good
+A14,24,3181,A72,A152,female,old,good
+A14,24,717,A75,A152,male,old,good
+A14,12,1262,A73,A152,male,young,good
+A14,12,1884,A75,A152,male,old,good
+A12,18,1924,A72,A151,female,old,bad
+A11,48,7476,A74,A153,male,old,good
+A13,12,939,A74,A152,male,old,bad
+A12,12,841,A74,A151,female,young,good
+A12,24,4113,A72,A151,female,old,bad
+A13,24,5152,A74,A152,male,young,good
+A11,33,4281,A73,A152,female,young,bad
+A12,12,1567,A73,A152,female,young,good
+A11,12,339,A75,A152,male,old,good
+A11,15,1275,A73,A151,female,young,bad
+A12,12,639,A73,A152,male,old,bad
+A14,36,2299,A75,A152,male,old,good
+A11,24,1823,A71,A152,male,old,bad
+A11,48,7763,A75,A153,male,old,bad
+A14,6,1595,A74,A152,male,old,good
+A11,12,385,A74,A152,female,old,good
+A14,12,776,A73,A152,male,old,good
+A12,24,3069,A75,A153,male,old,good
+A11,9,1364,A74,A152,male,old,good
+A14,6,1221,A73,A152,male,old,good
+A11,8,3398,A74,A152,male,old,good
+A11,18,4153,A73,A152,male,old,bad
+A12,15,1778,A72,A151,female,old,bad
+A14,9,1236,A72,A151,female,young,good
+A14,6,1766,A73,A151,male,young,good
+A11,30,8072,A72,A152,male,young,good
+A12,9,5129,A75,A153,female,old,bad
+A14,24,8648,A72,A152,male,old,bad
+A11,24,4169,A73,A152,male,old,good
+A14,12,1935,A75,A152,male,old,good
+A14,24,2397,A75,A152,male,old,bad
+A14,15,1459,A73,A152,female,old,good
+A12,12,685,A74,A152,male,young,bad
+A11,48,4605,A75,A153,male,young,bad
+A14,11,2142,A75,A152,male,old,good
+A11,15,950,A75,A151,male,old,bad
+A12,16,1175,A71,A153,male,old,good
+A11,12,1498,A73,A152,female,young,good
+A14,18,1984,A73,A153,male,old,good
+A12,36,3804,A73,A152,female,old,bad
+A11,24,1987,A73,A151,male,young,bad
+A14,24,4139,A73,A152,male,old,good
+A14,12,701,A73,A152,male,old,good
+A14,54,9436,A73,A152,male,old,good
+A11,18,3190,A73,A152,female,young,bad
+A12,13,882,A72,A152,male,young,good
+A12,24,2760,A75,A153,male,old,good
+A14,15,2221,A73,A151,female,young,good
+A11,12,626,A73,A152,female,young,bad
+A12,48,18424,A73,A152,female,old,bad
+A11,18,1217,A73,A152,male,old,bad
+A12,7,2576,A73,A152,male,old,good
+A11,6,2647,A73,A151,male,old,good
+A14,27,2570,A73,A151,female,young,bad
+A13,24,3148,A73,A152,male,old,good
+A11,6,666,A74,A152,female,old,good
+A14,12,1185,A73,A152,female,old,good
+A14,30,5954,A74,A152,male,old,good
+A12,45,4746,A72,A152,male,young,bad
+A14,15,1213,A75,A152,male,old,good
+A14,6,1898,A73,A152,male,old,good
+A14,48,11590,A73,A151,female,young,bad
+A14,4,601,A72,A151,female,young,good
+A12,18,1245,A73,A152,male,old,bad
+A14,36,5742,A74,A152,male,old,good
+A12,14,1410,A75,A152,male,old,good
+A11,12,900,A73,A152,male,young,bad
+A14,6,362,A73,A152,female,old,good
+A14,15,1979,A75,A152,male,old,good
+A14,36,3349,A73,A152,female,old,bad
+A14,12,797,A75,A152,female,old,bad
+A14,6,518,A73,A152,female,old,good
+A11,24,2303,A75,A152,male,old,bad
+A11,36,2712,A75,A152,male,old,bad
+A14,24,1474,A72,A152,male,old,good
+A14,18,1820,A73,A152,male,old,good
+A14,24,1927,A73,A152,female,old,good
+A11,15,1433,A73,A151,female,young,good
+A12,12,958,A74,A152,male,old,good
+A14,36,3835,A75,A152,female,old,good
+A14,12,763,A73,A152,female,old,good
+A14,6,2978,A73,A152,male,old,good
+A14,12,1291,A73,A152,female,old,good
+A11,12,674,A74,A152,male,young,bad
+A14,60,6527,A73,A153,male,old,good
+A14,12,976,A75,A152,male,old,good
+A13,10,3949,A72,A152,male,old,good
+A12,10,1048,A73,A152,male,young,good
+A14,24,2670,A75,A152,male,old,good
+A11,30,4583,A73,A152,male,old,good
+A14,24,2375,A73,A152,male,old,good
+A11,12,1282,A73,A151,female,young,bad
+A14,24,3757,A75,A153,female,old,good
+A12,24,4241,A73,A152,male,old,bad
+A11,24,6579,A71,A153,male,old,good
+A11,12,2121,A73,A152,male,old,good
+A14,12,2171,A72,A152,female,old,good
+A14,15,3594,A72,A152,female,old,good
+A14,36,4686,A73,A153,male,old,good
+A14,15,4657,A73,A152,male,old,good
+A11,14,8978,A75,A152,male,old,bad
+A14,18,1055,A72,A152,female,old,good
+A11,48,3051,A73,A152,male,old,bad
+A12,30,3441,A73,A151,female,young,bad
+A14,24,5150,A75,A152,male,old,good
+A14,12,1963,A74,A151,male,old,good
+A12,18,3590,A71,A152,male,old,good
+A14,36,5842,A75,A152,male,old,good
+A14,24,1469,A75,A151,male,old,good
+A14,24,2603,A73,A151,female,old,good
+A11,12,3651,A73,A152,male,old,good
+A11,24,3149,A72,A153,male,young,good
+A11,24,4110,A75,A151,male,young,bad
+A12,9,1437,A74,A152,male,old,bad
+A13,24,1377,A75,A153,female,old,good
+A11,9,1422,A72,A153,male,old,bad
+A11,30,3622,A75,A151,female,old,good
+A11,24,3345,A75,A151,male,old,bad
+A12,12,625,A72,A152,male,old,good
+A12,12,983,A72,A151,female,young,good
+A12,30,2991,A75,A152,female,young,good
+A14,18,1149,A73,A152,male,old,good
+A12,9,276,A73,A151,male,young,good
+A11,24,1371,A73,A151,female,young,bad
+A11,12,795,A72,A152,female,old,bad
+A11,24,1282,A73,A152,female,old,bad
+A12,6,368,A75,A152,male,old,good
+A14,24,1258,A74,A152,male,young,good
+A14,15,2186,A74,A151,female,old,good
+A14,24,2835,A75,A152,male,old,good
+A11,24,3632,A73,A151,female,young,good
+A14,24,1311,A74,A152,male,old,good
+A11,36,2348,A73,A152,male,old,good
+A12,6,433,A72,A151,female,young,bad
+A11,15,3959,A73,A152,female,old,bad
+A12,15,2631,A73,A152,female,young,good
+A14,12,2445,A72,A151,male,old,good
+A11,10,1038,A74,A152,male,old,good
+A14,48,8858,A74,A153,male,old,good
+A13,24,1275,A73,A152,male,old,good
+A11,24,6568,A73,A152,male,young,good
+A13,24,1344,A74,A152,male,old,bad
+A12,27,5965,A75,A152,male,old,good
+A14,12,618,A75,A152,male,old,good
+A14,10,727,A75,A153,male,old,good
+A12,15,1444,A72,A152,male,young,good
+A11,36,3620,A73,A152,male,old,good
+A11,10,2241,A72,A151,male,old,good
+A12,60,14782,A75,A153,female,old,bad
+A14,24,2022,A73,A152,female,old,good
+A13,12,609,A72,A152,female,old,bad
+A11,24,3021,A73,A151,male,young,good
+A11,18,4380,A73,A152,male,old,good
+A14,10,1309,A73,A152,male,old,bad
+A11,24,2828,A73,A152,male,young,good
+A12,18,1795,A75,A151,female,old,good
+A14,18,1533,A72,A152,male,old,bad
+A14,24,5103,A72,A153,male,old,good
+A11,30,10623,A75,A153,male,old,good
+A14,12,1386,A73,A152,female,old,bad
+A14,24,4591,A73,A152,male,old,bad
+A13,12,3399,A75,A152,male,old,good
+A13,6,1047,A73,A152,female,old,good
+A11,48,7685,A74,A151,female,old,bad
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-val.csv
new file mode 100644
index 0000000..e0b67a7
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS-val.csv
@@ -0,0 +1,301 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,18,1553,A73,A152,male,old,bad
+A14,15,1360,A73,A152,male,old,good
+A12,8,760,A74,A152,female,old,good
+A12,18,2899,A75,A152,male,old,good
+A12,45,3031,A73,A151,male,young,bad
+A12,24,4712,A73,A152,male,old,good
+A12,18,1887,A73,A152,male,old,good
+A12,36,2273,A74,A152,male,old,good
+A12,12,888,A75,A152,male,old,bad
+A11,12,1620,A73,A152,female,old,good
+A14,15,2708,A72,A152,male,old,good
+A12,18,866,A73,A152,male,young,good
+A12,8,1414,A73,A152,male,old,good
+A14,9,1313,A75,A152,male,young,good
+A14,24,2578,A75,A152,male,old,good
+A14,15,3568,A75,A151,female,old,good
+A14,24,1393,A73,A152,male,old,good
+A13,6,2116,A73,A152,male,old,good
+A14,12,2012,A74,A152,female,old,good
+A14,24,1940,A75,A152,male,old,good
+A13,6,1323,A75,A152,male,old,good
+A11,36,1977,A75,A152,male,old,bad
+A11,12,1108,A74,A152,male,old,bad
+A14,15,1300,A75,A153,male,old,good
+A11,6,1169,A75,A152,male,old,good
+A11,36,3249,A74,A153,male,old,good
+A14,39,2569,A73,A152,male,young,good
+A12,9,1549,A72,A152,male,old,good
+A14,6,700,A75,A153,male,old,good
+A12,12,6078,A74,A152,male,old,good
+A12,9,1206,A75,A152,female,young,good
+A12,24,1355,A72,A152,female,young,bad
+A14,24,2978,A73,A152,male,old,good
+A14,24,1525,A74,A152,female,old,good
+A14,6,932,A73,A152,female,young,good
+A12,27,2528,A72,A152,female,old,good
+A11,18,3114,A72,A151,female,old,bad
+A13,30,1908,A75,A152,male,old,bad
+A14,12,1240,A75,A152,female,old,good
+A14,12,717,A75,A152,male,old,good
+A12,15,1514,A73,A152,male,young,good
+A14,15,1829,A75,A152,male,old,good
+A14,12,640,A73,A152,male,old,good
+A12,36,7432,A73,A151,female,old,good
+A14,12,1542,A74,A152,male,old,good
+A11,36,9629,A74,A152,male,young,bad
+A14,6,426,A75,A152,male,old,good
+A14,15,4623,A73,A152,male,old,bad
+A11,18,8471,A73,A151,female,young,good
+A12,36,3711,A73,A152,male,old,good
+A14,48,12749,A74,A152,male,old,good
+A11,42,7882,A74,A153,male,old,good
+A14,60,10144,A74,A152,female,young,good
+A12,24,1246,A72,A152,male,young,bad
+A14,36,3595,A75,A152,male,old,good
+A12,12,3617,A75,A151,male,old,good
+A14,12,886,A73,A152,female,young,good
+A11,24,1546,A74,A151,male,young,bad
+A11,12,2171,A73,A152,male,old,good
+A14,36,2613,A73,A152,male,old,good
+A14,12,1101,A73,A152,male,old,good
+A12,18,3001,A74,A151,female,old,good
+A14,12,2279,A73,A153,male,old,good
+A12,10,1521,A73,A152,male,old,good
+A12,18,1042,A73,A152,female,old,bad
+A12,48,6681,A73,A153,male,old,good
+A11,36,2145,A74,A152,male,young,bad
+A11,24,4817,A74,A152,male,old,bad
+A12,48,14421,A73,A152,male,young,bad
+A11,12,691,A75,A152,male,old,bad
+A12,18,1239,A73,A153,male,old,good
+A12,21,2767,A75,A151,male,old,bad
+A12,18,6260,A74,A151,male,old,good
+A11,18,750,A71,A152,female,old,bad
+A14,12,1503,A73,A151,male,old,good
+A11,12,701,A73,A152,male,old,good
+A14,11,1154,A71,A152,female,old,good
+A14,21,3160,A75,A152,male,old,good
+A11,18,2124,A73,A151,female,young,bad
+A12,12,766,A73,A152,male,old,bad
+A12,36,6948,A73,A151,male,old,good
+A13,9,1126,A75,A152,male,old,good
+A11,18,1216,A72,A151,female,young,bad
+A11,12,2214,A73,A152,male,young,good
+A14,24,4042,A74,A152,male,old,good
+A11,15,1053,A72,A152,male,old,good
+A12,24,2896,A72,A152,male,old,good
+A11,12,6199,A73,A151,male,old,bad
+A12,24,7758,A75,A151,female,old,good
+A11,12,1372,A74,A152,male,old,bad
+A11,30,6350,A75,A152,male,old,bad
+A12,18,4439,A75,A152,male,old,good
+A12,48,15672,A73,A152,male,young,bad
+A14,12,804,A75,A152,male,old,good
+A11,21,1602,A75,A152,male,old,good
+A12,12,6468,A71,A152,male,old,bad
+A14,18,1473,A72,A152,male,old,good
+A14,24,5433,A71,A151,female,old,good
+A14,6,1238,A71,A152,male,old,good
+A14,33,2764,A73,A152,female,old,good
+A12,12,1922,A73,A152,male,old,bad
+A13,30,3017,A75,A152,male,old,good
+A11,24,1207,A72,A151,female,young,bad
+A11,16,2625,A75,A151,male,old,bad
+A11,12,684,A73,A151,male,old,bad
+A12,9,1670,A72,A152,female,young,bad
+A14,6,672,A71,A152,female,old,good
+A14,21,2580,A72,A152,male,old,bad
+A14,24,6314,A71,A152,male,old,good
+A11,24,3234,A72,A151,female,young,bad
+A12,10,7308,A71,A153,male,old,good
+A12,13,2101,A72,A152,female,young,good
+A11,36,6229,A72,A151,female,young,bad
+A13,12,3016,A73,A152,male,young,good
+A14,12,522,A75,A152,male,old,good
+A14,12,1934,A75,A152,male,old,good
+A14,48,2751,A75,A152,male,old,good
+A12,15,1512,A73,A152,male,old,bad
+A12,30,2503,A75,A152,male,old,good
+A14,24,1287,A75,A152,female,old,good
+A12,36,12389,A73,A153,male,old,bad
+A13,9,1337,A72,A152,male,old,bad
+A11,27,2442,A75,A152,male,old,good
+A14,10,1364,A73,A152,female,old,good
+A11,24,2384,A75,A151,male,old,good
+A11,9,2799,A73,A151,male,old,good
+A13,18,1445,A74,A152,male,old,good
+A14,6,1743,A73,A152,male,old,good
+A13,10,1347,A74,A152,male,old,good
+A11,24,4870,A73,A153,male,old,bad
+A14,12,1736,A74,A152,female,old,good
+A12,12,754,A75,A152,male,old,good
+A14,36,3535,A74,A152,male,old,good
+A14,15,1478,A73,A152,male,old,good
+A14,30,6742,A74,A152,male,old,good
+A14,36,7127,A72,A151,female,young,bad
+A11,15,806,A73,A152,female,young,good
+A13,24,2892,A75,A153,male,old,good
+A12,36,5800,A73,A152,male,old,good
+A14,24,3105,A72,A152,male,young,good
+A14,15,1386,A73,A151,male,old,good
+A14,18,2404,A73,A152,female,old,good
+A12,24,3878,A72,A152,male,old,good
+A12,24,3092,A72,A151,male,young,bad
+A14,28,2743,A75,A152,male,old,good
+A14,12,1574,A73,A152,male,old,good
+A11,21,1835,A73,A152,female,young,bad
+A12,24,2718,A73,A151,female,young,bad
+A14,48,4844,A71,A151,male,old,bad
+A11,24,1199,A75,A152,male,old,bad
+A11,14,3973,A71,A153,male,young,good
+A14,24,2679,A72,A152,female,old,good
+A14,12,2133,A75,A153,female,old,good
+A11,24,1231,A75,A151,female,old,good
+A14,24,2872,A75,A152,male,old,good
+A12,18,2427,A75,A152,male,old,good
+A11,60,7297,A75,A151,male,old,bad
+A14,30,7596,A75,A152,male,old,good
+A11,39,14179,A74,A152,male,old,good
+A12,9,1501,A75,A152,female,old,bad
+A12,48,3844,A74,A153,male,old,bad
+A14,42,4042,A73,A152,male,old,good
+A11,6,1374,A71,A152,female,old,good
+A14,24,7814,A74,A152,male,old,good
+A14,36,909,A75,A152,male,old,good
+A14,18,6458,A75,A152,male,old,bad
+A14,24,2424,A75,A152,male,old,good
+A14,15,2788,A74,A152,female,young,good
+A11,12,697,A72,A152,male,old,bad
+A14,21,2476,A75,A152,male,old,good
+A12,12,1534,A72,A151,male,young,bad
+A12,6,1068,A75,A152,male,old,good
+A12,12,3573,A73,A152,female,young,good
+A12,18,1301,A75,A152,male,old,good
+A14,42,7166,A74,A151,male,old,good
+A12,24,2039,A72,A152,male,young,bad
+A11,24,2483,A73,A152,male,young,good
+A13,24,3617,A75,A151,male,young,good
+A14,24,2058,A73,A152,male,old,good
+A11,24,1755,A75,A152,female,old,good
+A14,24,1559,A74,A152,male,old,good
+A11,24,3161,A73,A151,male,old,bad
+A14,48,10222,A74,A152,male,old,good
+A11,24,1381,A73,A152,female,old,bad
+A14,21,2241,A75,A152,male,old,good
+A14,24,2538,A75,A152,male,old,bad
+A11,36,8065,A73,A152,female,young,bad
+A12,48,6416,A75,A151,female,old,bad
+A12,24,4736,A72,A152,female,young,bad
+A14,12,930,A75,A152,male,old,good
+A11,24,2910,A74,A153,male,old,good
+A12,18,2278,A72,A152,female,old,bad
+A14,36,4463,A73,A152,male,old,bad
+A14,36,2394,A73,A152,female,young,good
+A14,24,2255,A72,A152,male,old,good
+A14,39,8588,A75,A152,male,old,good
+A12,24,5743,A72,A153,female,young,good
+A11,12,1200,A73,A151,female,young,good
+A14,30,1867,A75,A152,male,old,good
+A11,6,860,A75,A152,female,old,good
+A11,24,1024,A72,A152,male,old,bad
+A11,60,6836,A75,A152,male,old,bad
+A11,12,1680,A75,A152,male,old,good
+A14,24,1413,A73,A152,male,old,good
+A12,8,907,A72,A152,male,old,good
+A11,24,1193,A71,A151,female,old,bad
+A11,12,2577,A73,A152,male,old,good
+A14,24,3972,A74,A151,female,young,good
+A12,24,3758,A71,A151,female,young,good
+A14,18,2320,A71,A152,male,old,good
+A14,24,2463,A74,A152,male,old,good
+A13,6,1343,A75,A152,male,old,good
+A14,28,7824,A72,A151,male,old,good
+A14,21,3275,A75,A152,male,old,good
+A12,18,3244,A73,A152,female,old,good
+A14,27,8613,A73,A152,male,old,good
+A11,36,1842,A72,A152,female,old,bad
+A14,24,2611,A75,A152,male,old,good
+A12,9,2030,A74,A152,male,young,good
+A14,24,9277,A73,A153,male,old,good
+A14,12,1076,A73,A152,male,old,good
+A14,10,1546,A73,A152,male,old,good
+A11,12,1893,A73,A152,female,old,good
+A14,12,2141,A74,A152,male,old,good
+A14,36,7980,A72,A151,male,old,bad
+A14,12,1264,A75,A151,male,old,good
+A11,18,2249,A74,A152,male,old,good
+A11,48,6143,A75,A153,female,old,bad
+A11,21,3599,A74,A151,female,old,good
+A12,15,1537,A75,A152,male,old,good
+A14,36,10974,A71,A152,female,old,bad
+A12,18,4297,A75,A152,male,old,bad
+A11,18,1190,A71,A153,female,old,bad
+A11,12,1274,A72,A152,female,old,bad
+A14,9,2406,A71,A152,male,old,good
+A12,24,6403,A72,A152,male,old,good
+A14,10,1287,A75,A152,male,old,good
+A12,42,5954,A74,A152,female,old,good
+A11,6,1957,A74,A152,female,old,good
+A14,24,1552,A74,A152,male,old,good
+A14,24,937,A72,A152,male,old,good
+A14,24,2032,A75,A153,male,old,good
+A14,36,10875,A75,A152,male,old,good
+A12,24,5084,A75,A152,female,old,good
+A11,36,5179,A74,A152,male,old,bad
+A12,6,1050,A71,A152,male,old,good
+A13,36,3913,A73,A152,male,young,good
+A13,18,3049,A72,A152,female,old,good
+A11,36,8335,A75,A153,male,old,bad
+A11,12,1858,A72,A151,female,young,good
+A11,12,4843,A75,A151,male,old,bad
+A14,18,3780,A72,A152,male,old,good
+A12,12,2028,A73,A152,male,old,good
+A14,24,2284,A74,A152,male,old,good
+A12,15,2728,A74,A152,male,old,good
+A11,27,5293,A71,A152,male,old,bad
+A12,60,9157,A73,A153,male,old,good
+A14,18,1943,A72,A152,female,young,bad
+A11,24,2957,A75,A152,male,old,good
+A11,18,2389,A72,A152,female,old,good
+A11,18,1882,A73,A151,female,young,bad
+A12,30,1715,A73,A152,female,old,good
+A11,48,6331,A75,A153,male,old,bad
+A14,15,1316,A73,A152,male,old,good
+A11,6,14896,A75,A152,male,old,bad
+A14,30,3077,A75,A152,male,old,good
+A14,4,1544,A74,A152,male,old,good
+A12,12,1007,A73,A152,male,young,good
+A14,10,2069,A73,A152,male,old,good
+A12,18,12976,A71,A153,female,old,bad
+A11,18,3104,A74,A152,male,old,good
+A12,36,3990,A72,A152,female,old,good
+A11,12,3499,A73,A152,female,old,bad
+A12,9,1136,A75,A153,male,old,bad
+A13,15,392,A72,A151,female,young,good
+A11,18,1131,A71,A152,female,old,bad
+A12,12,1424,A74,A152,male,old,good
+A11,42,3394,A71,A152,male,old,good
+A11,48,6999,A74,A152,male,old,bad
+A11,30,11998,A72,A152,male,old,bad
+A14,30,2831,A73,A152,female,old,good
+A12,24,1965,A73,A151,female,old,good
+A14,24,1597,A75,A153,male,old,good
+A12,6,2063,A72,A151,male,old,good
+A12,26,7966,A72,A152,male,old,good
+A12,48,7582,A71,A153,male,old,good
+A14,24,4679,A74,A152,male,old,good
+A14,12,5801,A75,A151,male,old,good
+A14,24,3868,A75,A151,female,old,good
+A14,12,2390,A75,A152,male,old,good
+A12,15,6850,A71,A152,male,old,bad
+A14,9,1478,A74,A152,male,young,bad
+A14,6,2080,A73,A152,male,young,good
+A12,6,931,A72,A152,female,old,bad
+A14,9,1244,A75,A151,female,old,good
+A11,10,2132,A72,A151,female,old,good
+A14,18,1950,A74,A152,male,old,good
+A11,6,1374,A73,A152,male,old,good
+A12,6,1449,A75,A152,male,old,good
+A14,18,1582,A75,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-test.csv
new file mode 100644
index 0000000..37c539e
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-test.csv
@@ -0,0 +1,201 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A13,36,5848,A73,A152,male,young,good
+A11,36,2302,A73,A151,male,old,bad
+A12,12,1103,A74,A152,male,old,good
+A14,18,1530,A73,A152,male,old,bad
+A11,15,1721,A72,A152,male,old,good
+A14,18,2775,A74,A152,male,old,bad
+A13,12,1881,A73,A151,female,old,good
+A14,15,3029,A74,A152,male,old,good
+A11,12,1657,A73,A152,male,old,good
+A12,21,2353,A73,A152,male,old,good
+A11,18,2659,A73,A152,male,old,good
+A14,12,1592,A74,A152,female,old,good
+A13,24,1925,A73,A152,male,old,good
+A14,36,11054,A73,A152,male,old,good
+A12,24,1967,A75,A152,female,young,good
+A12,36,2671,A73,A153,female,old,bad
+A14,24,3430,A75,A152,male,old,good
+A13,36,4473,A75,A152,male,old,good
+A14,12,4675,A72,A151,female,young,good
+A14,24,999,A75,A152,male,young,good
+A12,48,10961,A74,A152,male,old,bad
+A12,12,1318,A75,A152,male,old,good
+A11,6,609,A74,A152,female,old,good
+A14,21,2993,A73,A152,male,old,good
+A14,60,15653,A74,A152,male,young,good
+A14,15,3368,A75,A151,male,young,good
+A11,48,4788,A74,A152,male,old,good
+A11,12,2246,A75,A152,male,old,bad
+A11,24,1747,A72,A152,male,young,good
+A12,48,5096,A73,A152,female,old,bad
+A12,18,884,A75,A152,male,old,bad
+A14,18,6070,A75,A152,male,old,good
+A11,24,1382,A74,A152,male,old,good
+A12,6,753,A73,A152,female,old,good
+A14,9,2697,A73,A152,male,old,good
+A12,6,590,A72,A152,male,old,good
+A12,30,1919,A72,A152,male,old,bad
+A11,21,1647,A73,A152,male,old,bad
+A12,18,5866,A73,A152,male,old,good
+A14,24,5943,A72,A152,female,old,bad
+A11,12,727,A72,A152,male,old,bad
+A11,24,2812,A75,A151,female,old,good
+A11,48,4308,A72,A151,female,young,bad
+A11,30,6187,A74,A151,male,young,good
+A12,18,1295,A72,A152,female,old,good
+A14,36,5711,A75,A152,male,old,good
+A14,48,6110,A73,A153,male,old,good
+A12,30,8386,A74,A152,male,old,bad
+A13,18,1961,A75,A152,female,young,good
+A12,12,1158,A73,A152,male,old,good
+A12,12,1410,A73,A152,male,old,good
+A14,33,7253,A74,A152,male,old,good
+A13,6,683,A72,A152,female,old,good
+A14,12,1413,A74,A152,male,old,good
+A11,36,5371,A73,A152,male,old,good
+A12,30,4280,A73,A151,female,old,bad
+A11,24,1358,A75,A152,male,old,bad
+A13,10,781,A75,A153,male,old,good
+A14,4,3380,A74,A152,female,old,good
+A12,9,1082,A75,A152,male,old,good
+A12,36,2337,A75,A152,male,old,good
+A14,7,846,A75,A153,male,old,good
+A12,48,8487,A74,A152,female,young,good
+A13,15,1905,A75,A151,male,old,good
+A12,12,1804,A72,A152,male,old,good
+A11,12,652,A75,A151,female,young,good
+A12,12,1331,A72,A152,male,young,bad
+A14,36,3079,A73,A152,male,old,good
+A13,15,2687,A74,A151,male,old,good
+A13,12,1330,A72,A152,male,old,good
+A11,12,1168,A73,A152,male,old,good
+A11,36,3446,A75,A152,male,old,bad
+A12,9,1199,A74,A152,female,old,good
+A11,12,741,A71,A152,female,young,bad
+A12,18,7374,A71,A152,male,old,good
+A14,18,2169,A73,A152,male,old,bad
+A14,6,1338,A73,A152,male,old,good
+A11,24,6419,A75,A153,female,old,good
+A11,15,975,A73,A152,male,young,good
+A14,24,2684,A73,A152,male,old,good
+A14,24,4526,A73,A152,male,old,good
+A12,12,951,A72,A151,female,old,bad
+A12,54,15945,A72,A151,male,old,bad
+A14,15,3556,A73,A152,male,old,good
+A14,4,1455,A74,A152,male,old,good
+A12,24,1935,A75,A152,male,old,bad
+A12,12,3124,A72,A152,male,old,good
+A14,6,1543,A73,A152,male,old,good
+A11,24,2964,A75,A153,male,old,good
+A11,18,1442,A74,A153,male,old,bad
+A14,60,13756,A75,A153,male,old,good
+A12,20,6148,A75,A152,male,old,good
+A14,24,6313,A75,A152,male,old,good
+A11,12,7865,A75,A153,male,old,bad
+A13,12,1297,A73,A151,male,young,good
+A11,18,2462,A73,A152,male,young,bad
+A14,18,6761,A73,A151,male,old,bad
+A14,36,9572,A72,A152,male,old,bad
+A14,36,7409,A75,A152,male,old,good
+A11,36,6887,A73,A152,male,old,bad
+A14,24,3621,A75,A152,male,old,bad
+A11,6,1352,A71,A151,female,young,good
+A13,12,1480,A71,A153,male,old,good
+A14,18,1028,A73,A152,female,old,good
+A11,12,902,A74,A151,male,young,bad
+A11,10,2315,A75,A152,male,old,good
+A14,9,1980,A72,A151,female,young,bad
+A12,24,1553,A74,A151,female,young,good
+A13,10,1240,A75,A153,female,old,bad
+A12,12,1223,A75,A151,male,old,bad
+A11,21,2606,A72,A151,female,old,good
+A11,12,1409,A75,A152,male,old,good
+A14,18,1568,A73,A151,female,young,good
+A14,5,3448,A74,A152,male,old,good
+A12,18,3872,A71,A152,female,old,good
+A12,12,2366,A74,A152,male,old,good
+A14,36,7678,A74,A152,female,old,good
+A12,36,9398,A72,A151,male,old,bad
+A14,36,9566,A73,A152,female,old,good
+A11,24,1938,A72,A152,male,old,bad
+A11,12,2149,A73,A153,male,old,bad
+A14,36,6614,A75,A152,male,old,good
+A14,12,2859,A71,A152,male,old,good
+A12,20,2629,A73,A152,male,old,good
+A12,18,6204,A73,A152,male,old,good
+A11,6,3676,A73,A151,male,old,good
+A12,15,2326,A73,A152,male,old,good
+A11,6,3384,A73,A151,male,old,bad
+A12,48,3060,A74,A152,male,old,bad
+A13,6,1299,A73,A152,male,old,good
+A14,10,1418,A73,A151,male,old,good
+A14,12,996,A74,A152,female,young,good
+A14,20,3485,A72,A152,male,old,good
+A14,18,3378,A73,A152,male,old,good
+A14,9,1224,A73,A152,male,old,good
+A13,9,745,A73,A152,female,old,bad
+A12,12,1860,A71,A152,male,old,good
+A11,20,4272,A75,A152,female,young,good
+A12,7,2415,A73,A152,male,old,good
+A14,24,3777,A73,A152,male,old,good
+A12,30,2181,A75,A152,male,old,good
+A14,18,4594,A72,A152,male,old,good
+A11,12,759,A74,A152,male,old,bad
+A14,30,4530,A74,A151,female,old,good
+A14,10,1924,A73,A152,male,old,good
+A14,12,3447,A73,A152,female,old,good
+A12,60,14027,A74,A152,male,old,bad
+A13,18,2864,A73,A152,male,old,bad
+A14,12,1255,A75,A152,male,old,good
+A11,42,4370,A74,A152,male,old,bad
+A14,12,926,A71,A152,female,old,good
+A12,15,802,A75,A152,male,old,bad
+A12,12,1092,A73,A152,female,old,good
+A14,21,1591,A74,A152,male,old,good
+A14,36,4454,A73,A152,female,old,good
+A14,12,682,A74,A152,female,old,good
+A12,18,6361,A75,A152,male,old,good
+A14,12,1258,A72,A151,female,young,good
+A12,27,2520,A73,A152,male,young,bad
+A11,6,338,A75,A152,male,old,good
+A12,20,7057,A74,A151,male,old,good
+A11,20,2235,A73,A151,male,old,bad
+A13,12,1424,A75,A152,female,old,good
+A12,24,11328,A73,A152,male,old,bad
+A14,24,1249,A72,A152,male,old,good
+A12,9,2118,A73,A152,male,old,good
+A11,24,1285,A74,A151,female,old,bad
+A14,48,3578,A75,A152,male,old,good
+A14,10,2848,A73,A152,male,old,good
+A11,18,3509,A74,A152,female,young,good
+A11,8,1164,A75,A153,male,old,good
+A12,21,3976,A74,A152,male,old,good
+A12,18,3612,A75,A152,female,old,good
+A12,72,5595,A73,A152,male,young,bad
+A11,12,709,A75,A152,male,old,bad
+A11,36,8229,A73,A152,male,old,bad
+A11,21,3414,A72,A152,male,old,bad
+A12,36,12612,A73,A153,male,old,bad
+A14,6,1750,A75,A152,male,old,good
+A14,12,3077,A73,A152,male,old,good
+A14,15,5324,A75,A153,female,old,good
+A13,4,1494,A72,A152,male,old,good
+A11,12,2578,A71,A153,female,old,good
+A12,7,2329,A72,A152,female,old,good
+A11,9,654,A73,A152,male,old,bad
+A14,36,8133,A73,A152,female,old,good
+A14,24,3863,A73,A153,male,old,good
+A12,12,7472,A71,A151,female,young,good
+A14,12,1555,A75,A153,male,old,bad
+A11,12,1262,A75,A152,male,old,good
+A12,9,790,A73,A152,female,old,good
+A12,30,5234,A71,A152,male,old,bad
+A11,30,2406,A74,A151,female,young,bad
+A12,12,2969,A72,A151,female,young,bad
+A14,21,12680,A75,A153,male,old,bad
+A11,12,1082,A73,A152,male,old,bad
+A12,27,3915,A73,A152,male,old,bad
+A14,9,3832,A75,A152,male,old,good
+A12,18,1928,A72,A152,male,old,bad
+A12,36,9857,A74,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-train.csv
new file mode 100644
index 0000000..1045c8d
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-train.csv
@@ -0,0 +1,501 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,36,3959,A71,A152,male,old,good
+A14,9,3577,A73,A151,male,old,good
+A14,18,2515,A73,A152,male,old,good
+A12,12,1995,A72,A152,male,old,good
+A14,60,10366,A75,A152,male,old,good
+A14,24,7393,A73,A152,male,old,good
+A13,12,409,A73,A151,female,old,good
+A14,9,2507,A75,A153,male,old,good
+A14,30,4811,A74,A151,female,young,good
+A12,48,6560,A74,A152,male,young,bad
+A12,36,9034,A72,A151,male,old,bad
+A11,24,3552,A74,A152,male,old,bad
+A12,20,6468,A71,A152,male,old,good
+A11,30,2522,A75,A152,male,old,good
+A12,36,2820,A72,A152,male,old,bad
+A14,6,783,A73,A152,male,old,good
+A11,21,3763,A74,A152,male,young,good
+A14,30,7485,A71,A152,female,old,bad
+A11,15,1403,A73,A151,female,old,good
+A12,11,4771,A74,A152,male,old,good
+A11,24,4020,A73,A152,male,old,good
+A12,24,1201,A72,A152,male,old,good
+A12,48,5381,A71,A153,male,old,good
+A14,24,5511,A73,A152,male,young,good
+A14,18,1098,A71,A152,female,old,good
+A12,48,6224,A75,A153,male,old,bad
+A14,11,1393,A72,A152,female,old,good
+A12,9,1919,A74,A151,male,old,good
+A14,15,5045,A75,A152,female,old,good
+A12,48,5951,A73,A152,female,young,bad
+A11,8,731,A75,A152,male,old,good
+A14,18,629,A75,A152,male,old,good
+A14,18,1808,A74,A152,female,young,bad
+A12,21,1188,A75,A152,female,old,bad
+A14,24,3235,A75,A152,male,old,good
+A12,24,11560,A73,A151,female,young,bad
+A12,9,1154,A75,A152,male,old,good
+A14,18,3850,A74,A152,male,old,good
+A11,24,2359,A71,A152,male,old,bad
+A14,24,1533,A72,A152,female,old,good
+A13,36,4210,A73,A152,male,old,bad
+A11,30,3857,A73,A152,male,old,good
+A14,12,1393,A75,A152,male,old,good
+A14,18,1800,A73,A152,male,young,good
+A13,12,2251,A73,A152,female,old,good
+A12,12,836,A72,A152,female,young,bad
+A11,18,1345,A73,A152,male,old,bad
+A12,36,4795,A72,A152,female,old,good
+A11,12,708,A73,A152,male,old,good
+A11,12,1289,A73,A152,male,young,good
+A14,18,1864,A73,A152,female,old,bad
+A12,48,9960,A72,A152,female,old,bad
+A11,18,5302,A75,A153,male,old,good
+A11,6,1872,A71,A153,male,old,good
+A11,24,2439,A72,A152,female,old,bad
+A14,9,1388,A73,A151,female,old,good
+A12,48,12169,A71,A153,male,old,good
+A14,18,2662,A74,A152,male,old,good
+A14,13,1409,A71,A152,female,old,good
+A11,15,3643,A75,A152,female,old,good
+A11,48,6758,A73,A152,female,old,bad
+A14,24,1851,A74,A152,male,old,good
+A11,27,3416,A73,A152,male,old,good
+A14,27,5117,A74,A152,male,old,good
+A14,10,1597,A73,A151,male,old,good
+A14,6,1382,A73,A152,female,old,good
+A14,10,1231,A75,A152,male,old,good
+A12,15,2631,A73,A151,female,old,bad
+A12,30,2150,A73,A152,female,young,bad
+A12,12,2930,A74,A152,female,old,good
+A14,6,6761,A74,A152,male,old,good
+A12,21,2745,A74,A152,male,old,good
+A11,24,2325,A74,A152,male,old,good
+A14,6,1204,A73,A151,male,old,good
+A11,18,1049,A72,A151,female,young,good
+A14,12,2292,A71,A152,male,old,bad
+A14,9,1449,A74,A152,female,old,good
+A11,18,7511,A75,A153,male,old,bad
+A12,9,918,A73,A152,female,old,bad
+A14,24,2346,A74,A152,male,old,good
+A11,24,1603,A75,A152,female,old,good
+A11,11,3905,A73,A151,male,old,good
+A14,18,1126,A72,A151,female,young,good
+A13,21,2923,A73,A152,female,old,good
+A12,9,3195,A73,A152,female,old,good
+A11,6,4716,A72,A152,male,old,good
+A12,6,14555,A71,A152,male,young,bad
+A11,36,5493,A75,A153,male,old,good
+A14,24,1585,A74,A152,male,old,good
+A11,12,3590,A73,A152,male,old,good
+A14,12,2759,A75,A152,male,old,good
+A11,12,2122,A73,A151,male,old,good
+A12,12,2002,A74,A151,male,old,good
+A14,24,1901,A73,A151,male,old,good
+A14,12,1493,A72,A152,female,old,good
+A12,6,1209,A71,A152,male,old,bad
+A12,48,3979,A74,A152,male,old,good
+A11,24,7721,A72,A152,female,old,good
+A11,18,3650,A72,A151,female,young,good
+A11,18,1880,A74,A152,male,old,good
+A12,18,1941,A73,A152,male,old,good
+A14,15,874,A72,A152,female,young,good
+A14,27,5190,A75,A152,male,old,good
+A14,36,10477,A75,A153,male,old,good
+A12,24,4351,A73,A152,female,old,good
+A14,24,1376,A74,A152,female,old,good
+A12,30,3496,A73,A152,male,old,good
+A11,40,5998,A73,A152,male,old,bad
+A11,28,4006,A73,A152,male,old,bad
+A14,36,3342,A75,A152,male,old,good
+A12,18,2622,A73,A152,male,old,good
+A12,36,2384,A72,A151,male,old,bad
+A11,24,1659,A72,A151,female,old,bad
+A14,24,3488,A74,A152,female,young,good
+A12,60,7418,A73,A152,male,old,good
+A14,18,2238,A73,A152,female,young,good
+A11,24,1333,A71,A153,male,old,bad
+A14,22,1283,A74,A151,female,young,good
+A14,6,3518,A73,A151,male,old,good
+A12,12,585,A73,A151,male,young,good
+A12,39,4933,A74,A152,male,young,bad
+A14,24,1278,A75,A152,male,old,good
+A14,12,1768,A73,A151,male,young,good
+A11,24,3349,A72,A153,male,old,bad
+A12,9,1391,A73,A152,male,old,good
+A14,18,3422,A75,A152,male,old,good
+A14,10,2146,A72,A151,female,young,good
+A14,15,1520,A75,A152,male,old,good
+A14,24,6842,A73,A152,male,old,good
+A14,18,2051,A72,A152,male,old,good
+A11,24,2996,A73,A152,male,young,bad
+A11,24,3660,A73,A152,female,old,good
+A11,24,1442,A74,A151,female,young,bad
+A11,9,1138,A73,A152,male,young,good
+A14,30,5771,A74,A152,female,young,good
+A11,18,2600,A73,A153,male,old,bad
+A13,24,3749,A72,A152,female,old,good
+A14,12,3059,A74,A152,male,old,good
+A11,6,662,A72,A152,male,old,good
+A11,18,1940,A72,A153,male,old,good
+A12,36,2323,A74,A151,male,young,good
+A12,48,12204,A73,A152,male,old,good
+A14,24,5507,A75,A153,male,old,good
+A14,18,433,A71,A151,female,young,bad
+A14,6,4611,A72,A152,female,old,bad
+A12,12,2762,A75,A152,female,young,bad
+A14,30,4272,A73,A152,male,old,good
+A12,39,11760,A74,A151,male,old,good
+A12,36,2225,A75,A153,male,old,bad
+A13,12,1474,A72,A152,female,old,good
+A14,12,1402,A74,A151,female,old,good
+A12,18,1913,A72,A152,male,old,good
+A14,14,802,A73,A152,male,old,good
+A13,15,1271,A73,A153,male,old,bad
+A13,24,1258,A73,A152,female,old,good
+A14,12,3565,A72,A152,male,old,good
+A11,12,3386,A75,A153,male,old,bad
+A13,12,2247,A73,A152,female,old,good
+A14,24,5804,A73,A152,male,old,good
+A14,15,960,A74,A152,female,old,good
+A12,18,1056,A75,A152,male,old,bad
+A13,42,6289,A72,A152,male,old,good
+A14,9,2753,A75,A152,male,old,good
+A14,10,2210,A73,A151,male,young,bad
+A11,48,10297,A74,A153,male,old,bad
+A12,36,14318,A75,A153,male,old,bad
+A12,30,4249,A71,A152,male,old,bad
+A14,24,929,A74,A152,male,old,good
+A12,9,959,A73,A152,female,old,bad
+A14,21,2782,A74,A152,female,old,good
+A12,48,8358,A72,A152,female,old,good
+A14,9,2301,A72,A151,female,young,good
+A12,48,3566,A74,A152,male,old,good
+A11,18,3966,A75,A151,female,old,bad
+A14,6,1237,A73,A152,female,old,good
+A11,18,976,A72,A152,female,young,bad
+A11,24,915,A75,A152,female,old,bad
+A13,24,947,A74,A153,male,old,bad
+A12,24,2825,A74,A152,male,old,good
+A14,21,2288,A72,A152,female,young,good
+A14,10,2901,A72,A151,female,old,good
+A12,45,4576,A71,A152,male,old,good
+A14,48,7238,A75,A152,male,old,good
+A11,36,2746,A75,A152,male,old,bad
+A12,24,1743,A75,A152,male,old,good
+A11,9,1366,A72,A151,female,young,bad
+A14,18,1505,A73,A153,male,old,good
+A11,6,343,A72,A152,female,old,good
+A14,24,2223,A75,A152,male,old,good
+A14,12,1495,A75,A152,male,old,good
+A13,15,2327,A72,A152,female,young,bad
+A11,45,11816,A75,A151,male,old,bad
+A12,12,1484,A73,A152,male,young,bad
+A14,36,9055,A73,A153,male,old,good
+A14,12,1655,A75,A152,male,old,good
+A14,18,1453,A72,A152,female,old,good
+A11,12,1228,A73,A152,female,young,bad
+A12,6,932,A74,A152,female,old,good
+A12,36,8086,A75,A152,male,old,bad
+A14,6,1346,A75,A153,male,old,good
+A12,36,2862,A75,A153,male,old,good
+A11,24,3123,A72,A152,female,old,bad
+A12,60,7408,A72,A152,female,young,bad
+A12,15,1264,A73,A151,male,young,bad
+A14,6,1554,A74,A151,female,young,good
+A12,21,3652,A74,A152,male,old,good
+A11,48,3931,A74,A153,male,old,bad
+A11,6,1361,A72,A152,male,old,good
+A11,6,1203,A75,A152,male,old,good
+A14,12,3527,A72,A152,male,old,good
+A14,36,7855,A73,A152,female,young,bad
+A11,21,3357,A72,A152,female,old,good
+A13,18,2100,A73,A152,male,old,bad
+A14,18,1169,A73,A152,male,old,good
+A14,36,8947,A74,A152,male,old,good
+A11,18,1936,A74,A151,male,young,good
+A12,24,3512,A74,A152,male,old,good
+A12,24,6967,A74,A151,male,old,good
+A14,30,2333,A75,A152,male,old,good
+A11,18,2039,A73,A151,female,young,bad
+A11,36,9271,A74,A152,male,young,bad
+A14,21,1572,A75,A152,female,old,good
+A12,6,454,A72,A152,male,young,good
+A14,36,6304,A75,A152,male,old,good
+A14,24,2197,A74,A152,male,old,good
+A14,48,7629,A75,A152,male,old,good
+A13,10,1275,A72,A152,female,young,good
+A11,48,7119,A73,A153,male,old,bad
+A12,24,12579,A75,A153,female,old,bad
+A12,18,2779,A73,A151,male,young,good
+A14,4,1503,A74,A152,male,old,good
+A12,8,1237,A73,A152,female,young,bad
+A14,48,3914,A73,A152,male,old,bad
+A14,24,3062,A75,A151,male,old,good
+A12,36,1953,A75,A153,male,old,bad
+A12,30,3832,A72,A152,male,young,good
+A14,6,1538,A72,A152,female,old,good
+A14,6,2108,A74,A151,male,old,good
+A11,6,428,A75,A152,female,old,good
+A12,36,4455,A73,A152,male,old,bad
+A13,21,2319,A72,A151,male,old,bad
+A14,12,1163,A73,A152,male,old,good
+A14,9,936,A75,A152,male,old,good
+A11,18,2473,A71,A152,male,young,bad
+A14,15,1569,A75,A152,male,old,good
+A12,12,3017,A72,A151,female,old,good
+A14,12,2331,A75,A152,male,old,good
+A14,12,2096,A74,A152,male,old,good
+A12,12,1037,A74,A152,male,old,good
+A14,6,1236,A73,A151,male,old,good
+A14,24,4151,A73,A152,male,old,good
+A14,48,3609,A73,A152,female,old,good
+A14,12,1412,A73,A152,female,old,good
+A11,11,3939,A73,A152,male,old,good
+A11,12,1526,A75,A153,male,old,good
+A14,22,2675,A75,A152,male,old,good
+A12,30,4221,A73,A152,female,old,good
+A11,24,2924,A73,A152,male,old,good
+A14,18,1817,A73,A152,female,old,good
+A13,10,1225,A73,A152,male,old,good
+A11,15,1845,A72,A151,female,old,good
+A14,6,250,A73,A152,female,old,good
+A14,6,660,A74,A151,male,young,good
+A13,6,709,A72,A152,male,old,good
+A11,36,15857,A71,A152,male,old,good
+A12,24,1837,A74,A153,female,old,bad
+A11,13,1797,A72,A152,male,old,good
+A11,6,448,A72,A152,female,young,bad
+A12,18,3213,A72,A151,male,young,good
+A11,12,2579,A72,A152,male,old,bad
+A11,6,1198,A75,A153,female,old,bad
+A14,21,5003,A73,A152,female,old,bad
+A12,24,2333,A72,A152,male,old,good
+A14,15,1262,A74,A152,male,old,good
+A12,12,1155,A75,A152,male,old,good
+A14,10,894,A74,A152,female,old,good
+A14,12,3331,A75,A152,male,old,good
+A12,18,1113,A73,A152,female,old,good
+A14,12,719,A75,A152,male,old,bad
+A12,24,2064,A71,A152,female,old,bad
+A13,30,3656,A75,A152,male,old,good
+A11,24,6615,A71,A153,male,old,good
+A14,18,3229,A71,A152,male,old,good
+A14,15,3343,A73,A153,male,old,good
+A11,30,3108,A72,A152,male,old,bad
+A12,9,458,A73,A152,male,young,good
+A14,15,1471,A73,A153,male,old,good
+A12,11,1322,A73,A152,female,old,good
+A11,9,1288,A75,A152,male,old,good
+A14,12,707,A73,A152,male,old,good
+A12,15,1308,A75,A152,male,old,good
+A14,9,3074,A73,A152,male,old,good
+A11,15,2511,A71,A151,female,young,good
+A14,6,1740,A75,A151,male,old,good
+A14,12,1123,A73,A151,female,old,bad
+A12,11,1577,A72,A152,female,young,good
+A11,45,1845,A73,A153,male,young,bad
+A14,24,2028,A74,A152,male,old,good
+A12,27,8318,A75,A153,female,old,bad
+A12,12,1295,A72,A151,female,young,bad
+A12,24,11938,A73,A152,male,old,bad
+A14,36,1819,A73,A153,male,old,bad
+A11,15,1478,A75,A152,male,old,good
+A14,24,1516,A73,A152,female,old,good
+A11,9,2136,A73,A152,male,young,good
+A14,15,3812,A72,A152,female,young,good
+A14,18,4165,A73,A152,male,old,bad
+A12,24,4057,A74,A152,male,old,bad
+A14,48,10127,A73,A153,male,old,bad
+A12,6,484,A74,A152,male,old,good
+A12,60,6288,A73,A153,male,old,bad
+A11,24,6872,A72,A152,male,old,bad
+A14,15,1532,A73,A152,female,old,good
+A14,9,2134,A73,A152,male,old,good
+A11,12,1107,A73,A151,male,young,good
+A11,20,2212,A74,A152,male,old,good
+A14,15,3186,A74,A151,female,young,good
+A11,47,10722,A72,A152,female,old,good
+A14,7,730,A75,A151,male,old,good
+A14,12,2073,A73,A152,female,old,good
+A11,21,571,A75,A152,male,old,good
+A13,15,2360,A73,A152,male,old,good
+A11,12,1344,A73,A152,male,old,good
+A11,42,7174,A74,A152,female,old,bad
+A13,42,4796,A75,A153,male,old,good
+A11,42,3965,A72,A152,male,old,bad
+A14,11,7228,A73,A152,male,old,good
+A12,24,1216,A72,A152,male,old,bad
+A14,12,2748,A75,A153,female,old,good
+A12,42,9283,A71,A153,male,old,good
+A14,27,4526,A72,A152,male,old,good
+A14,21,5248,A73,A152,male,old,good
+A14,24,3181,A72,A152,female,old,good
+A14,24,717,A75,A152,male,old,good
+A14,12,1262,A73,A152,male,young,good
+A14,12,1884,A75,A152,male,old,good
+A12,18,1924,A72,A151,female,old,bad
+A11,48,7476,A74,A153,male,old,good
+A13,12,939,A74,A152,male,old,bad
+A12,12,841,A74,A151,female,young,good
+A12,24,4113,A72,A151,female,old,bad
+A13,24,5152,A74,A152,male,young,good
+A11,33,4281,A73,A152,female,young,bad
+A12,12,1567,A73,A152,female,young,good
+A11,12,339,A75,A152,male,old,good
+A11,15,1275,A73,A151,female,young,bad
+A12,12,639,A73,A152,male,old,bad
+A14,36,2299,A75,A152,male,old,good
+A11,24,1823,A71,A152,male,old,bad
+A11,48,7763,A75,A153,male,old,bad
+A14,6,1595,A74,A152,male,old,good
+A11,12,385,A74,A152,female,old,good
+A14,12,776,A73,A152,male,old,good
+A12,24,3069,A75,A153,male,old,good
+A11,9,1364,A74,A152,male,old,good
+A14,6,1221,A73,A152,male,old,good
+A11,8,3398,A74,A152,male,old,good
+A11,18,4153,A73,A152,male,old,bad
+A12,15,1778,A72,A151,female,old,bad
+A14,9,1236,A72,A151,female,young,good
+A14,6,1766,A73,A151,male,young,good
+A11,30,8072,A72,A152,male,young,good
+A12,9,5129,A75,A153,female,old,bad
+A14,24,8648,A72,A152,male,old,bad
+A11,24,4169,A73,A152,male,old,good
+A14,12,1935,A75,A152,male,old,good
+A14,24,2397,A75,A152,male,old,bad
+A14,15,1459,A73,A152,female,old,good
+A12,12,685,A74,A152,male,young,bad
+A11,48,4605,A75,A153,male,young,bad
+A14,11,2142,A75,A152,male,old,good
+A11,15,950,A75,A151,male,old,bad
+A12,16,1175,A71,A153,male,old,good
+A11,12,1498,A73,A152,female,young,good
+A14,18,1984,A73,A153,male,old,good
+A12,36,3804,A73,A152,female,old,bad
+A11,24,1987,A73,A151,male,young,bad
+A14,24,4139,A73,A152,male,old,good
+A14,12,701,A73,A152,male,old,good
+A14,54,9436,A73,A152,male,old,good
+A11,18,3190,A73,A152,female,young,bad
+A12,13,882,A72,A152,male,young,good
+A12,24,2760,A75,A153,male,old,good
+A14,15,2221,A73,A151,female,young,good
+A11,12,626,A73,A152,female,young,bad
+A12,48,18424,A73,A152,female,old,bad
+A11,18,1217,A73,A152,male,old,bad
+A12,7,2576,A73,A152,male,old,good
+A11,6,2647,A73,A151,male,old,good
+A14,27,2570,A73,A151,female,young,bad
+A13,24,3148,A73,A152,male,old,good
+A11,6,666,A74,A152,female,old,good
+A14,12,1185,A73,A152,female,old,good
+A14,30,5954,A74,A152,male,old,good
+A12,45,4746,A72,A152,male,young,bad
+A14,15,1213,A75,A152,male,old,good
+A14,6,1898,A73,A152,male,old,good
+A14,48,11590,A73,A151,female,young,bad
+A14,4,601,A72,A151,female,young,good
+A12,18,1245,A73,A152,male,old,bad
+A14,36,5742,A74,A152,male,old,good
+A12,14,1410,A75,A152,male,old,good
+A11,12,900,A73,A152,male,young,bad
+A14,6,362,A73,A152,female,old,good
+A14,15,1979,A75,A152,male,old,good
+A14,36,3349,A73,A152,female,old,bad
+A14,12,797,A75,A152,female,old,bad
+A14,6,518,A73,A152,female,old,good
+A11,24,2303,A75,A152,male,old,bad
+A11,36,2712,A75,A152,male,old,bad
+A14,24,1474,A72,A152,male,old,good
+A14,18,1820,A73,A152,male,old,good
+A14,24,1927,A73,A152,female,old,good
+A11,15,1433,A73,A151,female,young,good
+A12,12,958,A74,A152,male,old,good
+A14,36,3835,A75,A152,female,old,good
+A14,12,763,A73,A152,female,old,good
+A14,6,2978,A73,A152,male,old,good
+A14,12,1291,A73,A152,female,old,good
+A11,12,674,A74,A152,male,young,bad
+A14,60,6527,A73,A153,male,old,good
+A14,12,976,A75,A152,male,old,good
+A13,10,3949,A72,A152,male,old,good
+A12,10,1048,A73,A152,male,young,good
+A14,24,2670,A75,A152,male,old,good
+A11,30,4583,A73,A152,male,old,good
+A14,24,2375,A73,A152,male,old,good
+A11,12,1282,A73,A151,female,young,bad
+A14,24,3757,A75,A153,female,old,good
+A12,24,4241,A73,A152,male,old,bad
+A11,24,6579,A71,A153,male,old,good
+A11,12,2121,A73,A152,male,old,good
+A14,12,2171,A72,A152,female,old,good
+A14,15,3594,A72,A152,female,old,good
+A14,36,4686,A73,A153,male,old,good
+A14,15,4657,A73,A152,male,old,good
+A11,14,8978,A75,A152,male,old,bad
+A14,18,1055,A72,A152,female,old,good
+A11,48,3051,A73,A152,male,old,bad
+A12,30,3441,A73,A151,female,young,bad
+A14,24,5150,A75,A152,male,old,good
+A14,12,1963,A74,A151,male,old,good
+A12,18,3590,A71,A152,male,old,good
+A14,36,5842,A75,A152,male,old,good
+A14,24,1469,A75,A151,male,old,good
+A14,24,2603,A73,A151,female,old,good
+A11,12,3651,A73,A152,male,old,good
+A11,24,3149,A72,A153,male,young,good
+A11,24,4110,A75,A151,male,young,bad
+A12,9,1437,A74,A152,male,old,bad
+A13,24,1377,A75,A153,female,old,good
+A11,9,1422,A72,A153,male,old,bad
+A11,30,3622,A75,A151,female,old,good
+A11,24,3345,A75,A151,male,old,bad
+A12,12,625,A72,A152,male,old,good
+A12,12,983,A72,A151,female,young,good
+A12,30,2991,A75,A152,female,young,good
+A14,18,1149,A73,A152,male,old,good
+A12,9,276,A73,A151,male,young,good
+A11,24,1371,A73,A151,female,young,bad
+A11,12,795,A72,A152,female,old,bad
+A11,24,1282,A73,A152,female,old,bad
+A12,6,368,A75,A152,male,old,good
+A14,24,1258,A74,A152,male,young,good
+A14,15,2186,A74,A151,female,old,good
+A14,24,2835,A75,A152,male,old,good
+A11,24,3632,A73,A151,female,young,good
+A14,24,1311,A74,A152,male,old,good
+A11,36,2348,A73,A152,male,old,good
+A12,6,433,A72,A151,female,young,bad
+A11,15,3959,A73,A152,female,old,bad
+A12,15,2631,A73,A152,female,young,good
+A14,12,2445,A72,A151,male,old,good
+A11,10,1038,A74,A152,male,old,good
+A14,48,8858,A74,A153,male,old,good
+A13,24,1275,A73,A152,male,old,good
+A11,24,6568,A73,A152,male,young,good
+A13,24,1344,A74,A152,male,old,bad
+A12,27,5965,A75,A152,male,old,good
+A14,12,618,A75,A152,male,old,good
+A14,10,727,A75,A153,male,old,good
+A12,15,1444,A72,A152,male,young,good
+A11,36,3620,A73,A152,male,old,good
+A11,10,2241,A72,A151,male,old,good
+A12,60,14782,A75,A153,female,old,bad
+A14,24,2022,A73,A152,female,old,good
+A13,12,609,A72,A152,female,old,bad
+A11,24,3021,A73,A151,male,young,good
+A11,18,4380,A73,A152,male,old,good
+A14,10,1309,A73,A152,male,old,bad
+A11,24,2828,A73,A152,male,young,good
+A12,18,1795,A75,A151,female,old,good
+A14,18,1533,A72,A152,male,old,bad
+A14,24,5103,A72,A153,male,old,good
+A11,30,10623,A75,A153,male,old,good
+A14,12,1386,A73,A152,female,old,bad
+A14,24,4591,A73,A152,male,old,bad
+A13,12,3399,A75,A152,male,old,good
+A13,6,1047,A73,A152,female,old,good
+A11,48,7685,A74,A151,female,old,bad
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-val.csv
new file mode 100644
index 0000000..e0b67a7
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI-val.csv
@@ -0,0 +1,301 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,18,1553,A73,A152,male,old,bad
+A14,15,1360,A73,A152,male,old,good
+A12,8,760,A74,A152,female,old,good
+A12,18,2899,A75,A152,male,old,good
+A12,45,3031,A73,A151,male,young,bad
+A12,24,4712,A73,A152,male,old,good
+A12,18,1887,A73,A152,male,old,good
+A12,36,2273,A74,A152,male,old,good
+A12,12,888,A75,A152,male,old,bad
+A11,12,1620,A73,A152,female,old,good
+A14,15,2708,A72,A152,male,old,good
+A12,18,866,A73,A152,male,young,good
+A12,8,1414,A73,A152,male,old,good
+A14,9,1313,A75,A152,male,young,good
+A14,24,2578,A75,A152,male,old,good
+A14,15,3568,A75,A151,female,old,good
+A14,24,1393,A73,A152,male,old,good
+A13,6,2116,A73,A152,male,old,good
+A14,12,2012,A74,A152,female,old,good
+A14,24,1940,A75,A152,male,old,good
+A13,6,1323,A75,A152,male,old,good
+A11,36,1977,A75,A152,male,old,bad
+A11,12,1108,A74,A152,male,old,bad
+A14,15,1300,A75,A153,male,old,good
+A11,6,1169,A75,A152,male,old,good
+A11,36,3249,A74,A153,male,old,good
+A14,39,2569,A73,A152,male,young,good
+A12,9,1549,A72,A152,male,old,good
+A14,6,700,A75,A153,male,old,good
+A12,12,6078,A74,A152,male,old,good
+A12,9,1206,A75,A152,female,young,good
+A12,24,1355,A72,A152,female,young,bad
+A14,24,2978,A73,A152,male,old,good
+A14,24,1525,A74,A152,female,old,good
+A14,6,932,A73,A152,female,young,good
+A12,27,2528,A72,A152,female,old,good
+A11,18,3114,A72,A151,female,old,bad
+A13,30,1908,A75,A152,male,old,bad
+A14,12,1240,A75,A152,female,old,good
+A14,12,717,A75,A152,male,old,good
+A12,15,1514,A73,A152,male,young,good
+A14,15,1829,A75,A152,male,old,good
+A14,12,640,A73,A152,male,old,good
+A12,36,7432,A73,A151,female,old,good
+A14,12,1542,A74,A152,male,old,good
+A11,36,9629,A74,A152,male,young,bad
+A14,6,426,A75,A152,male,old,good
+A14,15,4623,A73,A152,male,old,bad
+A11,18,8471,A73,A151,female,young,good
+A12,36,3711,A73,A152,male,old,good
+A14,48,12749,A74,A152,male,old,good
+A11,42,7882,A74,A153,male,old,good
+A14,60,10144,A74,A152,female,young,good
+A12,24,1246,A72,A152,male,young,bad
+A14,36,3595,A75,A152,male,old,good
+A12,12,3617,A75,A151,male,old,good
+A14,12,886,A73,A152,female,young,good
+A11,24,1546,A74,A151,male,young,bad
+A11,12,2171,A73,A152,male,old,good
+A14,36,2613,A73,A152,male,old,good
+A14,12,1101,A73,A152,male,old,good
+A12,18,3001,A74,A151,female,old,good
+A14,12,2279,A73,A153,male,old,good
+A12,10,1521,A73,A152,male,old,good
+A12,18,1042,A73,A152,female,old,bad
+A12,48,6681,A73,A153,male,old,good
+A11,36,2145,A74,A152,male,young,bad
+A11,24,4817,A74,A152,male,old,bad
+A12,48,14421,A73,A152,male,young,bad
+A11,12,691,A75,A152,male,old,bad
+A12,18,1239,A73,A153,male,old,good
+A12,21,2767,A75,A151,male,old,bad
+A12,18,6260,A74,A151,male,old,good
+A11,18,750,A71,A152,female,old,bad
+A14,12,1503,A73,A151,male,old,good
+A11,12,701,A73,A152,male,old,good
+A14,11,1154,A71,A152,female,old,good
+A14,21,3160,A75,A152,male,old,good
+A11,18,2124,A73,A151,female,young,bad
+A12,12,766,A73,A152,male,old,bad
+A12,36,6948,A73,A151,male,old,good
+A13,9,1126,A75,A152,male,old,good
+A11,18,1216,A72,A151,female,young,bad
+A11,12,2214,A73,A152,male,young,good
+A14,24,4042,A74,A152,male,old,good
+A11,15,1053,A72,A152,male,old,good
+A12,24,2896,A72,A152,male,old,good
+A11,12,6199,A73,A151,male,old,bad
+A12,24,7758,A75,A151,female,old,good
+A11,12,1372,A74,A152,male,old,bad
+A11,30,6350,A75,A152,male,old,bad
+A12,18,4439,A75,A152,male,old,good
+A12,48,15672,A73,A152,male,young,bad
+A14,12,804,A75,A152,male,old,good
+A11,21,1602,A75,A152,male,old,good
+A12,12,6468,A71,A152,male,old,bad
+A14,18,1473,A72,A152,male,old,good
+A14,24,5433,A71,A151,female,old,good
+A14,6,1238,A71,A152,male,old,good
+A14,33,2764,A73,A152,female,old,good
+A12,12,1922,A73,A152,male,old,bad
+A13,30,3017,A75,A152,male,old,good
+A11,24,1207,A72,A151,female,young,bad
+A11,16,2625,A75,A151,male,old,bad
+A11,12,684,A73,A151,male,old,bad
+A12,9,1670,A72,A152,female,young,bad
+A14,6,672,A71,A152,female,old,good
+A14,21,2580,A72,A152,male,old,bad
+A14,24,6314,A71,A152,male,old,good
+A11,24,3234,A72,A151,female,young,bad
+A12,10,7308,A71,A153,male,old,good
+A12,13,2101,A72,A152,female,young,good
+A11,36,6229,A72,A151,female,young,bad
+A13,12,3016,A73,A152,male,young,good
+A14,12,522,A75,A152,male,old,good
+A14,12,1934,A75,A152,male,old,good
+A14,48,2751,A75,A152,male,old,good
+A12,15,1512,A73,A152,male,old,bad
+A12,30,2503,A75,A152,male,old,good
+A14,24,1287,A75,A152,female,old,good
+A12,36,12389,A73,A153,male,old,bad
+A13,9,1337,A72,A152,male,old,bad
+A11,27,2442,A75,A152,male,old,good
+A14,10,1364,A73,A152,female,old,good
+A11,24,2384,A75,A151,male,old,good
+A11,9,2799,A73,A151,male,old,good
+A13,18,1445,A74,A152,male,old,good
+A14,6,1743,A73,A152,male,old,good
+A13,10,1347,A74,A152,male,old,good
+A11,24,4870,A73,A153,male,old,bad
+A14,12,1736,A74,A152,female,old,good
+A12,12,754,A75,A152,male,old,good
+A14,36,3535,A74,A152,male,old,good
+A14,15,1478,A73,A152,male,old,good
+A14,30,6742,A74,A152,male,old,good
+A14,36,7127,A72,A151,female,young,bad
+A11,15,806,A73,A152,female,young,good
+A13,24,2892,A75,A153,male,old,good
+A12,36,5800,A73,A152,male,old,good
+A14,24,3105,A72,A152,male,young,good
+A14,15,1386,A73,A151,male,old,good
+A14,18,2404,A73,A152,female,old,good
+A12,24,3878,A72,A152,male,old,good
+A12,24,3092,A72,A151,male,young,bad
+A14,28,2743,A75,A152,male,old,good
+A14,12,1574,A73,A152,male,old,good
+A11,21,1835,A73,A152,female,young,bad
+A12,24,2718,A73,A151,female,young,bad
+A14,48,4844,A71,A151,male,old,bad
+A11,24,1199,A75,A152,male,old,bad
+A11,14,3973,A71,A153,male,young,good
+A14,24,2679,A72,A152,female,old,good
+A14,12,2133,A75,A153,female,old,good
+A11,24,1231,A75,A151,female,old,good
+A14,24,2872,A75,A152,male,old,good
+A12,18,2427,A75,A152,male,old,good
+A11,60,7297,A75,A151,male,old,bad
+A14,30,7596,A75,A152,male,old,good
+A11,39,14179,A74,A152,male,old,good
+A12,9,1501,A75,A152,female,old,bad
+A12,48,3844,A74,A153,male,old,bad
+A14,42,4042,A73,A152,male,old,good
+A11,6,1374,A71,A152,female,old,good
+A14,24,7814,A74,A152,male,old,good
+A14,36,909,A75,A152,male,old,good
+A14,18,6458,A75,A152,male,old,bad
+A14,24,2424,A75,A152,male,old,good
+A14,15,2788,A74,A152,female,young,good
+A11,12,697,A72,A152,male,old,bad
+A14,21,2476,A75,A152,male,old,good
+A12,12,1534,A72,A151,male,young,bad
+A12,6,1068,A75,A152,male,old,good
+A12,12,3573,A73,A152,female,young,good
+A12,18,1301,A75,A152,male,old,good
+A14,42,7166,A74,A151,male,old,good
+A12,24,2039,A72,A152,male,young,bad
+A11,24,2483,A73,A152,male,young,good
+A13,24,3617,A75,A151,male,young,good
+A14,24,2058,A73,A152,male,old,good
+A11,24,1755,A75,A152,female,old,good
+A14,24,1559,A74,A152,male,old,good
+A11,24,3161,A73,A151,male,old,bad
+A14,48,10222,A74,A152,male,old,good
+A11,24,1381,A73,A152,female,old,bad
+A14,21,2241,A75,A152,male,old,good
+A14,24,2538,A75,A152,male,old,bad
+A11,36,8065,A73,A152,female,young,bad
+A12,48,6416,A75,A151,female,old,bad
+A12,24,4736,A72,A152,female,young,bad
+A14,12,930,A75,A152,male,old,good
+A11,24,2910,A74,A153,male,old,good
+A12,18,2278,A72,A152,female,old,bad
+A14,36,4463,A73,A152,male,old,bad
+A14,36,2394,A73,A152,female,young,good
+A14,24,2255,A72,A152,male,old,good
+A14,39,8588,A75,A152,male,old,good
+A12,24,5743,A72,A153,female,young,good
+A11,12,1200,A73,A151,female,young,good
+A14,30,1867,A75,A152,male,old,good
+A11,6,860,A75,A152,female,old,good
+A11,24,1024,A72,A152,male,old,bad
+A11,60,6836,A75,A152,male,old,bad
+A11,12,1680,A75,A152,male,old,good
+A14,24,1413,A73,A152,male,old,good
+A12,8,907,A72,A152,male,old,good
+A11,24,1193,A71,A151,female,old,bad
+A11,12,2577,A73,A152,male,old,good
+A14,24,3972,A74,A151,female,young,good
+A12,24,3758,A71,A151,female,young,good
+A14,18,2320,A71,A152,male,old,good
+A14,24,2463,A74,A152,male,old,good
+A13,6,1343,A75,A152,male,old,good
+A14,28,7824,A72,A151,male,old,good
+A14,21,3275,A75,A152,male,old,good
+A12,18,3244,A73,A152,female,old,good
+A14,27,8613,A73,A152,male,old,good
+A11,36,1842,A72,A152,female,old,bad
+A14,24,2611,A75,A152,male,old,good
+A12,9,2030,A74,A152,male,young,good
+A14,24,9277,A73,A153,male,old,good
+A14,12,1076,A73,A152,male,old,good
+A14,10,1546,A73,A152,male,old,good
+A11,12,1893,A73,A152,female,old,good
+A14,12,2141,A74,A152,male,old,good
+A14,36,7980,A72,A151,male,old,bad
+A14,12,1264,A75,A151,male,old,good
+A11,18,2249,A74,A152,male,old,good
+A11,48,6143,A75,A153,female,old,bad
+A11,21,3599,A74,A151,female,old,good
+A12,15,1537,A75,A152,male,old,good
+A14,36,10974,A71,A152,female,old,bad
+A12,18,4297,A75,A152,male,old,bad
+A11,18,1190,A71,A153,female,old,bad
+A11,12,1274,A72,A152,female,old,bad
+A14,9,2406,A71,A152,male,old,good
+A12,24,6403,A72,A152,male,old,good
+A14,10,1287,A75,A152,male,old,good
+A12,42,5954,A74,A152,female,old,good
+A11,6,1957,A74,A152,female,old,good
+A14,24,1552,A74,A152,male,old,good
+A14,24,937,A72,A152,male,old,good
+A14,24,2032,A75,A153,male,old,good
+A14,36,10875,A75,A152,male,old,good
+A12,24,5084,A75,A152,female,old,good
+A11,36,5179,A74,A152,male,old,bad
+A12,6,1050,A71,A152,male,old,good
+A13,36,3913,A73,A152,male,young,good
+A13,18,3049,A72,A152,female,old,good
+A11,36,8335,A75,A153,male,old,bad
+A11,12,1858,A72,A151,female,young,good
+A11,12,4843,A75,A151,male,old,bad
+A14,18,3780,A72,A152,male,old,good
+A12,12,2028,A73,A152,male,old,good
+A14,24,2284,A74,A152,male,old,good
+A12,15,2728,A74,A152,male,old,good
+A11,27,5293,A71,A152,male,old,bad
+A12,60,9157,A73,A153,male,old,good
+A14,18,1943,A72,A152,female,young,bad
+A11,24,2957,A75,A152,male,old,good
+A11,18,2389,A72,A152,female,old,good
+A11,18,1882,A73,A151,female,young,bad
+A12,30,1715,A73,A152,female,old,good
+A11,48,6331,A75,A153,male,old,bad
+A14,15,1316,A73,A152,male,old,good
+A11,6,14896,A75,A152,male,old,bad
+A14,30,3077,A75,A152,male,old,good
+A14,4,1544,A74,A152,male,old,good
+A12,12,1007,A73,A152,male,young,good
+A14,10,2069,A73,A152,male,old,good
+A12,18,12976,A71,A153,female,old,bad
+A11,18,3104,A74,A152,male,old,good
+A12,36,3990,A72,A152,female,old,good
+A11,12,3499,A73,A152,female,old,bad
+A12,9,1136,A75,A153,male,old,bad
+A13,15,392,A72,A151,female,young,good
+A11,18,1131,A71,A152,female,old,bad
+A12,12,1424,A74,A152,male,old,good
+A11,42,3394,A71,A152,male,old,good
+A11,48,6999,A74,A152,male,old,bad
+A11,30,11998,A72,A152,male,old,bad
+A14,30,2831,A73,A152,female,old,good
+A12,24,1965,A73,A151,female,old,good
+A14,24,1597,A75,A153,male,old,good
+A12,6,2063,A72,A151,male,old,good
+A12,26,7966,A72,A152,male,old,good
+A12,48,7582,A71,A153,male,old,good
+A14,24,4679,A74,A152,male,old,good
+A14,12,5801,A75,A151,male,old,good
+A14,24,3868,A75,A151,female,old,good
+A14,12,2390,A75,A152,male,old,good
+A12,15,6850,A71,A152,male,old,bad
+A14,9,1478,A74,A152,male,young,bad
+A14,6,2080,A73,A152,male,young,good
+A12,6,931,A72,A152,female,old,bad
+A14,9,1244,A75,A151,female,old,good
+A11,10,2132,A72,A151,female,old,good
+A14,18,1950,A74,A152,male,old,good
+A11,6,1374,A73,A152,male,old,good
+A12,6,1449,A75,A152,male,old,good
+A14,18,1582,A75,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-test.csv
new file mode 100644
index 0000000..1b64b92
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-test.csv
@@ -0,0 +1,201 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A13,0.5714285714285714,0.30802244965335096,A73,A152,male,young,good
+A11,0.5714285714285714,0.11290855067679102,A73,A151,male,old,bad
+A12,0.14285714285714285,0.04693518212831517,A74,A152,male,old,good
+A14,0.24999999999999997,0.0704302850225597,A73,A152,male,old,bad
+A11,0.19642857142857142,0.08093980411576977,A72,A152,male,old,good
+A14,0.24999999999999997,0.1389347419390338,A74,A152,male,old,bad
+A13,0.14285714285714285,0.08974358974358974,A73,A151,female,old,good
+A14,0.19642857142857142,0.15291075162319798,A74,A152,male,old,good
+A11,0.14285714285714285,0.07741828986464179,A73,A152,male,old,good
+A12,0.3035714285714286,0.11571475734565863,A73,A152,male,old,good
+A11,0.24999999999999997,0.13255199735886433,A73,A152,male,old,good
+A14,0.14285714285714285,0.07384175195333993,A74,A152,female,old,good
+A13,0.3571428571428571,0.09216463079124024,A73,A152,male,old,good
+A14,0.5714285714285714,0.594475624518543,A73,A152,male,old,good
+A12,0.3571428571428571,0.09447562451854297,A75,A152,female,young,good
+A12,0.5714285714285714,0.13321228128095083,A73,A153,female,old,bad
+A14,0.3571428571428571,0.17497523935292178,A75,A152,male,old,good
+A13,0.5714285714285714,0.23236491691427316,A75,A152,male,old,good
+A14,0.14285714285714285,0.24347969626939586,A72,A151,female,young,good
+A14,0.3571428571428571,0.041212721470232194,A75,A152,male,young,good
+A12,0.7857142857142857,0.5893584241223726,A74,A152,male,old,bad
+A12,0.14285714285714285,0.05876526906569825,A75,A152,male,old,good
+A11,0.03571428571428571,0.019753494002421042,A74,A152,female,old,good
+A14,0.3035714285714286,0.1509298998569385,A73,A152,male,old,good
+A14,1.0,0.847529437658193,A74,A152,male,young,good
+A14,0.19642857142857142,0.17156377242214152,A75,A151,male,young,good
+A11,0.7857142857142857,0.24969736986904373,A74,A152,male,old,good
+A11,0.14285714285714285,0.10982722570705403,A75,A152,male,old,bad
+A11,0.3571428571428571,0.08237041928029053,A72,A152,male,young,good
+A12,0.7857142857142857,0.2666446572025971,A73,A152,female,old,bad
+A12,0.24999999999999997,0.034885000550236606,A75,A152,male,old,bad
+A14,0.24999999999999997,0.32023770221195114,A75,A152,male,old,good
+A11,0.3571428571428571,0.06228678331682623,A74,A152,male,old,good
+A12,0.03571428571428571,0.02767690106745901,A73,A152,female,old,good
+A14,0.08928571428571427,0.13464289644547156,A73,A152,male,old,good
+A12,0.03571428571428571,0.01870804445911742,A72,A152,male,old,good
+A12,0.4642857142857143,0.09183448883019699,A72,A152,male,old,bad
+A11,0.3035714285714286,0.07686805326290305,A73,A152,male,old,bad
+A12,0.24999999999999997,0.3090128755364807,A73,A152,male,old,good
+A14,0.3571428571428571,0.3132496973698691,A72,A152,female,old,bad
+A11,0.14285714285714285,0.026246285902938263,A72,A152,male,old,bad
+A11,0.3571428571428571,0.14097061736546715,A75,A151,female,old,good
+A11,0.7857142857142857,0.22328601298558381,A72,A151,female,young,bad
+A11,0.4642857142857143,0.3266754704522945,A74,A151,male,young,good
+A12,0.24999999999999997,0.05749972488169913,A72,A152,female,old,good
+A14,0.5714285714285714,0.3004842082095301,A75,A152,male,old,good
+A14,0.7857142857142857,0.32243864861890614,A73,A153,male,old,good
+A12,0.4642857142857143,0.44767249917464513,A74,A152,male,old,bad
+A13,0.24999999999999997,0.09414548255749973,A75,A152,female,young,good
+A12,0.14285714285714285,0.04996148343787829,A73,A152,male,old,good
+A12,0.14285714285714285,0.06382744580169472,A73,A152,male,old,good
+A14,0.5178571428571429,0.385330692197645,A74,A152,male,old,good
+A13,0.03571428571428571,0.023825244855287777,A72,A152,female,old,good
+A14,0.14285714285714285,0.06399251678221635,A74,A152,male,old,good
+A11,0.5714285714285714,0.2817761637504127,A73,A152,male,old,good
+A12,0.4642857142857143,0.2217453505007153,A73,A151,female,old,bad
+A11,0.3571428571428571,0.06096621547265324,A75,A152,male,old,bad
+A13,0.10714285714285712,0.0292175635523275,A75,A153,male,old,good
+A14,0.0,0.17222405634422802,A74,A152,female,old,good
+A12,0.08928571428571427,0.045779685264663805,A75,A152,male,old,good
+A12,0.5714285714285714,0.11483437878287663,A75,A152,male,old,good
+A14,0.053571428571428575,0.03279410146362936,A75,A153,male,old,good
+A12,0.7857142857142857,0.45322988885220644,A74,A152,female,young,good
+A13,0.19642857142857142,0.09106415758776273,A75,A151,male,old,good
+A12,0.14285714285714285,0.08550676791020138,A72,A152,male,old,good
+A11,0.14285714285714285,0.022119511389897654,A75,A151,female,young,good
+A12,0.14285714285714285,0.059480576647958625,A72,A152,male,young,bad
+A14,0.5714285714285714,0.15566193463189174,A73,A152,male,old,good
+A13,0.19642857142857142,0.1340926598437328,A74,A151,male,old,good
+A13,0.14285714285714285,0.05942555298778475,A72,A152,male,old,good
+A11,0.14285714285714285,0.05051172003961703,A73,A152,male,old,good
+A11,0.5714285714285714,0.17585561791570375,A75,A152,male,old,bad
+A12,0.08928571428571427,0.052217453505007144,A74,A152,female,old,good
+A11,0.14285714285714285,0.02701661714537251,A71,A152,female,young,bad
+A12,0.24999999999999997,0.39198855507868385,A71,A152,male,old,good
+A14,0.24999999999999997,0.10559040387366568,A73,A152,male,old,bad
+A14,0.03571428571428571,0.05986574226917574,A73,A152,male,old,good
+A11,0.3571428571428571,0.33944095961263343,A75,A153,female,old,good
+A11,0.19642857142857142,0.039892153626059204,A73,A152,male,young,good
+A14,0.3571428571428571,0.13392758886321118,A73,A152,male,old,good
+A14,0.3571428571428571,0.23528117090348852,A73,A152,male,old,good
+A12,0.14285714285714285,0.038571585781886214,A72,A151,female,old,bad
+A12,0.8928571428571428,0.8635963464289644,A72,A151,male,old,bad
+A14,0.19642857142857142,0.18190822053483,A73,A152,male,old,good
+A14,0.0,0.06630351050951909,A74,A152,male,old,good
+A12,0.3571428571428571,0.09271486739297898,A75,A152,male,old,bad
+A12,0.14285714285714285,0.1581379993397161,A72,A152,male,old,good
+A14,0.03571428571428571,0.07114559260482008,A73,A152,male,old,good
+A11,0.3571428571428571,0.14933421371189612,A75,A153,male,old,good
+A11,0.24999999999999997,0.06558820292725871,A74,A153,male,old,bad
+A14,1.0,0.7431495543083525,A75,A153,male,old,good
+A12,0.2857142857142857,0.3245295477055134,A75,A152,male,old,good
+A14,0.3571428571428571,0.3336084516342027,A75,A152,male,old,good
+A11,0.14285714285714285,0.41900517222405637,A75,A153,male,old,bad
+A13,0.14285714285714285,0.05760977220204688,A73,A151,male,young,good
+A11,0.24999999999999997,0.121712336304611,A73,A152,male,young,bad
+A14,0.24999999999999997,0.3582590513920986,A73,A151,male,old,bad
+A14,0.5714285714285714,0.5129305601408606,A72,A152,male,old,bad
+A14,0.5714285714285714,0.39391438318476946,A75,A152,male,old,good
+A11,0.5714285714285714,0.36519203257400684,A73,A152,male,old,bad
+A14,0.3571428571428571,0.18548475844613185,A75,A152,male,old,bad
+A11,0.03571428571428571,0.06063607351160999,A71,A151,female,young,good
+A13,0.14285714285714285,0.06767910201386595,A71,A153,male,old,good
+A14,0.24999999999999997,0.042808407615274574,A73,A152,female,old,good
+A11,0.14285714285714285,0.035875426433366345,A74,A151,male,young,bad
+A11,0.10714285714285712,0.1136238582590514,A75,A152,male,old,good
+A14,0.08928571428571427,0.09519093210080334,A72,A151,female,young,bad
+A12,0.3571428571428571,0.07169582920655881,A74,A151,female,young,good
+A13,0.10714285714285712,0.05447342357213601,A75,A153,female,old,bad
+A12,0.14285714285714285,0.05353802134918015,A75,A151,male,old,bad
+A11,0.3035714285714286,0.12963574336964895,A72,A151,female,old,good
+A11,0.14285714285714285,0.06377242214152085,A75,A152,male,old,good
+A14,0.24999999999999997,0.07252118410916694,A73,A151,female,young,good
+A14,0.01785714285714285,0.1759656652360515,A74,A152,male,old,good
+A12,0.24999999999999997,0.1992956971497744,A71,A152,female,old,good
+A12,0.14285714285714285,0.116430064927919,A74,A152,male,old,good
+A14,0.5714285714285714,0.4087157477715418,A74,A152,female,old,good
+A12,0.5714285714285714,0.5033564432706064,A72,A151,male,old,bad
+A14,0.5714285714285714,0.5126004181798173,A73,A152,female,old,good
+A11,0.3571428571428571,0.0928799383735006,A72,A152,male,old,bad
+A11,0.14285714285714285,0.10448993067018818,A73,A153,male,old,bad
+A14,0.5714285714285714,0.350170573346539,A75,A152,male,old,good
+A14,0.14285714285714285,0.14355672939363928,A71,A152,male,old,good
+A12,0.2857142857142857,0.13090128755364808,A73,A152,male,old,good
+A12,0.24999999999999997,0.32761087267525035,A73,A152,male,old,good
+A11,0.03571428571428571,0.18851105975569496,A73,A151,male,old,good
+A12,0.19642857142857142,0.114229118520964,A73,A152,male,old,good
+A11,0.03571428571428571,0.17244415098492352,A73,A151,male,old,bad
+A12,0.7857142857142857,0.1546164850885881,A74,A152,male,old,bad
+A13,0.03571428571428571,0.05771981952239463,A73,A152,male,old,good
+A14,0.10714285714285712,0.06426763508308572,A73,A151,male,old,good
+A14,0.14285714285714285,0.04104765048971058,A74,A152,female,young,good
+A14,0.2857142857142857,0.1780015406624849,A72,A152,male,old,good
+A14,0.24999999999999997,0.17211400902388027,A73,A152,male,old,good
+A14,0.08928571428571427,0.05359304500935402,A73,A152,male,old,good
+A13,0.08928571428571427,0.02723671178606801,A73,A152,female,old,bad
+A12,0.14285714285714285,0.08858809287993837,A71,A152,male,old,good
+A11,0.2857142857142857,0.2213051612193243,A75,A152,female,young,good
+A12,0.053571428571428575,0.11912622427643886,A73,A152,male,old,good
+A14,0.3571428571428571,0.19406844943325632,A73,A152,male,old,good
+A12,0.4642857142857143,0.10625068779575217,A75,A152,male,old,good
+A14,0.24999999999999997,0.239022779795312,A72,A152,male,old,good
+A11,0.14285714285714285,0.028007043028502255,A74,A152,male,old,bad
+A14,0.4642857142857143,0.23550126554418402,A74,A151,female,old,good
+A14,0.10714285714285712,0.09210960713106636,A73,A152,male,old,good
+A14,0.14285714285714285,0.17591064157587763,A73,A152,female,old,good
+A12,1.0,0.7580609662154726,A74,A152,male,old,bad
+A13,0.24999999999999997,0.14383184769450866,A73,A152,male,old,bad
+A14,0.14285714285714285,0.05529877847474414,A75,A152,male,old,good
+A11,0.6785714285714286,0.22669747991636405,A74,A152,male,old,bad
+A14,0.14285714285714285,0.03719599427753935,A71,A152,female,old,good
+A12,0.19642857142857142,0.030373060415978873,A75,A152,male,old,bad
+A12,0.14285714285714285,0.04632992186640256,A73,A152,female,old,good
+A14,0.3035714285714286,0.07378672829316606,A74,A152,male,old,good
+A14,0.5714285714285714,0.23131946737096953,A73,A152,female,old,good
+A14,0.14285714285714285,0.023770221195113902,A74,A152,female,old,good
+A12,0.24999999999999997,0.3362495873225487,A75,A152,male,old,good
+A14,0.14285714285714285,0.055463849455265765,A72,A151,female,young,good
+A12,0.4107142857142857,0.12490370859469573,A73,A152,male,young,bad
+A11,0.03571428571428571,0.00484208209530098,A75,A152,male,old,good
+A12,0.2857142857142857,0.37454605480356556,A74,A151,male,old,good
+A11,0.2857142857142857,0.10922196544514141,A73,A151,male,old,bad
+A13,0.14285714285714285,0.06459777704412897,A75,A152,female,old,good
+A12,0.3571428571428571,0.6095521074061846,A73,A152,male,old,bad
+A14,0.3571428571428571,0.05496863651370089,A72,A152,male,old,good
+A12,0.08928571428571427,0.10278419720479806,A73,A152,male,old,good
+A11,0.3571428571428571,0.05694948827996038,A74,A151,female,old,bad
+A14,0.7857142857142857,0.18311874105865522,A75,A152,male,old,good
+A14,0.10714285714285712,0.14295146913172666,A73,A152,male,old,good
+A11,0.24999999999999997,0.17932210850665786,A74,A152,female,young,good
+A11,0.07142857142857142,0.05029162539892153,A75,A153,male,old,good
+A12,0.3035714285714286,0.2050181578078574,A74,A152,male,old,good
+A12,0.24999999999999997,0.18498954550456698,A75,A152,female,old,good
+A12,1.2142857142857142,0.2941014636293606,A73,A152,male,young,bad
+A11,0.14285714285714285,0.025255860019808517,A75,A152,male,old,bad
+A11,0.5714285714285714,0.43903378452734676,A73,A152,male,old,bad
+A11,0.3035714285714286,0.17409486079013978,A72,A152,male,old,bad
+A12,0.5714285714285714,0.6802024870694399,A73,A153,male,old,bad
+A14,0.03571428571428571,0.08253549026081215,A75,A152,male,old,good
+A14,0.14285714285714285,0.15555188731154398,A73,A152,male,old,good
+A14,0.19642857142857142,0.2791900517222406,A75,A153,female,old,good
+A13,0.0,0.06844943325630021,A72,A152,male,old,good
+A11,0.14285714285714285,0.12809508088478047,A71,A153,female,old,good
+A12,0.053571428571428575,0.11439418950148562,A72,A152,female,old,good
+A11,0.08928571428571427,0.022229558710245404,A73,A152,male,old,bad
+A14,0.5714285714285714,0.4337515131506548,A73,A152,female,old,good
+A14,0.3571428571428571,0.19880048420820953,A73,A153,male,old,good
+A12,0.14285714285714285,0.39738087377572356,A71,A151,female,young,good
+A14,0.14285714285714285,0.07180587652690656,A75,A153,male,old,bad
+A11,0.14285714285714285,0.055683944095961266,A75,A152,male,old,good
+A12,0.08928571428571427,0.02971277649389237,A73,A152,female,old,good
+A12,0.4642857142857143,0.27423792230659183,A71,A152,male,old,bad
+A11,0.4642857142857143,0.11863101133487398,A74,A151,female,young,bad
+A12,0.14285714285714285,0.1496093320127655,A72,A151,female,young,bad
+A14,0.3035714285714286,0.6839440959612633,A75,A153,male,old,bad
+A11,0.14285714285714285,0.045779685264663805,A73,A152,male,old,bad
+A12,0.4107142857142857,0.20166171453725104,A73,A152,male,old,bad
+A14,0.08928571428571427,0.19709475074281943,A75,A152,male,old,good
+A12,0.24999999999999997,0.09232970177176185,A72,A152,male,old,bad
+A12,0.5714285714285714,0.5286123032904149,A74,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-train.csv
new file mode 100644
index 0000000..b15c2c9
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-train.csv
@@ -0,0 +1,501 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,0.5714285714285714,0.20408275558490152,A71,A152,male,old,good
+A14,0.08928571428571427,0.18306371739848135,A73,A151,male,old,good
+A14,0.24999999999999997,0.12462859029382635,A73,A152,male,old,good
+A12,0.14285714285714285,0.09601628700341147,A72,A152,male,old,good
+A14,1.0,0.5566193463189171,A75,A152,male,old,good
+A14,0.3571428571428571,0.39303400462198745,A73,A152,male,old,good
+A13,0.14285714285714285,0.008748761967646089,A73,A151,female,old,good
+A14,0.08928571428571427,0.12418840101243535,A75,A153,male,old,good
+A14,0.4642857142857143,0.25096291405304283,A74,A151,female,young,good
+A12,0.7857142857142857,0.3471992956971498,A74,A152,male,young,bad
+A12,0.5714285714285714,0.48332783096731596,A72,A151,male,old,bad
+A11,0.3571428571428571,0.1816881258941345,A74,A152,male,old,bad
+A12,0.2857142857142857,0.3421371189611533,A71,A152,male,old,good
+A11,0.4642857142857143,0.1250137559150435,A75,A152,male,old,good
+A12,0.5714285714285714,0.14141080664685815,A72,A152,male,old,bad
+A14,0.03571428571428571,0.029327610872675252,A73,A152,male,old,good
+A11,0.3035714285714286,0.19329811819082207,A74,A152,male,young,good
+A14,0.4642857142857143,0.39809618135798397,A71,A152,female,old,bad
+A11,0.19642857142857142,0.06344228018047761,A73,A151,female,old,good
+A12,0.125,0.24876196764608782,A74,A152,male,old,good
+A11,0.3571428571428571,0.20743919885550788,A73,A152,male,old,good
+A12,0.3571428571428571,0.052327500825354895,A72,A152,male,old,good
+A12,0.7857142857142857,0.28232640035215145,A71,A153,male,old,good
+A14,0.3571428571428571,0.28947947617475517,A73,A152,male,young,good
+A14,0.24999999999999997,0.04666006382744581,A71,A152,female,old,good
+A12,0.7857142857142857,0.32871134587872786,A75,A153,male,old,bad
+A14,0.125,0.06289204357873886,A72,A152,female,old,good
+A12,0.08928571428571427,0.09183448883019699,A74,A151,male,old,good
+A14,0.19642857142857142,0.2638384505337295,A75,A152,female,old,good
+A12,0.7857142857142857,0.3136898866512601,A73,A152,female,young,bad
+A11,0.07142857142857142,0.026466380543633764,A75,A152,male,old,good
+A14,0.24999999999999997,0.02085396720589854,A75,A152,male,old,good
+A14,0.24999999999999997,0.08572686255089688,A74,A152,female,young,bad
+A12,0.3035714285714286,0.05161219324309453,A75,A152,female,old,bad
+A14,0.3571428571428571,0.16424562561901618,A75,A152,male,old,good
+A12,0.3571428571428571,0.6223175965665235,A73,A151,female,young,bad
+A12,0.08928571428571427,0.04974138879718279,A75,A152,male,old,good
+A14,0.24999999999999997,0.19808517662594918,A74,A152,male,old,good
+A11,0.3571428571428571,0.11604489930670188,A71,A152,male,old,bad
+A14,0.3571428571428571,0.07059535600308132,A72,A152,female,old,good
+A13,0.5714285714285714,0.21789369428854408,A73,A152,male,old,bad
+A11,0.4642857142857143,0.19847034224716628,A73,A152,male,old,good
+A14,0.14285714285714285,0.06289204357873886,A75,A152,male,old,good
+A14,0.24999999999999997,0.08528667326950588,A73,A152,male,young,good
+A13,0.14285714285714285,0.1101023440079234,A73,A152,female,old,good
+A12,0.14285714285714285,0.03224386486189061,A72,A152,female,young,bad
+A11,0.24999999999999997,0.06025090789039286,A73,A152,male,old,bad
+A12,0.5714285714285714,0.2500825354902608,A72,A152,female,old,good
+A11,0.14285714285714285,0.025200836359634642,A73,A152,male,old,good
+A11,0.14285714285714285,0.05716958292065588,A73,A152,male,young,good
+A14,0.24999999999999997,0.08880818752063387,A73,A152,female,old,bad
+A12,0.7857142857142857,0.5342797402883239,A72,A152,female,old,bad
+A11,0.24999999999999997,0.27797953119841534,A75,A153,male,old,good
+A11,0.03571428571428571,0.08924837680202487,A71,A153,male,old,good
+A11,0.3571428571428571,0.12044679212061186,A72,A152,female,old,bad
+A14,0.08928571428571427,0.06261692527786948,A73,A151,female,old,good
+A12,0.7857142857142857,0.6558270056124132,A71,A153,male,old,good
+A14,0.24999999999999997,0.13271706833938596,A74,A152,male,old,good
+A14,0.1607142857142857,0.06377242214152085,A71,A152,female,old,good
+A11,0.19642857142857142,0.1866952789699571,A75,A152,female,old,good
+A11,0.7857142857142857,0.358093980411577,A73,A152,female,old,bad
+A14,0.3571428571428571,0.0880928799383735,A74,A152,male,old,good
+A11,0.4107142857142857,0.17420490811048753,A73,A152,male,old,good
+A14,0.4107142857142857,0.2678001540662485,A74,A152,male,old,good
+A14,0.10714285714285712,0.07411687025420931,A73,A151,male,old,good
+A14,0.03571428571428571,0.06228678331682623,A73,A152,female,old,good
+A14,0.10714285714285712,0.05397821063057115,A75,A152,male,old,good
+A12,0.19642857142857142,0.13101133487399583,A73,A151,female,old,bad
+A12,0.4642857142857143,0.10454495433036205,A73,A152,female,young,bad
+A12,0.14285714285714285,0.1474634092659844,A74,A152,female,old,good
+A14,0.03571428571428571,0.3582590513920986,A74,A152,male,old,good
+A12,0.3035714285714286,0.13728403213381754,A74,A152,male,old,good
+A11,0.3571428571428571,0.11417409486079015,A74,A152,male,old,good
+A14,0.03571428571428571,0.05249257180587652,A73,A151,male,old,good
+A11,0.24999999999999997,0.04396390447892594,A72,A151,female,young,good
+A14,0.14285714285714285,0.11235831407505227,A71,A152,male,old,bad
+A14,0.08928571428571427,0.06597336854847584,A74,A152,female,old,good
+A11,0.24999999999999997,0.39952679652250467,A75,A153,male,old,bad
+A12,0.08928571428571427,0.03675580499614835,A73,A152,female,old,bad
+A14,0.3571428571428571,0.1153295917244415,A74,A152,male,old,good
+A11,0.3571428571428571,0.07444701221525256,A75,A152,female,old,good
+A11,0.125,0.2011114779355123,A73,A151,male,old,good
+A14,0.24999999999999997,0.0482007263123143,A72,A151,female,young,good
+A13,0.3035714285714286,0.14707824364476726,A73,A152,female,old,good
+A12,0.08928571428571427,0.1620446792120612,A73,A152,female,old,good
+A11,0.03571428571428571,0.24573566633652472,A72,A152,male,old,good
+A12,0.03571428571428571,0.7871134587872785,A71,A152,male,young,bad
+A11,0.5714285714285714,0.2884890502916254,A75,A153,male,old,good
+A14,0.3571428571428571,0.0734565863321228,A74,A152,male,old,good
+A11,0.14285714285714285,0.18377902498074172,A73,A152,male,old,good
+A14,0.14285714285714285,0.1380543633762518,A75,A152,male,old,good
+A11,0.14285714285714285,0.10300429184549356,A73,A151,male,old,good
+A12,0.14285714285714285,0.09640145262462858,A74,A151,male,old,good
+A14,0.3571428571428571,0.09084406294706723,A73,A151,male,old,good
+A14,0.14285714285714285,0.06839440959612633,A72,A152,female,old,good
+A12,0.03571428571428571,0.052767690106745896,A71,A152,male,old,bad
+A12,0.7857142857142857,0.20518322878837902,A74,A152,male,old,good
+A11,0.3571428571428571,0.4110817651590184,A72,A152,female,old,good
+A11,0.24999999999999997,0.1870804445911742,A72,A151,female,young,good
+A11,0.24999999999999997,0.08968856608341587,A74,A152,male,old,good
+A12,0.24999999999999997,0.09304500935402223,A73,A152,male,old,good
+A14,0.19642857142857142,0.034334763948497854,A72,A152,female,young,good
+A14,0.4107142857142857,0.2718168812589414,A75,A152,male,old,good
+A14,0.5714285714285714,0.5627269725982172,A75,A153,male,old,good
+A12,0.3571428571428571,0.22565203037306042,A73,A152,female,old,good
+A14,0.3571428571428571,0.06195664135578298,A74,A152,female,old,good
+A12,0.4642857142857143,0.1786068009243975,A73,A152,male,old,good
+A11,0.6428571428571428,0.3162759986794322,A73,A152,male,old,bad
+A11,0.4285714285714286,0.20666886761307363,A73,A152,male,old,bad
+A14,0.5714285714285714,0.1701331572576208,A75,A152,male,old,good
+A12,0.24999999999999997,0.13051612193243095,A73,A152,male,old,good
+A12,0.5714285714285714,0.11742049081104876,A72,A151,male,old,bad
+A11,0.3571428571428571,0.07752833718498954,A72,A151,female,old,bad
+A14,0.3571428571428571,0.1781666116430065,A74,A152,female,young,good
+A12,1.0,0.39440959612633436,A73,A152,male,old,good
+A14,0.24999999999999997,0.10938703642566304,A73,A152,female,young,good
+A11,0.3571428571428571,0.059590623968306375,A71,A153,male,old,bad
+A14,0.3214285714285714,0.05683944095961263,A74,A151,female,young,good
+A14,0.03571428571428571,0.17981732144822274,A73,A151,male,old,good
+A12,0.14285714285714285,0.018432926158248045,A73,A151,male,young,good
+A12,0.625,0.25767580059425554,A74,A152,male,young,bad
+A14,0.3571428571428571,0.056564322658743255,A75,A152,male,old,good
+A14,0.14285714285714285,0.08352591614394189,A73,A151,male,young,good
+A11,0.3571428571428571,0.17051832287883792,A72,A153,male,old,bad
+A12,0.08928571428571427,0.0627819962583911,A73,A152,male,old,good
+A14,0.24999999999999997,0.17453505007153078,A75,A152,male,old,good
+A14,0.10714285714285712,0.10432485968966655,A72,A151,female,young,good
+A14,0.19642857142857142,0.06988004842082095,A75,A152,male,old,good
+A14,0.3571428571428571,0.3627159678661825,A73,A152,male,old,good
+A14,0.24999999999999997,0.09909761197314845,A72,A152,male,old,good
+A11,0.3571428571428571,0.15109497083746012,A73,A152,male,young,bad
+A11,0.3571428571428571,0.18763068119291296,A73,A152,female,old,good
+A11,0.3571428571428571,0.06558820292725871,A74,A151,female,young,bad
+A11,0.08928571428571427,0.04886101023440079,A73,A152,male,young,good
+A14,0.4642857142857143,0.3037856278199626,A74,A152,female,young,good
+A11,0.24999999999999997,0.12930560140860572,A73,A153,male,old,bad
+A13,0.3571428571428571,0.19252778694838782,A72,A152,female,old,good
+A14,0.14285714285714285,0.15456146142841423,A74,A152,male,old,good
+A11,0.03571428571428571,0.022669747991636405,A72,A152,male,old,good
+A11,0.24999999999999997,0.09298998569384835,A72,A153,male,old,good
+A12,0.5714285714285714,0.1140640475404424,A74,A151,male,young,good
+A12,0.7857142857142857,0.6577528337184989,A73,A152,male,old,good
+A14,0.3571428571428571,0.28925938153405967,A75,A153,male,old,good
+A14,0.24999999999999997,0.010069329811819083,A71,A151,female,young,bad
+A14,0.03571428571428571,0.23995818201826785,A72,A152,female,old,bad
+A12,0.14285714285714285,0.13821943435677342,A75,A152,female,young,bad
+A14,0.4642857142857143,0.2213051612193243,A73,A152,male,old,good
+A12,0.625,0.6333223286012986,A74,A151,male,old,good
+A12,0.5714285714285714,0.10867172884340266,A75,A153,male,old,bad
+A13,0.14285714285714285,0.06734896005282272,A72,A152,female,old,good
+A14,0.14285714285714285,0.06338725652030373,A74,A151,female,old,good
+A12,0.24999999999999997,0.09150434686915374,A72,A152,male,old,good
+A14,0.17857142857142858,0.030373060415978873,A73,A152,male,old,good
+A13,0.19642857142857142,0.05617915703752613,A73,A153,male,old,bad
+A13,0.3571428571428571,0.055463849455265765,A73,A152,female,old,good
+A14,0.14285714285714285,0.18240343347639487,A72,A152,male,old,good
+A11,0.14285714285714285,0.17255419830527127,A75,A153,male,old,bad
+A13,0.14285714285714285,0.1098822493672279,A73,A152,female,old,good
+A14,0.3571428571428571,0.30560140860570045,A73,A152,male,old,good
+A14,0.19642857142857142,0.03906679872345109,A74,A152,female,old,good
+A12,0.24999999999999997,0.044349070100143065,A75,A152,male,old,bad
+A13,0.6785714285714286,0.3322878837900297,A72,A152,male,old,good
+A14,0.08928571428571427,0.13772422141520854,A75,A152,male,old,good
+A14,0.10714285714285712,0.10784637394079453,A73,A151,male,young,bad
+A11,0.7857142857142857,0.5528227137669197,A74,A153,male,old,bad
+A12,0.5714285714285714,0.7740728513260702,A75,A153,male,old,bad
+A12,0.4642857142857143,0.2200396170353252,A71,A152,male,old,bad
+A14,0.3571428571428571,0.03736106525806096,A74,A152,male,old,good
+A12,0.08928571428571427,0.039011775063277215,A73,A152,female,old,bad
+A14,0.3035714285714286,0.13931990756025092,A74,A152,female,old,good
+A12,0.7857142857142857,0.44613183668977663,A72,A152,female,old,good
+A14,0.08928571428571427,0.11285352701661715,A72,A151,female,young,good
+A12,0.7857142857142857,0.18245845713656875,A74,A152,male,old,good
+A11,0.24999999999999997,0.20446792120611865,A75,A151,female,old,bad
+A14,0.03571428571428571,0.05430835259161439,A73,A152,female,old,good
+A11,0.24999999999999997,0.03994717728623308,A72,A152,female,young,bad
+A11,0.3571428571428571,0.03659073401562672,A75,A152,female,old,bad
+A13,0.3571428571428571,0.03835149114119071,A74,A153,male,old,bad
+A12,0.3571428571428571,0.14168592494772753,A74,A152,male,old,good
+A14,0.3035714285714286,0.11213821943435677,A72,A152,female,young,good
+A14,0.10714285714285712,0.145867723120942,A72,A151,female,old,good
+A12,0.7321428571428571,0.23803235391218225,A71,A152,male,old,good
+A14,0.7857142857142857,0.3845053372950369,A75,A152,male,old,good
+A11,0.5714285714285714,0.13733905579399142,A75,A152,male,old,bad
+A12,0.3571428571428571,0.08215032463959503,A75,A152,male,old,good
+A11,0.08928571428571427,0.06140640475404424,A72,A151,female,young,bad
+A14,0.24999999999999997,0.06905469351821283,A73,A153,male,old,good
+A11,0.03571428571428571,0.005117200396170352,A72,A152,female,old,good
+A14,0.3571428571428571,0.10856168152305491,A75,A152,male,old,good
+A14,0.14285714285714285,0.06850445691647408,A75,A152,male,old,good
+A13,0.19642857142857142,0.11428414218113787,A72,A152,female,young,bad
+A11,0.7321428571428571,0.6364036535710355,A75,A151,male,old,bad
+A12,0.14285714285714285,0.06789919665456146,A73,A152,male,young,bad
+A14,0.5714285714285714,0.4844833278309673,A73,A153,male,old,good
+A14,0.14285714285714285,0.07730824254429404,A75,A152,male,old,good
+A14,0.24999999999999997,0.06619346318917134,A72,A152,female,old,good
+A11,0.14285714285714285,0.053813139650049524,A73,A152,female,young,bad
+A12,0.03571428571428571,0.037526136238582586,A74,A152,female,old,good
+A12,0.5714285714285714,0.4311654011224827,A75,A152,male,old,bad
+A14,0.03571428571428571,0.06030593155056674,A75,A153,male,old,good
+A12,0.5714285714285714,0.1437218003741609,A75,A153,male,old,good
+A11,0.3571428571428571,0.15808297567954221,A72,A152,female,old,bad
+A12,1.0,0.3938593595245956,A72,A152,female,young,bad
+A12,0.19642857142857142,0.055793991416309016,A73,A151,male,young,bad
+A14,0.03571428571428571,0.07175085286673269,A74,A151,female,young,good
+A12,0.3035714285714286,0.18719049191152196,A74,A152,male,old,good
+A11,0.7857142857142857,0.20254209310003302,A74,A153,male,old,bad
+A11,0.03571428571428571,0.061131286453174866,A72,A152,male,old,good
+A11,0.03571428571428571,0.052437548145702645,A75,A152,male,old,good
+A14,0.14285714285714285,0.18031253438978762,A72,A152,male,old,good
+A14,0.5714285714285714,0.4184549356223176,A73,A152,female,young,bad
+A11,0.3035714285714286,0.17095851216022892,A72,A152,female,old,good
+A13,0.24999999999999997,0.10179377132166832,A73,A152,male,old,bad
+A14,0.24999999999999997,0.0505667436997909,A73,A152,male,old,good
+A14,0.5714285714285714,0.47854077253218885,A74,A152,male,old,good
+A11,0.24999999999999997,0.09276989105315285,A74,A151,male,young,good
+A12,0.3571428571428571,0.1794871794871795,A74,A152,male,old,good
+A12,0.3571428571428571,0.3695939253879168,A74,A151,male,old,good
+A14,0.4642857142857143,0.11461428414218112,A75,A152,male,old,good
+A11,0.24999999999999997,0.09843732805106195,A73,A151,female,young,bad
+A11,0.5714285714285714,0.49636843842852424,A74,A152,male,young,bad
+A14,0.3035714285714286,0.07274127874986244,A75,A152,female,old,good
+A12,0.03571428571428571,0.011224826675470454,A72,A152,male,young,good
+A14,0.5714285714285714,0.33311323869263787,A75,A152,male,old,good
+A14,0.3571428571428571,0.10713106635853417,A74,A152,male,old,good
+A14,0.7857142857142857,0.40601958842302194,A75,A152,male,old,good
+A13,0.10714285714285712,0.05639925167822163,A72,A152,female,young,good
+A11,0.7857142857142857,0.37795752173434577,A73,A153,male,old,bad
+A12,0.3571428571428571,0.6783867062837019,A75,A153,female,old,bad
+A12,0.24999999999999997,0.1391548365797293,A73,A151,male,young,good
+A14,0.0,0.06894464619786508,A74,A152,male,old,good
+A12,0.07142857142857142,0.05430835259161439,A73,A152,female,young,bad
+A14,0.7857142857142857,0.20160669087707717,A73,A152,male,old,bad
+A14,0.3571428571428571,0.15472653240893586,A75,A151,male,old,good
+A12,0.5714285714285714,0.09370529327610873,A75,A153,male,old,bad
+A12,0.4642857142857143,0.19709475074281943,A72,A152,male,young,good
+A14,0.03571428571428571,0.0708704743039507,A72,A152,female,old,good
+A14,0.03571428571428571,0.10223396060305931,A74,A151,male,old,good
+A11,0.03571428571428571,0.00979421151094971,A75,A152,female,old,good
+A12,0.5714285714285714,0.2313744910311434,A73,A152,male,old,bad
+A13,0.3035714285714286,0.1138439528997469,A72,A151,male,old,bad
+A14,0.14285714285714285,0.05023660173874765,A73,A152,male,old,good
+A14,0.08928571428571427,0.037746230879278087,A75,A152,male,old,good
+A11,0.24999999999999997,0.12231759656652359,A71,A152,male,young,bad
+A14,0.19642857142857142,0.07257620776934082,A75,A152,male,old,good
+A12,0.14285714285714285,0.1522504677011115,A72,A151,female,old,good
+A14,0.14285714285714285,0.11450423682183337,A75,A152,male,old,good
+A14,0.14285714285714285,0.10157367668097282,A74,A152,male,old,good
+A12,0.14285714285714285,0.04330362055683944,A74,A152,male,old,good
+A14,0.03571428571428571,0.05425332893144051,A73,A151,male,old,good
+A14,0.3571428571428571,0.21464729833828547,A73,A152,male,old,good
+A14,0.7857142857142857,0.18482447452404535,A73,A152,female,old,good
+A14,0.14285714285714285,0.06393749312204247,A73,A152,female,old,good
+A11,0.125,0.20298228238142402,A73,A152,male,old,good
+A11,0.14285714285714285,0.0702101903818642,A75,A153,male,old,good
+A14,0.3214285714285714,0.1334323759216463,A75,A152,male,old,good
+A12,0.4642857142857143,0.2184989545504567,A73,A152,female,old,good
+A11,0.3571428571428571,0.14713326730494114,A73,A152,male,old,good
+A14,0.24999999999999997,0.08622207549246176,A73,A152,female,old,good
+A13,0.10714285714285712,0.0536480686695279,A73,A152,male,old,good
+A11,0.19642857142857142,0.08776273797733025,A72,A151,female,old,good
+A14,0.03571428571428571,0.0,A73,A152,female,old,good
+A14,0.03571428571428571,0.022559700671288655,A74,A151,male,young,good
+A13,0.03571428571428571,0.025255860019808517,A72,A152,male,old,good
+A11,0.5714285714285714,0.8587542643336634,A71,A152,male,old,good
+A12,0.3571428571428571,0.08732254869593925,A74,A153,female,old,bad
+A11,0.1607142857142857,0.08512160228898426,A72,A152,male,old,good
+A11,0.03571428571428571,0.010894684714427203,A72,A152,female,young,bad
+A12,0.24999999999999997,0.16303510509519095,A72,A151,male,young,good
+A11,0.14285714285714285,0.12815010454495435,A72,A152,male,old,bad
+A11,0.03571428571428571,0.05216242984483327,A75,A153,female,old,bad
+A14,0.3035714285714286,0.26152745680642675,A73,A152,female,old,bad
+A12,0.3571428571428571,0.11461428414218112,A72,A152,male,old,good
+A14,0.19642857142857142,0.055683944095961266,A74,A152,male,old,good
+A12,0.14285714285714285,0.049796412457356665,A75,A152,male,old,good
+A14,0.10714285714285712,0.03543523715197534,A74,A152,female,old,good
+A14,0.14285714285714285,0.16952789699570817,A75,A152,male,old,good
+A12,0.24999999999999997,0.04748541873005392,A73,A152,female,old,good
+A14,0.14285714285714285,0.02580609662154727,A75,A152,male,old,bad
+A12,0.3571428571428571,0.09981291955540883,A71,A152,female,old,bad
+A13,0.4642857142857143,0.18741058655221746,A75,A152,male,old,good
+A11,0.3571428571428571,0.3502255970067129,A71,A153,male,old,good
+A14,0.24999999999999997,0.16391548365797293,A71,A152,male,old,good
+A14,0.19642857142857142,0.17018818091779467,A73,A153,male,old,good
+A11,0.4642857142857143,0.1572576207769341,A72,A152,male,old,bad
+A12,0.08928571428571427,0.011444921316165951,A73,A152,male,young,good
+A14,0.19642857142857142,0.06718388907230109,A73,A153,male,old,good
+A12,0.125,0.05898536370639375,A73,A152,female,old,good
+A11,0.08928571428571427,0.057114559260482006,A75,A152,male,old,good
+A14,0.14285714285714285,0.025145812699460767,A73,A152,male,old,good
+A12,0.19642857142857142,0.058215032463959496,A75,A152,male,old,good
+A14,0.08928571428571427,0.15538681633102236,A73,A152,male,old,good
+A11,0.19642857142857142,0.12440849565313085,A71,A151,female,young,good
+A14,0.03571428571428571,0.0819852536590734,A75,A151,male,old,good
+A14,0.14285714285714285,0.04803565533179267,A73,A151,female,old,bad
+A12,0.125,0.07301639705073182,A72,A152,female,young,good
+A11,0.7321428571428571,0.08776273797733025,A73,A153,male,young,bad
+A14,0.3571428571428571,0.09783206778914934,A74,A152,male,old,good
+A12,0.4107142857142857,0.4439308902828216,A75,A153,female,old,bad
+A12,0.14285714285714285,0.05749972488169913,A72,A151,female,young,bad
+A12,0.3571428571428571,0.6431165401122483,A73,A152,male,old,bad
+A14,0.5714285714285714,0.08633212281280951,A73,A153,male,old,bad
+A11,0.19642857142857142,0.0675690546935182,A75,A152,male,old,good
+A14,0.3571428571428571,0.06965995378012545,A73,A152,female,old,good
+A11,0.08928571428571427,0.1037746230879278,A73,A152,male,young,good
+A14,0.19642857142857142,0.19599427753934193,A72,A152,female,young,good
+A14,0.24999999999999997,0.21541762958071972,A73,A152,male,old,bad
+A12,0.3571428571428571,0.20947507428194126,A74,A152,male,old,bad
+A14,0.7857142857142857,0.543468691537361,A73,A153,male,old,bad
+A12,0.03571428571428571,0.012875536480686695,A74,A152,male,old,good
+A12,1.0,0.33223286012985587,A73,A153,male,old,bad
+A11,0.3571428571428571,0.36436667767139874,A72,A152,male,old,bad
+A14,0.19642857142857142,0.07054033234290745,A73,A152,female,old,good
+A14,0.08928571428571427,0.10366457576758006,A73,A152,male,old,good
+A11,0.14285714285714285,0.04715527676901067,A73,A151,male,young,good
+A11,0.2857142857142857,0.10795642126114229,A74,A152,male,old,good
+A14,0.19642857142857142,0.16154946627049632,A74,A151,female,young,good
+A11,0.7678571428571428,0.5762077693408165,A72,A152,female,old,good
+A14,0.053571428571428575,0.02641135688345989,A75,A151,male,old,good
+A14,0.14285714285714285,0.10030813249697369,A73,A152,female,old,good
+A11,0.3035714285714286,0.0176625949158138,A75,A152,male,old,good
+A13,0.19642857142857142,0.11609992296687575,A73,A152,male,old,good
+A11,0.14285714285714285,0.06019588423021899,A73,A152,male,old,good
+A11,0.6785714285714286,0.3809838230439089,A74,A152,female,old,bad
+A13,0.6785714285714286,0.2501375591504347,A75,A153,male,old,good
+A11,0.6785714285714286,0.20441289754594477,A72,A152,male,old,bad
+A14,0.125,0.38395510069329813,A73,A152,male,old,good
+A12,0.3571428571428571,0.05315285572796302,A72,A152,male,old,bad
+A14,0.14285714285714285,0.13744910311433917,A75,A153,female,old,good
+A12,0.6785714285714286,0.49702872235061074,A71,A153,male,old,good
+A14,0.4107142857142857,0.23528117090348852,A72,A152,male,old,good
+A14,0.3035714285714286,0.2750082535490261,A73,A152,male,old,good
+A14,0.3571428571428571,0.16127434796962695,A72,A152,female,old,good
+A14,0.3571428571428571,0.02569604930119952,A75,A152,male,old,good
+A14,0.14285714285714285,0.055683944095961266,A73,A152,male,young,good
+A14,0.14285714285714285,0.08990866072411137,A75,A152,male,old,good
+A12,0.24999999999999997,0.09210960713106636,A72,A151,female,old,bad
+A11,0.7857142857142857,0.39760096841641906,A74,A153,male,old,good
+A13,0.14285714285714285,0.03791130185979971,A74,A152,male,old,bad
+A12,0.14285714285714285,0.03251898316275999,A74,A151,female,young,good
+A12,0.3571428571428571,0.21255639925167824,A72,A151,female,old,bad
+A13,0.3571428571428571,0.2697259821723341,A74,A152,male,young,good
+A11,0.5178571428571429,0.22180037416088919,A73,A152,female,young,bad
+A12,0.14285714285714285,0.07246616044899307,A73,A152,female,young,good
+A11,0.14285714285714285,0.004897105755474855,A75,A152,male,old,good
+A11,0.19642857142857142,0.05639925167822163,A73,A151,female,young,bad
+A12,0.14285714285714285,0.021404203807637284,A73,A152,male,old,bad
+A14,0.5714285714285714,0.1127434796962694,A75,A152,male,old,good
+A11,0.3571428571428571,0.086552217453505,A71,A152,male,old,bad
+A11,0.7857142857142857,0.41339275888632115,A75,A153,male,old,bad
+A14,0.03571428571428571,0.07400682293386156,A74,A152,male,old,good
+A11,0.14285714285714285,0.007428194123473095,A74,A152,female,old,good
+A14,0.14285714285714285,0.028942445251458126,A73,A152,male,old,good
+A12,0.3571428571428571,0.15511169803015298,A75,A153,male,old,good
+A11,0.08928571428571427,0.06129635743369649,A74,A152,male,old,good
+A14,0.03571428571428571,0.0534279740288324,A73,A152,male,old,good
+A11,0.07142857142857142,0.17321448222735777,A74,A152,male,old,good
+A11,0.24999999999999997,0.21475734565863322,A73,A152,male,old,bad
+A12,0.19642857142857142,0.08407615274568064,A72,A151,female,old,bad
+A14,0.08928571428571427,0.05425332893144051,A72,A151,female,young,good
+A14,0.03571428571428571,0.08341586882359414,A73,A151,male,young,good
+A11,0.4642857142857143,0.43039506988004844,A72,A152,male,young,good
+A12,0.08928571428571427,0.268460437988335,A75,A153,female,old,bad
+A14,0.3571428571428571,0.4620886981402003,A72,A152,male,old,bad
+A11,0.3571428571428571,0.21563772422141522,A73,A152,male,old,good
+A14,0.14285714285714285,0.09271486739297898,A75,A152,male,old,good
+A14,0.3571428571428571,0.11813579839330911,A75,A152,male,old,bad
+A14,0.19642857142857142,0.06652360515021459,A73,A152,female,old,good
+A12,0.14285714285714285,0.023935292175635527,A74,A152,male,young,bad
+A11,0.7857142857142857,0.2396280400572246,A75,A153,male,young,bad
+A14,0.125,0.10410476504897105,A75,A152,male,old,good
+A11,0.19642857142857142,0.03851656212171234,A75,A151,male,old,bad
+A12,0.21428571428571427,0.050896885660834154,A71,A153,male,old,good
+A11,0.14285714285714285,0.06866952789699571,A73,A152,female,young,good
+A14,0.24999999999999997,0.09541102674149884,A73,A153,male,old,good
+A12,0.5714285714285714,0.19555408825795093,A73,A152,female,old,bad
+A11,0.3571428571428571,0.09557609772202047,A73,A151,male,young,bad
+A14,0.3571428571428571,0.21398701441619897,A73,A152,male,old,good
+A14,0.14285714285714285,0.024815670738417523,A73,A152,male,old,good
+A14,0.8928571428571428,0.5054473423572136,A73,A152,male,old,good
+A11,0.24999999999999997,0.16176956091119182,A73,A152,female,young,bad
+A12,0.1607142857142857,0.034774953229888855,A72,A152,male,young,good
+A12,0.3571428571428571,0.13810938703642567,A75,A153,male,old,good
+A14,0.19642857142857142,0.10845163420270716,A73,A151,female,young,good
+A11,0.14285714285714285,0.020688896225376913,A73,A152,female,young,bad
+A12,0.7857142857142857,1.0,A73,A152,female,old,bad
+A11,0.24999999999999997,0.0532078793881369,A73,A152,male,old,bad
+A12,0.053571428571428575,0.12798503356443272,A73,A152,male,old,good
+A11,0.03571428571428571,0.13189171343677783,A73,A151,male,old,good
+A14,0.4107142857142857,0.12765489160338947,A73,A151,female,young,bad
+A13,0.3571428571428571,0.1594585671838891,A73,A152,male,old,good
+A11,0.03571428571428571,0.022889842632331906,A74,A152,female,old,good
+A14,0.14285714285714285,0.051447122262572906,A73,A152,female,old,good
+A14,0.4642857142857143,0.3138549576317817,A74,A152,male,old,good
+A12,0.7321428571428571,0.24738637614174097,A72,A152,male,young,bad
+A14,0.19642857142857142,0.0529877847474414,A75,A152,male,old,good
+A14,0.03571428571428571,0.09067899196654561,A73,A152,male,old,good
+A14,0.7857142857142857,0.6239683063717398,A73,A151,female,young,bad
+A14,0.0,0.01931330472103004,A72,A151,female,young,good
+A12,0.24999999999999997,0.05474854187300539,A73,A152,male,old,bad
+A14,0.5714285714285714,0.30218994167492025,A74,A152,male,old,good
+A12,0.17857142857142858,0.06382744580169472,A75,A152,male,old,good
+A11,0.14285714285714285,0.035765379113018594,A73,A152,male,young,bad
+A14,0.03571428571428571,0.0061626499394739735,A73,A152,female,old,good
+A14,0.19642857142857142,0.09513590844062947,A75,A152,male,old,good
+A14,0.5714285714285714,0.17051832287883792,A73,A152,female,old,bad
+A14,0.14285714285714285,0.030097942115109497,A75,A152,female,old,bad
+A14,0.03571428571428571,0.014746340926598437,A73,A152,female,old,good
+A11,0.3571428571428571,0.1129635743369649,A75,A152,male,old,bad
+A11,0.5714285714285714,0.1354682513480797,A75,A152,male,old,bad
+A14,0.3571428571428571,0.06734896005282272,A72,A152,male,old,good
+A14,0.24999999999999997,0.08638714647298339,A73,A152,male,old,good
+A14,0.3571428571428571,0.09227467811158797,A73,A152,female,old,good
+A11,0.19642857142857142,0.06509298998569385,A73,A151,female,young,good
+A12,0.14285714285714285,0.03895675140310334,A74,A152,male,old,good
+A14,0.5714285714285714,0.19725982172334106,A75,A152,female,old,good
+A14,0.14285714285714285,0.028227137669197756,A73,A152,female,old,good
+A14,0.03571428571428571,0.15010454495433037,A73,A152,male,old,good
+A14,0.14285714285714285,0.05727963024100363,A73,A152,female,old,good
+A11,0.14285714285714285,0.0233300319137229,A74,A152,male,young,bad
+A14,1.0,0.3453835149114119,A73,A153,male,old,good
+A14,0.14285714285714285,0.03994717728623308,A75,A152,male,old,good
+A13,0.10714285714285712,0.20353251898316277,A72,A152,male,old,good
+A12,0.10714285714285712,0.043908880818752064,A73,A152,male,young,good
+A14,0.3571428571428571,0.13315725762077696,A75,A152,male,old,good
+A11,0.4642857142857143,0.23841751953339935,A73,A152,male,old,good
+A14,0.3571428571428571,0.11692527786948388,A73,A152,male,old,good
+A11,0.14285714285714285,0.056784417299438755,A73,A151,female,young,bad
+A14,0.3571428571428571,0.19296797622977882,A75,A153,female,old,good
+A12,0.3571428571428571,0.2195994277539342,A73,A152,male,old,bad
+A11,0.3571428571428571,0.3482447452404534,A71,A153,male,old,good
+A11,0.14285714285714285,0.10294926818531969,A73,A152,male,old,good
+A14,0.14285714285714285,0.10570045119401342,A72,A152,female,old,good
+A14,0.19642857142857142,0.18399911962143722,A72,A152,female,old,good
+A14,0.5714285714285714,0.24408495653130846,A73,A153,male,old,good
+A14,0.19642857142857142,0.2424892703862661,A73,A152,male,old,good
+A11,0.17857142857142858,0.48024650599757895,A75,A152,male,old,bad
+A14,0.24999999999999997,0.04429404643996919,A72,A152,female,old,good
+A11,0.7857142857142857,0.15412127214702323,A73,A152,male,old,bad
+A12,0.4642857142857143,0.17558049961483438,A73,A151,female,young,bad
+A14,0.3571428571428571,0.26961593485198637,A75,A152,male,old,good
+A14,0.14285714285714285,0.09425552987784747,A74,A151,male,old,good
+A12,0.24999999999999997,0.18377902498074172,A71,A152,male,old,good
+A14,0.5714285714285714,0.3076923076923077,A75,A152,male,old,good
+A14,0.3571428571428571,0.06707384175195334,A75,A151,male,old,good
+A14,0.3571428571428571,0.12947067238912735,A73,A151,female,old,good
+A11,0.14285714285714285,0.18713546825134808,A73,A152,male,old,good
+A11,0.3571428571428571,0.15951359084406297,A72,A153,male,young,good
+A11,0.3571428571428571,0.21239132827115662,A75,A151,male,young,bad
+A12,0.08928571428571427,0.06531308462638935,A74,A152,male,old,bad
+A13,0.3571428571428571,0.062011665015956854,A75,A153,female,old,good
+A11,0.08928571428571427,0.06448772972378122,A72,A153,male,old,bad
+A11,0.4642857142857143,0.18553978210630573,A75,A151,female,old,good
+A11,0.3571428571428571,0.17029822823814242,A75,A151,male,old,bad
+A12,0.14285714285714285,0.020633872565203038,A72,A152,male,old,good
+A12,0.14285714285714285,0.040332342907450205,A72,A151,female,young,good
+A12,0.4642857142857143,0.15081985253659075,A75,A152,female,young,good
+A14,0.24999999999999997,0.049466270496313414,A73,A152,male,old,good
+A12,0.08928571428571427,0.001430615164520744,A73,A151,male,young,good
+A11,0.3571428571428571,0.06168152305491362,A73,A151,female,young,bad
+A11,0.14285714285714285,0.029987894794761747,A72,A152,female,old,bad
+A11,0.3571428571428571,0.056784417299438755,A73,A152,female,old,bad
+A12,0.03571428571428571,0.0064927919005172245,A75,A152,male,old,good
+A14,0.3571428571428571,0.055463849455265765,A74,A152,male,young,good
+A14,0.19642857142857142,0.10652580609662154,A74,A151,female,old,good
+A14,0.3571428571428571,0.14223616154946628,A75,A152,male,old,good
+A11,0.3571428571428571,0.18609001870804448,A73,A151,female,young,good
+A14,0.3571428571428571,0.05838010344448112,A74,A152,male,old,good
+A11,0.5714285714285714,0.11543963904478925,A73,A152,male,old,good
+A12,0.03571428571428571,0.010069329811819083,A72,A151,female,young,bad
+A11,0.19642857142857142,0.20408275558490152,A73,A152,female,old,bad
+A12,0.19642857142857142,0.13101133487399583,A73,A152,female,young,good
+A14,0.14285714285714285,0.12077693408165512,A72,A151,male,old,good
+A11,0.10714285714285712,0.04335864421701331,A74,A152,male,old,good
+A14,0.7857142857142857,0.473643666776714,A74,A153,male,old,good
+A13,0.3571428571428571,0.05639925167822163,A73,A152,male,old,good
+A11,0.3571428571428571,0.3476394849785408,A73,A152,male,young,good
+A13,0.3571428571428571,0.06019588423021899,A74,A152,male,old,bad
+A12,0.4107142857142857,0.3144602178936943,A75,A152,male,old,good
+A14,0.14285714285714285,0.020248706943985912,A75,A152,male,old,good
+A14,0.10714285714285712,0.026246285902938263,A75,A153,male,old,good
+A12,0.19642857142857142,0.06569825024760646,A72,A152,male,young,good
+A11,0.5714285714285714,0.18542973478595798,A73,A152,male,old,good
+A11,0.10714285714285712,0.10955210740618465,A72,A151,male,old,good
+A12,1.0,0.799603829646748,A75,A153,female,old,bad
+A14,0.3571428571428571,0.09750192582810609,A73,A152,female,old,good
+A13,0.14285714285714285,0.019753494002421042,A72,A152,female,old,bad
+A11,0.3571428571428571,0.15247056234180698,A73,A151,male,young,good
+A11,0.24999999999999997,0.2272477165181028,A73,A152,male,old,good
+A14,0.10714285714285712,0.05827005612413337,A73,A152,male,old,bad
+A11,0.3571428571428571,0.14185099592824915,A73,A152,male,young,good
+A12,0.24999999999999997,0.08501155496863651,A75,A151,female,old,good
+A14,0.24999999999999997,0.07059535600308132,A72,A152,male,old,bad
+A14,0.3571428571428571,0.26702982282381427,A72,A153,male,old,good
+A11,0.4642857142857143,0.5707604269836029,A75,A153,male,old,good
+A14,0.14285714285714285,0.06250687795752173,A73,A152,female,old,bad
+A14,0.3571428571428571,0.23885770881479035,A73,A152,male,old,bad
+A13,0.14285714285714285,0.17326950588753165,A75,A152,male,old,good
+A13,0.03571428571428571,0.04385385715857819,A73,A152,female,old,good
+A11,0.7857142857142857,0.4091009133927589,A74,A151,female,old,bad
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-val.csv
new file mode 100644
index 0000000..757c07f
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS-val.csv
@@ -0,0 +1,301 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,0.24999999999999997,0.07169582920655881,A73,A152,male,old,bad
+A14,0.19642857142857142,0.06107626279300099,A73,A152,male,old,good
+A12,0.07142857142857142,0.02806206668867613,A74,A152,female,old,good
+A12,0.24999999999999997,0.14575767580059426,A75,A152,male,old,good
+A12,0.7321428571428571,0.15302079894354573,A73,A151,male,young,bad
+A12,0.3571428571428571,0.24551557169582922,A73,A152,male,old,good
+A12,0.24999999999999997,0.090073731704633,A73,A152,male,old,good
+A12,0.5714285714285714,0.11131286453174864,A74,A152,male,old,good
+A12,0.14285714285714285,0.035105095190932106,A75,A152,male,old,bad
+A11,0.14285714285714285,0.07538241443820842,A73,A152,female,old,good
+A14,0.19642857142857142,0.1352481567073842,A72,A152,male,old,good
+A12,0.24999999999999997,0.03389457466710685,A73,A152,male,young,good
+A12,0.07142857142857142,0.06404754044239022,A73,A152,male,old,good
+A14,0.08928571428571427,0.05849015076482887,A75,A152,male,young,good
+A14,0.3571428571428571,0.12809508088478047,A75,A152,male,old,good
+A14,0.19642857142857142,0.1825685044569165,A75,A151,female,old,good
+A14,0.3571428571428571,0.06289204357873886,A73,A152,male,old,good
+A13,0.03571428571428571,0.10267414988445031,A73,A152,male,old,good
+A14,0.14285714285714285,0.09695168922636734,A74,A152,female,old,good
+A14,0.3571428571428571,0.09298998569384835,A75,A152,male,old,good
+A13,0.03571428571428571,0.05904038736656762,A75,A152,male,old,good
+A11,0.5714285714285714,0.09502586112028172,A75,A152,male,old,bad
+A11,0.14285714285714285,0.047210300429184546,A74,A152,male,old,bad
+A14,0.19642857142857142,0.05777484318256851,A75,A153,male,old,good
+A11,0.03571428571428571,0.0505667436997909,A75,A152,male,old,good
+A11,0.5714285714285714,0.16501595686145043,A74,A153,male,old,good
+A14,0.625,0.1275998679432156,A73,A152,male,young,good
+A12,0.08928571428571427,0.07147573456586331,A72,A152,male,old,good
+A14,0.03571428571428571,0.024760647078243648,A75,A153,male,old,good
+A12,0.14285714285714285,0.32067789149334214,A74,A152,male,old,good
+A12,0.08928571428571427,0.05260261912622427,A75,A152,female,young,good
+A12,0.3571428571428571,0.060801144492131615,A72,A152,female,young,bad
+A14,0.3571428571428571,0.15010454495433037,A73,A152,male,old,good
+A14,0.3571428571428571,0.07015516672169032,A74,A152,female,old,good
+A14,0.03571428571428571,0.037526136238582586,A73,A152,female,young,good
+A12,0.4107142857142857,0.12534389787608674,A72,A152,female,old,good
+A11,0.24999999999999997,0.15758776273797734,A72,A151,female,old,bad
+A13,0.4642857142857143,0.09122922856828436,A75,A152,male,old,bad
+A14,0.14285714285714285,0.05447342357213601,A75,A152,female,old,good
+A14,0.14285714285714285,0.02569604930119952,A75,A152,male,old,good
+A12,0.19642857142857142,0.0695499064597777,A73,A152,male,young,good
+A14,0.19642857142857142,0.08688235941454825,A75,A152,male,old,good
+A14,0.14285714285714285,0.02145922746781116,A73,A152,male,old,good
+A12,0.5714285714285714,0.39517992736876856,A73,A151,female,old,good
+A14,0.14285714285714285,0.0710905689446462,A74,A152,male,old,good
+A11,0.5714285714285714,0.5160669087707714,A74,A152,male,young,bad
+A14,0.03571428571428571,0.00968416419060196,A75,A152,male,old,good
+A14,0.19642857142857142,0.24061846594035435,A73,A152,male,old,bad
+A11,0.24999999999999997,0.45234951028942444,A73,A151,female,young,good
+A12,0.5714285714285714,0.1904368878617806,A73,A152,male,old,good
+A14,0.7857142857142857,0.6877407285132606,A74,A152,male,old,good
+A11,0.6785714285714286,0.4199405744470122,A74,A153,male,old,good
+A14,1.0,0.5444040937603168,A74,A152,female,young,good
+A12,0.3571428571428571,0.05480356553317926,A72,A152,male,young,bad
+A14,0.5714285714285714,0.1840541432816111,A75,A152,male,old,good
+A12,0.14285714285714285,0.18526466380543635,A75,A151,male,old,good
+A14,0.14285714285714285,0.034995047870584356,A73,A152,female,young,good
+A11,0.3571428571428571,0.07131066358534169,A74,A151,male,young,bad
+A11,0.14285714285714285,0.10570045119401342,A73,A152,male,old,good
+A14,0.5714285714285714,0.13002090899086607,A73,A152,male,old,good
+A14,0.14285714285714285,0.04682513480796743,A73,A152,male,old,good
+A12,0.24999999999999997,0.1513700891383295,A74,A151,female,old,good
+A14,0.14285714285714285,0.1116430064927919,A73,A153,male,old,good
+A12,0.10714285714285712,0.06993507208099482,A73,A152,male,old,good
+A12,0.24999999999999997,0.04357873885770881,A73,A152,female,old,bad
+A12,0.7857142857142857,0.3538571585781886,A73,A153,male,old,good
+A11,0.5714285714285714,0.10426983602949268,A74,A152,male,young,bad
+A11,0.3571428571428571,0.2512930560140861,A74,A152,male,old,bad
+A12,0.7857142857142857,0.7797402883239792,A73,A152,male,young,bad
+A11,0.14285714285714285,0.02426543413667877,A75,A152,male,old,bad
+A12,0.24999999999999997,0.05441839991196214,A73,A153,male,old,good
+A12,0.3035714285714286,0.1384945526576428,A75,A151,male,old,bad
+A12,0.24999999999999997,0.33069219764498736,A74,A151,male,old,good
+A11,0.24999999999999997,0.027511830086937385,A71,A152,female,old,bad
+A14,0.14285714285714285,0.06894464619786508,A73,A151,male,old,good
+A11,0.14285714285714285,0.024815670738417523,A73,A152,male,old,good
+A14,0.125,0.04974138879718279,A71,A152,female,old,good
+A14,0.3035714285714286,0.16011885110597557,A75,A152,male,old,good
+A11,0.24999999999999997,0.10311433916584131,A73,A151,female,young,bad
+A12,0.14285714285714285,0.02839220864971938,A73,A152,male,old,bad
+A12,0.5714285714285714,0.3685484758446132,A73,A151,male,old,good
+A13,0.08928571428571427,0.0482007263123143,A75,A152,male,old,good
+A11,0.24999999999999997,0.05315285572796302,A72,A151,female,young,bad
+A11,0.14285714285714285,0.10806646858149004,A73,A152,male,young,good
+A14,0.3571428571428571,0.20864971937933313,A74,A152,male,old,good
+A11,0.19642857142857142,0.04418399911962144,A72,A152,male,old,good
+A12,0.3571428571428571,0.14559260482007264,A72,A152,male,old,good
+A11,0.14285714285714285,0.327335754374381,A73,A151,male,old,bad
+A12,0.3571428571428571,0.41311764058545175,A75,A151,female,old,good
+A11,0.14285714285714285,0.06173654671508748,A74,A152,male,old,bad
+A11,0.4642857142857143,0.33564432706063607,A75,A152,male,old,bad
+A12,0.24999999999999997,0.2304941124683614,A75,A152,male,old,good
+A12,0.7857142857142857,0.8485748872014967,A73,A152,male,young,bad
+A14,0.14285714285714285,0.030483107736326624,A75,A152,male,old,good
+A11,0.3035714285714286,0.07439198855507868,A75,A152,male,old,good
+A12,0.14285714285714285,0.3421371189611533,A71,A152,male,old,bad
+A14,0.24999999999999997,0.06729393639264884,A72,A152,male,old,good
+A14,0.3571428571428571,0.2851876306811929,A71,A151,female,old,good
+A14,0.03571428571428571,0.05436337625178826,A71,A152,male,old,good
+A14,0.5178571428571429,0.13832948167712117,A73,A152,female,old,good
+A12,0.14285714285714285,0.09199955981071861,A73,A152,male,old,bad
+A13,0.4642857142857143,0.1522504677011115,A75,A152,male,old,good
+A11,0.3571428571428571,0.052657642786398146,A72,A151,female,young,bad
+A11,0.21428571428571427,0.13068119291295258,A75,A151,male,old,bad
+A11,0.14285714285714285,0.023880268515461652,A73,A151,male,old,bad
+A12,0.08928571428571427,0.07813359744690217,A72,A152,female,young,bad
+A14,0.03571428571428571,0.02321998459337515,A71,A152,female,old,good
+A14,0.3035714285714286,0.12820512820512822,A72,A152,male,old,bad
+A14,0.3571428571428571,0.3336634752943766,A71,A152,male,old,good
+A11,0.3571428571428571,0.1641906019588423,A72,A151,female,young,bad
+A12,0.10714285714285712,0.3883569935072081,A71,A153,male,old,good
+A12,0.1607142857142857,0.10184879498184218,A72,A152,female,young,good
+A11,0.5714285714285714,0.32898646417959726,A72,A151,female,young,bad
+A13,0.14285714285714285,0.15219544404093763,A73,A152,male,young,good
+A14,0.14285714285714285,0.014966435567293938,A75,A152,male,old,good
+A14,0.14285714285714285,0.0926598437328051,A75,A152,male,old,good
+A14,0.7857142857142857,0.1376141740948608,A75,A152,male,old,good
+A12,0.19642857142857142,0.06943985913942995,A73,A152,male,old,bad
+A12,0.4642857142857143,0.12396830637173985,A75,A152,male,old,good
+A14,0.3571428571428571,0.05705953560030813,A75,A152,female,old,good
+A12,0.5714285714285714,0.6679322108506658,A73,A153,male,old,bad
+A13,0.08928571428571427,0.05981071860900186,A72,A152,male,old,bad
+A11,0.4107142857142857,0.12061186310113349,A75,A152,male,old,good
+A14,0.10714285714285712,0.06129635743369649,A73,A152,female,old,good
+A11,0.3571428571428571,0.11742049081104876,A75,A151,male,old,good
+A11,0.08928571428571427,0.1402553097832068,A73,A151,male,old,good
+A13,0.24999999999999997,0.06575327390778034,A74,A152,male,old,good
+A14,0.03571428571428571,0.08215032463959503,A73,A152,male,old,good
+A13,0.10714285714285712,0.06036095521074061,A74,A152,male,old,good
+A11,0.3571428571428571,0.25420931000330144,A73,A153,male,old,bad
+A14,0.14285714285714285,0.0817651590183779,A74,A152,female,old,good
+A12,0.14285714285714285,0.027731924727632886,A75,A152,male,old,good
+A14,0.5714285714285714,0.18075272367117862,A74,A152,male,old,good
+A14,0.19642857142857142,0.0675690546935182,A73,A152,male,old,good
+A14,0.4642857142857143,0.35721360184879497,A74,A152,male,old,good
+A14,0.5714285714285714,0.37839771101573677,A72,A151,female,young,bad
+A11,0.19642857142857142,0.030593155056674374,A73,A152,female,young,good
+A13,0.3571428571428571,0.14537251017937713,A75,A153,male,old,good
+A12,0.5714285714285714,0.30538131396500495,A73,A152,male,old,good
+A14,0.3571428571428571,0.15709254979641246,A72,A152,male,young,good
+A14,0.19642857142857142,0.06250687795752173,A73,A151,male,old,good
+A14,0.24999999999999997,0.11852096401452623,A73,A152,female,old,good
+A12,0.3571428571428571,0.19962583911081766,A72,A152,male,old,good
+A12,0.3571428571428571,0.15637724221415208,A72,A151,male,young,bad
+A14,0.4285714285714286,0.1371739848134698,A75,A152,male,old,good
+A14,0.14285714285714285,0.07285132607021019,A73,A152,male,old,good
+A11,0.3035714285714286,0.0872125013755915,A73,A152,female,young,bad
+A12,0.3571428571428571,0.13579839330912294,A73,A151,female,young,bad
+A14,0.7857142857142857,0.2527786948387807,A71,A151,male,old,bad
+A11,0.3571428571428571,0.052217453505007144,A75,A152,male,old,bad
+A11,0.17857142857142858,0.20485308682733577,A71,A153,male,young,good
+A14,0.3571428571428571,0.1336524705623418,A72,A152,female,old,good
+A14,0.14285714285714285,0.10360955210740619,A75,A153,female,old,good
+A11,0.3571428571428571,0.05397821063057115,A75,A151,female,old,good
+A14,0.3571428571428571,0.14427203697589966,A75,A152,male,old,good
+A12,0.24999999999999997,0.11978650819852536,A75,A152,male,old,good
+A11,1.0,0.3877517332452955,A75,A151,male,old,bad
+A14,0.4642857142857143,0.40420380763728403,A75,A152,male,old,good
+A11,0.625,0.7664245625619016,A74,A152,male,old,good
+A12,0.08928571428571427,0.06883459887751733,A75,A152,female,old,bad
+A12,0.7857142857142857,0.19775503466490593,A74,A153,male,old,bad
+A14,0.6785714285714286,0.20864971937933313,A73,A152,male,old,good
+A11,0.03571428571428571,0.06184659403543523,A71,A152,female,old,good
+A14,0.3571428571428571,0.41619896555518876,A74,A152,male,old,good
+A14,0.5714285714285714,0.03626059205458347,A75,A152,male,old,good
+A14,0.24999999999999997,0.34158688235941453,A75,A152,male,old,bad
+A14,0.3571428571428571,0.11962143721800374,A75,A152,male,old,good
+A14,0.19642857142857142,0.13965004952129417,A74,A152,female,young,good
+A11,0.14285714285714285,0.024595576097722022,A72,A152,male,old,bad
+A14,0.3035714285714286,0.12248266754704522,A75,A152,male,old,good
+A12,0.14285714285714285,0.0706503796632552,A72,A151,male,young,bad
+A12,0.03571428571428571,0.04500935402222955,A75,A152,male,old,good
+A12,0.14285714285714285,0.18284362275778587,A73,A152,female,young,good
+A12,0.24999999999999997,0.05782986684274238,A75,A152,male,old,good
+A14,0.6785714285714286,0.3805436337625179,A74,A151,male,old,good
+A12,0.3571428571428571,0.09843732805106195,A72,A152,male,young,bad
+A11,0.3571428571428571,0.12286783316826234,A73,A152,male,young,good
+A13,0.3571428571428571,0.18526466380543635,A75,A151,male,young,good
+A14,0.3571428571428571,0.09948277759436558,A73,A152,male,old,good
+A11,0.3571428571428571,0.08281060856168151,A75,A152,female,old,good
+A14,0.3571428571428571,0.07202597116760207,A74,A152,male,old,good
+A11,0.3571428571428571,0.16017387476614944,A73,A151,male,old,bad
+A14,0.7857142857142857,0.5486959392538792,A74,A152,male,old,good
+A11,0.3571428571428571,0.062231759656652355,A73,A152,female,old,bad
+A14,0.3035714285714286,0.10955210740618465,A75,A152,male,old,good
+A14,0.3571428571428571,0.12589413447782546,A75,A152,male,old,bad
+A11,0.5714285714285714,0.4300099042588313,A73,A152,female,young,bad
+A12,0.7857142857142857,0.33927588863211183,A75,A151,female,old,bad
+A12,0.3571428571428571,0.24683613954000222,A72,A152,female,young,bad
+A14,0.14285714285714285,0.037416088918234836,A75,A152,male,old,good
+A11,0.3571428571428571,0.1463629360625069,A74,A153,male,old,good
+A12,0.24999999999999997,0.11158798283261802,A72,A152,female,old,bad
+A14,0.5714285714285714,0.2318146803125344,A73,A152,male,old,bad
+A14,0.5714285714285714,0.11797072741278751,A73,A152,female,young,good
+A14,0.3571428571428571,0.1103224386486189,A72,A152,male,old,good
+A14,0.625,0.4587872785297678,A75,A152,male,old,good
+A12,0.3571428571428571,0.3022449653350941,A72,A153,female,young,good
+A11,0.14285714285714285,0.05227247716518102,A73,A151,female,young,good
+A14,0.4642857142857143,0.08897325850115549,A75,A152,male,old,good
+A11,0.03571428571428571,0.0335644327060636,A75,A152,female,old,good
+A11,0.3571428571428571,0.04258831297457907,A72,A152,male,old,bad
+A11,1.0,0.36238582590513924,A75,A152,male,old,bad
+A11,0.14285714285714285,0.07868383404864092,A75,A152,male,old,good
+A14,0.3571428571428571,0.06399251678221635,A73,A152,male,old,good
+A12,0.07142857142857142,0.03615054473423572,A72,A152,male,old,good
+A11,0.3571428571428571,0.05188731154396391,A71,A151,female,old,bad
+A11,0.14285714285714285,0.1280400572246066,A73,A152,male,old,good
+A14,0.3571428571428571,0.2047980631671619,A74,A151,female,young,good
+A12,0.3571428571428571,0.1930229998899527,A71,A151,female,young,good
+A14,0.24999999999999997,0.11389897655992078,A71,A152,male,old,good
+A14,0.3571428571428571,0.12176735996478487,A74,A152,male,old,good
+A13,0.03571428571428571,0.06014086057004511,A75,A152,male,old,good
+A14,0.4285714285714286,0.4167492021569275,A72,A151,male,old,good
+A14,0.3035714285714286,0.16644657202597118,A75,A152,male,old,good
+A12,0.24999999999999997,0.16474083856058105,A73,A152,female,old,good
+A14,0.4107142857142857,0.46016287003411466,A73,A152,male,old,good
+A11,0.5714285714285714,0.08759766699680863,A72,A152,female,old,bad
+A14,0.3571428571428571,0.12991086167051832,A75,A152,male,old,good
+A12,0.08928571428571427,0.09794211510949709,A74,A152,male,young,good
+A14,0.3571428571428571,0.49669858038956755,A73,A153,male,old,good
+A14,0.14285714285714285,0.045449543303620554,A73,A152,male,old,good
+A14,0.10714285714285712,0.07131066358534169,A73,A152,male,old,good
+A11,0.14285714285714285,0.09040387366567623,A73,A152,female,old,good
+A14,0.14285714285714285,0.10404974138879718,A74,A152,male,old,good
+A14,0.5714285714285714,0.425332893144052,A72,A151,male,old,bad
+A14,0.14285714285714285,0.055793991416309016,A75,A151,male,old,good
+A11,0.24999999999999997,0.10999229668757565,A74,A152,male,old,good
+A11,0.7857142857142857,0.324254429404644,A75,A153,female,old,bad
+A11,0.3035714285714286,0.1842742379223066,A74,A151,female,old,good
+A12,0.19642857142857142,0.07081545064377683,A75,A152,male,old,good
+A14,0.5714285714285714,0.590073731704633,A71,A152,female,old,bad
+A12,0.24999999999999997,0.2226807527236712,A75,A152,male,old,bad
+A11,0.24999999999999997,0.05172224056344228,A71,A153,female,old,bad
+A11,0.14285714285714285,0.056344228018047754,A72,A152,female,old,bad
+A14,0.08928571428571427,0.11863101133487398,A71,A152,male,old,good
+A12,0.3571428571428571,0.33856058104985143,A72,A152,male,old,good
+A14,0.10714285714285712,0.05705953560030813,A75,A152,male,old,good
+A12,0.6785714285714286,0.3138549576317817,A74,A152,female,old,good
+A11,0.03571428571428571,0.09392538791680423,A74,A152,female,old,good
+A14,0.3571428571428571,0.07164080554638494,A74,A152,male,old,good
+A14,0.3571428571428571,0.03780125453945196,A72,A152,male,old,good
+A14,0.3571428571428571,0.09805216242984482,A75,A153,male,old,good
+A14,0.5714285714285714,0.5846263893474194,A75,A152,male,old,good
+A12,0.3571428571428571,0.2659843732805106,A75,A152,female,old,good
+A11,0.5714285714285714,0.2712116209970287,A74,A152,male,old,bad
+A12,0.03571428571428571,0.044018928139099814,A71,A152,male,old,good
+A13,0.5714285714285714,0.2015516672169033,A73,A152,male,young,good
+A13,0.24999999999999997,0.15401122482667548,A72,A152,female,old,good
+A11,0.5714285714285714,0.44486629250577747,A75,A153,male,old,bad
+A11,0.14285714285714285,0.08847804555959062,A72,A151,female,young,good
+A11,0.14285714285714285,0.25272367117860683,A75,A151,male,old,bad
+A14,0.24999999999999997,0.19423352041377795,A72,A152,male,old,good
+A12,0.14285714285714285,0.09783206778914934,A73,A152,male,old,good
+A14,0.3571428571428571,0.11191812479366127,A74,A152,male,old,good
+A12,0.19642857142857142,0.1363486299108617,A74,A152,male,old,good
+A11,0.4107142857142857,0.27748431825685044,A71,A152,male,old,bad
+A12,1.0,0.4900957411687025,A73,A153,male,old,good
+A14,0.24999999999999997,0.09315505667436998,A72,A152,female,young,bad
+A11,0.3571428571428571,0.148949048090679,A75,A152,male,old,good
+A11,0.24999999999999997,0.11769560911191813,A72,A152,female,old,good
+A11,0.24999999999999997,0.08979861340376362,A73,A151,female,young,bad
+A12,0.4642857142857143,0.08060966215472654,A73,A152,female,old,good
+A11,0.7857142857142857,0.33459887751733247,A75,A153,male,old,bad
+A14,0.19642857142857142,0.0586552217453505,A73,A152,male,old,good
+A11,0.03571428571428571,0.8058765269065697,A75,A152,male,old,bad
+A14,0.4642857142857143,0.15555188731154398,A75,A152,male,old,good
+A14,0.0,0.07120061626499395,A74,A152,male,old,good
+A12,0.14285714285714285,0.041652910751623196,A73,A152,male,young,good
+A14,0.10714285714285712,0.10008803785627819,A73,A152,male,old,good
+A12,0.24999999999999997,0.7002310993727302,A71,A153,female,old,bad
+A11,0.24999999999999997,0.1570375261362386,A74,A152,male,old,good
+A12,0.5714285714285714,0.20578848905029165,A72,A152,female,old,good
+A11,0.14285714285714285,0.17877187190491914,A73,A152,female,old,bad
+A12,0.08928571428571427,0.048750962914053037,A75,A153,male,old,bad
+A13,0.19642857142857142,0.007813359744690218,A72,A151,female,young,good
+A11,0.24999999999999997,0.048475844613183675,A71,A152,female,old,bad
+A12,0.14285714285714285,0.06459777704412897,A74,A152,male,old,good
+A11,0.6785714285714286,0.17299438758666227,A71,A152,male,old,good
+A11,0.7857142857142857,0.3713546825134808,A74,A152,male,old,bad
+A11,0.4642857142857143,0.6464179597226807,A72,A152,male,old,bad
+A14,0.4642857142857143,0.14201606690877078,A73,A152,female,old,good
+A12,0.3571428571428571,0.09436557719819522,A73,A151,female,old,good
+A14,0.3571428571428571,0.07411687025420931,A75,A153,male,old,good
+A12,0.03571428571428571,0.09975789589523495,A72,A151,male,old,good
+A12,0.3928571428571428,0.42456256190161773,A72,A152,male,old,good
+A12,0.7857142857142857,0.4034334763948498,A71,A153,male,old,good
+A14,0.3571428571428571,0.24369979091009136,A74,A152,male,old,good
+A14,0.14285714285714285,0.30543633762517886,A75,A151,male,old,good
+A14,0.3571428571428571,0.1990756025090789,A75,A151,female,old,good
+A14,0.14285714285714285,0.11775063277209201,A75,A152,male,old,good
+A12,0.19642857142857142,0.3631561571475735,A71,A152,male,old,bad
+A14,0.08928571428571427,0.0675690546935182,A74,A152,male,young,bad
+A14,0.03571428571428571,0.10069329811819082,A73,A152,male,young,good
+A12,0.03571428571428571,0.03747111257840871,A72,A152,female,old,bad
+A14,0.08928571428571427,0.05469351821283151,A75,A151,female,old,good
+A11,0.10714285714285712,0.10355452844723231,A72,A151,female,old,good
+A14,0.24999999999999997,0.0935402222955871,A74,A152,male,old,good
+A11,0.03571428571428571,0.06184659403543523,A73,A152,male,old,good
+A12,0.03571428571428571,0.06597336854847584,A75,A152,male,old,good
+A14,0.24999999999999997,0.07329151535160118,A75,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-test.csv
new file mode 100644
index 0000000..1b64b92
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-test.csv
@@ -0,0 +1,201 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A13,0.5714285714285714,0.30802244965335096,A73,A152,male,young,good
+A11,0.5714285714285714,0.11290855067679102,A73,A151,male,old,bad
+A12,0.14285714285714285,0.04693518212831517,A74,A152,male,old,good
+A14,0.24999999999999997,0.0704302850225597,A73,A152,male,old,bad
+A11,0.19642857142857142,0.08093980411576977,A72,A152,male,old,good
+A14,0.24999999999999997,0.1389347419390338,A74,A152,male,old,bad
+A13,0.14285714285714285,0.08974358974358974,A73,A151,female,old,good
+A14,0.19642857142857142,0.15291075162319798,A74,A152,male,old,good
+A11,0.14285714285714285,0.07741828986464179,A73,A152,male,old,good
+A12,0.3035714285714286,0.11571475734565863,A73,A152,male,old,good
+A11,0.24999999999999997,0.13255199735886433,A73,A152,male,old,good
+A14,0.14285714285714285,0.07384175195333993,A74,A152,female,old,good
+A13,0.3571428571428571,0.09216463079124024,A73,A152,male,old,good
+A14,0.5714285714285714,0.594475624518543,A73,A152,male,old,good
+A12,0.3571428571428571,0.09447562451854297,A75,A152,female,young,good
+A12,0.5714285714285714,0.13321228128095083,A73,A153,female,old,bad
+A14,0.3571428571428571,0.17497523935292178,A75,A152,male,old,good
+A13,0.5714285714285714,0.23236491691427316,A75,A152,male,old,good
+A14,0.14285714285714285,0.24347969626939586,A72,A151,female,young,good
+A14,0.3571428571428571,0.041212721470232194,A75,A152,male,young,good
+A12,0.7857142857142857,0.5893584241223726,A74,A152,male,old,bad
+A12,0.14285714285714285,0.05876526906569825,A75,A152,male,old,good
+A11,0.03571428571428571,0.019753494002421042,A74,A152,female,old,good
+A14,0.3035714285714286,0.1509298998569385,A73,A152,male,old,good
+A14,1.0,0.847529437658193,A74,A152,male,young,good
+A14,0.19642857142857142,0.17156377242214152,A75,A151,male,young,good
+A11,0.7857142857142857,0.24969736986904373,A74,A152,male,old,good
+A11,0.14285714285714285,0.10982722570705403,A75,A152,male,old,bad
+A11,0.3571428571428571,0.08237041928029053,A72,A152,male,young,good
+A12,0.7857142857142857,0.2666446572025971,A73,A152,female,old,bad
+A12,0.24999999999999997,0.034885000550236606,A75,A152,male,old,bad
+A14,0.24999999999999997,0.32023770221195114,A75,A152,male,old,good
+A11,0.3571428571428571,0.06228678331682623,A74,A152,male,old,good
+A12,0.03571428571428571,0.02767690106745901,A73,A152,female,old,good
+A14,0.08928571428571427,0.13464289644547156,A73,A152,male,old,good
+A12,0.03571428571428571,0.01870804445911742,A72,A152,male,old,good
+A12,0.4642857142857143,0.09183448883019699,A72,A152,male,old,bad
+A11,0.3035714285714286,0.07686805326290305,A73,A152,male,old,bad
+A12,0.24999999999999997,0.3090128755364807,A73,A152,male,old,good
+A14,0.3571428571428571,0.3132496973698691,A72,A152,female,old,bad
+A11,0.14285714285714285,0.026246285902938263,A72,A152,male,old,bad
+A11,0.3571428571428571,0.14097061736546715,A75,A151,female,old,good
+A11,0.7857142857142857,0.22328601298558381,A72,A151,female,young,bad
+A11,0.4642857142857143,0.3266754704522945,A74,A151,male,young,good
+A12,0.24999999999999997,0.05749972488169913,A72,A152,female,old,good
+A14,0.5714285714285714,0.3004842082095301,A75,A152,male,old,good
+A14,0.7857142857142857,0.32243864861890614,A73,A153,male,old,good
+A12,0.4642857142857143,0.44767249917464513,A74,A152,male,old,bad
+A13,0.24999999999999997,0.09414548255749973,A75,A152,female,young,good
+A12,0.14285714285714285,0.04996148343787829,A73,A152,male,old,good
+A12,0.14285714285714285,0.06382744580169472,A73,A152,male,old,good
+A14,0.5178571428571429,0.385330692197645,A74,A152,male,old,good
+A13,0.03571428571428571,0.023825244855287777,A72,A152,female,old,good
+A14,0.14285714285714285,0.06399251678221635,A74,A152,male,old,good
+A11,0.5714285714285714,0.2817761637504127,A73,A152,male,old,good
+A12,0.4642857142857143,0.2217453505007153,A73,A151,female,old,bad
+A11,0.3571428571428571,0.06096621547265324,A75,A152,male,old,bad
+A13,0.10714285714285712,0.0292175635523275,A75,A153,male,old,good
+A14,0.0,0.17222405634422802,A74,A152,female,old,good
+A12,0.08928571428571427,0.045779685264663805,A75,A152,male,old,good
+A12,0.5714285714285714,0.11483437878287663,A75,A152,male,old,good
+A14,0.053571428571428575,0.03279410146362936,A75,A153,male,old,good
+A12,0.7857142857142857,0.45322988885220644,A74,A152,female,young,good
+A13,0.19642857142857142,0.09106415758776273,A75,A151,male,old,good
+A12,0.14285714285714285,0.08550676791020138,A72,A152,male,old,good
+A11,0.14285714285714285,0.022119511389897654,A75,A151,female,young,good
+A12,0.14285714285714285,0.059480576647958625,A72,A152,male,young,bad
+A14,0.5714285714285714,0.15566193463189174,A73,A152,male,old,good
+A13,0.19642857142857142,0.1340926598437328,A74,A151,male,old,good
+A13,0.14285714285714285,0.05942555298778475,A72,A152,male,old,good
+A11,0.14285714285714285,0.05051172003961703,A73,A152,male,old,good
+A11,0.5714285714285714,0.17585561791570375,A75,A152,male,old,bad
+A12,0.08928571428571427,0.052217453505007144,A74,A152,female,old,good
+A11,0.14285714285714285,0.02701661714537251,A71,A152,female,young,bad
+A12,0.24999999999999997,0.39198855507868385,A71,A152,male,old,good
+A14,0.24999999999999997,0.10559040387366568,A73,A152,male,old,bad
+A14,0.03571428571428571,0.05986574226917574,A73,A152,male,old,good
+A11,0.3571428571428571,0.33944095961263343,A75,A153,female,old,good
+A11,0.19642857142857142,0.039892153626059204,A73,A152,male,young,good
+A14,0.3571428571428571,0.13392758886321118,A73,A152,male,old,good
+A14,0.3571428571428571,0.23528117090348852,A73,A152,male,old,good
+A12,0.14285714285714285,0.038571585781886214,A72,A151,female,old,bad
+A12,0.8928571428571428,0.8635963464289644,A72,A151,male,old,bad
+A14,0.19642857142857142,0.18190822053483,A73,A152,male,old,good
+A14,0.0,0.06630351050951909,A74,A152,male,old,good
+A12,0.3571428571428571,0.09271486739297898,A75,A152,male,old,bad
+A12,0.14285714285714285,0.1581379993397161,A72,A152,male,old,good
+A14,0.03571428571428571,0.07114559260482008,A73,A152,male,old,good
+A11,0.3571428571428571,0.14933421371189612,A75,A153,male,old,good
+A11,0.24999999999999997,0.06558820292725871,A74,A153,male,old,bad
+A14,1.0,0.7431495543083525,A75,A153,male,old,good
+A12,0.2857142857142857,0.3245295477055134,A75,A152,male,old,good
+A14,0.3571428571428571,0.3336084516342027,A75,A152,male,old,good
+A11,0.14285714285714285,0.41900517222405637,A75,A153,male,old,bad
+A13,0.14285714285714285,0.05760977220204688,A73,A151,male,young,good
+A11,0.24999999999999997,0.121712336304611,A73,A152,male,young,bad
+A14,0.24999999999999997,0.3582590513920986,A73,A151,male,old,bad
+A14,0.5714285714285714,0.5129305601408606,A72,A152,male,old,bad
+A14,0.5714285714285714,0.39391438318476946,A75,A152,male,old,good
+A11,0.5714285714285714,0.36519203257400684,A73,A152,male,old,bad
+A14,0.3571428571428571,0.18548475844613185,A75,A152,male,old,bad
+A11,0.03571428571428571,0.06063607351160999,A71,A151,female,young,good
+A13,0.14285714285714285,0.06767910201386595,A71,A153,male,old,good
+A14,0.24999999999999997,0.042808407615274574,A73,A152,female,old,good
+A11,0.14285714285714285,0.035875426433366345,A74,A151,male,young,bad
+A11,0.10714285714285712,0.1136238582590514,A75,A152,male,old,good
+A14,0.08928571428571427,0.09519093210080334,A72,A151,female,young,bad
+A12,0.3571428571428571,0.07169582920655881,A74,A151,female,young,good
+A13,0.10714285714285712,0.05447342357213601,A75,A153,female,old,bad
+A12,0.14285714285714285,0.05353802134918015,A75,A151,male,old,bad
+A11,0.3035714285714286,0.12963574336964895,A72,A151,female,old,good
+A11,0.14285714285714285,0.06377242214152085,A75,A152,male,old,good
+A14,0.24999999999999997,0.07252118410916694,A73,A151,female,young,good
+A14,0.01785714285714285,0.1759656652360515,A74,A152,male,old,good
+A12,0.24999999999999997,0.1992956971497744,A71,A152,female,old,good
+A12,0.14285714285714285,0.116430064927919,A74,A152,male,old,good
+A14,0.5714285714285714,0.4087157477715418,A74,A152,female,old,good
+A12,0.5714285714285714,0.5033564432706064,A72,A151,male,old,bad
+A14,0.5714285714285714,0.5126004181798173,A73,A152,female,old,good
+A11,0.3571428571428571,0.0928799383735006,A72,A152,male,old,bad
+A11,0.14285714285714285,0.10448993067018818,A73,A153,male,old,bad
+A14,0.5714285714285714,0.350170573346539,A75,A152,male,old,good
+A14,0.14285714285714285,0.14355672939363928,A71,A152,male,old,good
+A12,0.2857142857142857,0.13090128755364808,A73,A152,male,old,good
+A12,0.24999999999999997,0.32761087267525035,A73,A152,male,old,good
+A11,0.03571428571428571,0.18851105975569496,A73,A151,male,old,good
+A12,0.19642857142857142,0.114229118520964,A73,A152,male,old,good
+A11,0.03571428571428571,0.17244415098492352,A73,A151,male,old,bad
+A12,0.7857142857142857,0.1546164850885881,A74,A152,male,old,bad
+A13,0.03571428571428571,0.05771981952239463,A73,A152,male,old,good
+A14,0.10714285714285712,0.06426763508308572,A73,A151,male,old,good
+A14,0.14285714285714285,0.04104765048971058,A74,A152,female,young,good
+A14,0.2857142857142857,0.1780015406624849,A72,A152,male,old,good
+A14,0.24999999999999997,0.17211400902388027,A73,A152,male,old,good
+A14,0.08928571428571427,0.05359304500935402,A73,A152,male,old,good
+A13,0.08928571428571427,0.02723671178606801,A73,A152,female,old,bad
+A12,0.14285714285714285,0.08858809287993837,A71,A152,male,old,good
+A11,0.2857142857142857,0.2213051612193243,A75,A152,female,young,good
+A12,0.053571428571428575,0.11912622427643886,A73,A152,male,old,good
+A14,0.3571428571428571,0.19406844943325632,A73,A152,male,old,good
+A12,0.4642857142857143,0.10625068779575217,A75,A152,male,old,good
+A14,0.24999999999999997,0.239022779795312,A72,A152,male,old,good
+A11,0.14285714285714285,0.028007043028502255,A74,A152,male,old,bad
+A14,0.4642857142857143,0.23550126554418402,A74,A151,female,old,good
+A14,0.10714285714285712,0.09210960713106636,A73,A152,male,old,good
+A14,0.14285714285714285,0.17591064157587763,A73,A152,female,old,good
+A12,1.0,0.7580609662154726,A74,A152,male,old,bad
+A13,0.24999999999999997,0.14383184769450866,A73,A152,male,old,bad
+A14,0.14285714285714285,0.05529877847474414,A75,A152,male,old,good
+A11,0.6785714285714286,0.22669747991636405,A74,A152,male,old,bad
+A14,0.14285714285714285,0.03719599427753935,A71,A152,female,old,good
+A12,0.19642857142857142,0.030373060415978873,A75,A152,male,old,bad
+A12,0.14285714285714285,0.04632992186640256,A73,A152,female,old,good
+A14,0.3035714285714286,0.07378672829316606,A74,A152,male,old,good
+A14,0.5714285714285714,0.23131946737096953,A73,A152,female,old,good
+A14,0.14285714285714285,0.023770221195113902,A74,A152,female,old,good
+A12,0.24999999999999997,0.3362495873225487,A75,A152,male,old,good
+A14,0.14285714285714285,0.055463849455265765,A72,A151,female,young,good
+A12,0.4107142857142857,0.12490370859469573,A73,A152,male,young,bad
+A11,0.03571428571428571,0.00484208209530098,A75,A152,male,old,good
+A12,0.2857142857142857,0.37454605480356556,A74,A151,male,old,good
+A11,0.2857142857142857,0.10922196544514141,A73,A151,male,old,bad
+A13,0.14285714285714285,0.06459777704412897,A75,A152,female,old,good
+A12,0.3571428571428571,0.6095521074061846,A73,A152,male,old,bad
+A14,0.3571428571428571,0.05496863651370089,A72,A152,male,old,good
+A12,0.08928571428571427,0.10278419720479806,A73,A152,male,old,good
+A11,0.3571428571428571,0.05694948827996038,A74,A151,female,old,bad
+A14,0.7857142857142857,0.18311874105865522,A75,A152,male,old,good
+A14,0.10714285714285712,0.14295146913172666,A73,A152,male,old,good
+A11,0.24999999999999997,0.17932210850665786,A74,A152,female,young,good
+A11,0.07142857142857142,0.05029162539892153,A75,A153,male,old,good
+A12,0.3035714285714286,0.2050181578078574,A74,A152,male,old,good
+A12,0.24999999999999997,0.18498954550456698,A75,A152,female,old,good
+A12,1.2142857142857142,0.2941014636293606,A73,A152,male,young,bad
+A11,0.14285714285714285,0.025255860019808517,A75,A152,male,old,bad
+A11,0.5714285714285714,0.43903378452734676,A73,A152,male,old,bad
+A11,0.3035714285714286,0.17409486079013978,A72,A152,male,old,bad
+A12,0.5714285714285714,0.6802024870694399,A73,A153,male,old,bad
+A14,0.03571428571428571,0.08253549026081215,A75,A152,male,old,good
+A14,0.14285714285714285,0.15555188731154398,A73,A152,male,old,good
+A14,0.19642857142857142,0.2791900517222406,A75,A153,female,old,good
+A13,0.0,0.06844943325630021,A72,A152,male,old,good
+A11,0.14285714285714285,0.12809508088478047,A71,A153,female,old,good
+A12,0.053571428571428575,0.11439418950148562,A72,A152,female,old,good
+A11,0.08928571428571427,0.022229558710245404,A73,A152,male,old,bad
+A14,0.5714285714285714,0.4337515131506548,A73,A152,female,old,good
+A14,0.3571428571428571,0.19880048420820953,A73,A153,male,old,good
+A12,0.14285714285714285,0.39738087377572356,A71,A151,female,young,good
+A14,0.14285714285714285,0.07180587652690656,A75,A153,male,old,bad
+A11,0.14285714285714285,0.055683944095961266,A75,A152,male,old,good
+A12,0.08928571428571427,0.02971277649389237,A73,A152,female,old,good
+A12,0.4642857142857143,0.27423792230659183,A71,A152,male,old,bad
+A11,0.4642857142857143,0.11863101133487398,A74,A151,female,young,bad
+A12,0.14285714285714285,0.1496093320127655,A72,A151,female,young,bad
+A14,0.3035714285714286,0.6839440959612633,A75,A153,male,old,bad
+A11,0.14285714285714285,0.045779685264663805,A73,A152,male,old,bad
+A12,0.4107142857142857,0.20166171453725104,A73,A152,male,old,bad
+A14,0.08928571428571427,0.19709475074281943,A75,A152,male,old,good
+A12,0.24999999999999997,0.09232970177176185,A72,A152,male,old,bad
+A12,0.5714285714285714,0.5286123032904149,A74,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-train.csv
new file mode 100644
index 0000000..b15c2c9
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-train.csv
@@ -0,0 +1,501 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,0.5714285714285714,0.20408275558490152,A71,A152,male,old,good
+A14,0.08928571428571427,0.18306371739848135,A73,A151,male,old,good
+A14,0.24999999999999997,0.12462859029382635,A73,A152,male,old,good
+A12,0.14285714285714285,0.09601628700341147,A72,A152,male,old,good
+A14,1.0,0.5566193463189171,A75,A152,male,old,good
+A14,0.3571428571428571,0.39303400462198745,A73,A152,male,old,good
+A13,0.14285714285714285,0.008748761967646089,A73,A151,female,old,good
+A14,0.08928571428571427,0.12418840101243535,A75,A153,male,old,good
+A14,0.4642857142857143,0.25096291405304283,A74,A151,female,young,good
+A12,0.7857142857142857,0.3471992956971498,A74,A152,male,young,bad
+A12,0.5714285714285714,0.48332783096731596,A72,A151,male,old,bad
+A11,0.3571428571428571,0.1816881258941345,A74,A152,male,old,bad
+A12,0.2857142857142857,0.3421371189611533,A71,A152,male,old,good
+A11,0.4642857142857143,0.1250137559150435,A75,A152,male,old,good
+A12,0.5714285714285714,0.14141080664685815,A72,A152,male,old,bad
+A14,0.03571428571428571,0.029327610872675252,A73,A152,male,old,good
+A11,0.3035714285714286,0.19329811819082207,A74,A152,male,young,good
+A14,0.4642857142857143,0.39809618135798397,A71,A152,female,old,bad
+A11,0.19642857142857142,0.06344228018047761,A73,A151,female,old,good
+A12,0.125,0.24876196764608782,A74,A152,male,old,good
+A11,0.3571428571428571,0.20743919885550788,A73,A152,male,old,good
+A12,0.3571428571428571,0.052327500825354895,A72,A152,male,old,good
+A12,0.7857142857142857,0.28232640035215145,A71,A153,male,old,good
+A14,0.3571428571428571,0.28947947617475517,A73,A152,male,young,good
+A14,0.24999999999999997,0.04666006382744581,A71,A152,female,old,good
+A12,0.7857142857142857,0.32871134587872786,A75,A153,male,old,bad
+A14,0.125,0.06289204357873886,A72,A152,female,old,good
+A12,0.08928571428571427,0.09183448883019699,A74,A151,male,old,good
+A14,0.19642857142857142,0.2638384505337295,A75,A152,female,old,good
+A12,0.7857142857142857,0.3136898866512601,A73,A152,female,young,bad
+A11,0.07142857142857142,0.026466380543633764,A75,A152,male,old,good
+A14,0.24999999999999997,0.02085396720589854,A75,A152,male,old,good
+A14,0.24999999999999997,0.08572686255089688,A74,A152,female,young,bad
+A12,0.3035714285714286,0.05161219324309453,A75,A152,female,old,bad
+A14,0.3571428571428571,0.16424562561901618,A75,A152,male,old,good
+A12,0.3571428571428571,0.6223175965665235,A73,A151,female,young,bad
+A12,0.08928571428571427,0.04974138879718279,A75,A152,male,old,good
+A14,0.24999999999999997,0.19808517662594918,A74,A152,male,old,good
+A11,0.3571428571428571,0.11604489930670188,A71,A152,male,old,bad
+A14,0.3571428571428571,0.07059535600308132,A72,A152,female,old,good
+A13,0.5714285714285714,0.21789369428854408,A73,A152,male,old,bad
+A11,0.4642857142857143,0.19847034224716628,A73,A152,male,old,good
+A14,0.14285714285714285,0.06289204357873886,A75,A152,male,old,good
+A14,0.24999999999999997,0.08528667326950588,A73,A152,male,young,good
+A13,0.14285714285714285,0.1101023440079234,A73,A152,female,old,good
+A12,0.14285714285714285,0.03224386486189061,A72,A152,female,young,bad
+A11,0.24999999999999997,0.06025090789039286,A73,A152,male,old,bad
+A12,0.5714285714285714,0.2500825354902608,A72,A152,female,old,good
+A11,0.14285714285714285,0.025200836359634642,A73,A152,male,old,good
+A11,0.14285714285714285,0.05716958292065588,A73,A152,male,young,good
+A14,0.24999999999999997,0.08880818752063387,A73,A152,female,old,bad
+A12,0.7857142857142857,0.5342797402883239,A72,A152,female,old,bad
+A11,0.24999999999999997,0.27797953119841534,A75,A153,male,old,good
+A11,0.03571428571428571,0.08924837680202487,A71,A153,male,old,good
+A11,0.3571428571428571,0.12044679212061186,A72,A152,female,old,bad
+A14,0.08928571428571427,0.06261692527786948,A73,A151,female,old,good
+A12,0.7857142857142857,0.6558270056124132,A71,A153,male,old,good
+A14,0.24999999999999997,0.13271706833938596,A74,A152,male,old,good
+A14,0.1607142857142857,0.06377242214152085,A71,A152,female,old,good
+A11,0.19642857142857142,0.1866952789699571,A75,A152,female,old,good
+A11,0.7857142857142857,0.358093980411577,A73,A152,female,old,bad
+A14,0.3571428571428571,0.0880928799383735,A74,A152,male,old,good
+A11,0.4107142857142857,0.17420490811048753,A73,A152,male,old,good
+A14,0.4107142857142857,0.2678001540662485,A74,A152,male,old,good
+A14,0.10714285714285712,0.07411687025420931,A73,A151,male,old,good
+A14,0.03571428571428571,0.06228678331682623,A73,A152,female,old,good
+A14,0.10714285714285712,0.05397821063057115,A75,A152,male,old,good
+A12,0.19642857142857142,0.13101133487399583,A73,A151,female,old,bad
+A12,0.4642857142857143,0.10454495433036205,A73,A152,female,young,bad
+A12,0.14285714285714285,0.1474634092659844,A74,A152,female,old,good
+A14,0.03571428571428571,0.3582590513920986,A74,A152,male,old,good
+A12,0.3035714285714286,0.13728403213381754,A74,A152,male,old,good
+A11,0.3571428571428571,0.11417409486079015,A74,A152,male,old,good
+A14,0.03571428571428571,0.05249257180587652,A73,A151,male,old,good
+A11,0.24999999999999997,0.04396390447892594,A72,A151,female,young,good
+A14,0.14285714285714285,0.11235831407505227,A71,A152,male,old,bad
+A14,0.08928571428571427,0.06597336854847584,A74,A152,female,old,good
+A11,0.24999999999999997,0.39952679652250467,A75,A153,male,old,bad
+A12,0.08928571428571427,0.03675580499614835,A73,A152,female,old,bad
+A14,0.3571428571428571,0.1153295917244415,A74,A152,male,old,good
+A11,0.3571428571428571,0.07444701221525256,A75,A152,female,old,good
+A11,0.125,0.2011114779355123,A73,A151,male,old,good
+A14,0.24999999999999997,0.0482007263123143,A72,A151,female,young,good
+A13,0.3035714285714286,0.14707824364476726,A73,A152,female,old,good
+A12,0.08928571428571427,0.1620446792120612,A73,A152,female,old,good
+A11,0.03571428571428571,0.24573566633652472,A72,A152,male,old,good
+A12,0.03571428571428571,0.7871134587872785,A71,A152,male,young,bad
+A11,0.5714285714285714,0.2884890502916254,A75,A153,male,old,good
+A14,0.3571428571428571,0.0734565863321228,A74,A152,male,old,good
+A11,0.14285714285714285,0.18377902498074172,A73,A152,male,old,good
+A14,0.14285714285714285,0.1380543633762518,A75,A152,male,old,good
+A11,0.14285714285714285,0.10300429184549356,A73,A151,male,old,good
+A12,0.14285714285714285,0.09640145262462858,A74,A151,male,old,good
+A14,0.3571428571428571,0.09084406294706723,A73,A151,male,old,good
+A14,0.14285714285714285,0.06839440959612633,A72,A152,female,old,good
+A12,0.03571428571428571,0.052767690106745896,A71,A152,male,old,bad
+A12,0.7857142857142857,0.20518322878837902,A74,A152,male,old,good
+A11,0.3571428571428571,0.4110817651590184,A72,A152,female,old,good
+A11,0.24999999999999997,0.1870804445911742,A72,A151,female,young,good
+A11,0.24999999999999997,0.08968856608341587,A74,A152,male,old,good
+A12,0.24999999999999997,0.09304500935402223,A73,A152,male,old,good
+A14,0.19642857142857142,0.034334763948497854,A72,A152,female,young,good
+A14,0.4107142857142857,0.2718168812589414,A75,A152,male,old,good
+A14,0.5714285714285714,0.5627269725982172,A75,A153,male,old,good
+A12,0.3571428571428571,0.22565203037306042,A73,A152,female,old,good
+A14,0.3571428571428571,0.06195664135578298,A74,A152,female,old,good
+A12,0.4642857142857143,0.1786068009243975,A73,A152,male,old,good
+A11,0.6428571428571428,0.3162759986794322,A73,A152,male,old,bad
+A11,0.4285714285714286,0.20666886761307363,A73,A152,male,old,bad
+A14,0.5714285714285714,0.1701331572576208,A75,A152,male,old,good
+A12,0.24999999999999997,0.13051612193243095,A73,A152,male,old,good
+A12,0.5714285714285714,0.11742049081104876,A72,A151,male,old,bad
+A11,0.3571428571428571,0.07752833718498954,A72,A151,female,old,bad
+A14,0.3571428571428571,0.1781666116430065,A74,A152,female,young,good
+A12,1.0,0.39440959612633436,A73,A152,male,old,good
+A14,0.24999999999999997,0.10938703642566304,A73,A152,female,young,good
+A11,0.3571428571428571,0.059590623968306375,A71,A153,male,old,bad
+A14,0.3214285714285714,0.05683944095961263,A74,A151,female,young,good
+A14,0.03571428571428571,0.17981732144822274,A73,A151,male,old,good
+A12,0.14285714285714285,0.018432926158248045,A73,A151,male,young,good
+A12,0.625,0.25767580059425554,A74,A152,male,young,bad
+A14,0.3571428571428571,0.056564322658743255,A75,A152,male,old,good
+A14,0.14285714285714285,0.08352591614394189,A73,A151,male,young,good
+A11,0.3571428571428571,0.17051832287883792,A72,A153,male,old,bad
+A12,0.08928571428571427,0.0627819962583911,A73,A152,male,old,good
+A14,0.24999999999999997,0.17453505007153078,A75,A152,male,old,good
+A14,0.10714285714285712,0.10432485968966655,A72,A151,female,young,good
+A14,0.19642857142857142,0.06988004842082095,A75,A152,male,old,good
+A14,0.3571428571428571,0.3627159678661825,A73,A152,male,old,good
+A14,0.24999999999999997,0.09909761197314845,A72,A152,male,old,good
+A11,0.3571428571428571,0.15109497083746012,A73,A152,male,young,bad
+A11,0.3571428571428571,0.18763068119291296,A73,A152,female,old,good
+A11,0.3571428571428571,0.06558820292725871,A74,A151,female,young,bad
+A11,0.08928571428571427,0.04886101023440079,A73,A152,male,young,good
+A14,0.4642857142857143,0.3037856278199626,A74,A152,female,young,good
+A11,0.24999999999999997,0.12930560140860572,A73,A153,male,old,bad
+A13,0.3571428571428571,0.19252778694838782,A72,A152,female,old,good
+A14,0.14285714285714285,0.15456146142841423,A74,A152,male,old,good
+A11,0.03571428571428571,0.022669747991636405,A72,A152,male,old,good
+A11,0.24999999999999997,0.09298998569384835,A72,A153,male,old,good
+A12,0.5714285714285714,0.1140640475404424,A74,A151,male,young,good
+A12,0.7857142857142857,0.6577528337184989,A73,A152,male,old,good
+A14,0.3571428571428571,0.28925938153405967,A75,A153,male,old,good
+A14,0.24999999999999997,0.010069329811819083,A71,A151,female,young,bad
+A14,0.03571428571428571,0.23995818201826785,A72,A152,female,old,bad
+A12,0.14285714285714285,0.13821943435677342,A75,A152,female,young,bad
+A14,0.4642857142857143,0.2213051612193243,A73,A152,male,old,good
+A12,0.625,0.6333223286012986,A74,A151,male,old,good
+A12,0.5714285714285714,0.10867172884340266,A75,A153,male,old,bad
+A13,0.14285714285714285,0.06734896005282272,A72,A152,female,old,good
+A14,0.14285714285714285,0.06338725652030373,A74,A151,female,old,good
+A12,0.24999999999999997,0.09150434686915374,A72,A152,male,old,good
+A14,0.17857142857142858,0.030373060415978873,A73,A152,male,old,good
+A13,0.19642857142857142,0.05617915703752613,A73,A153,male,old,bad
+A13,0.3571428571428571,0.055463849455265765,A73,A152,female,old,good
+A14,0.14285714285714285,0.18240343347639487,A72,A152,male,old,good
+A11,0.14285714285714285,0.17255419830527127,A75,A153,male,old,bad
+A13,0.14285714285714285,0.1098822493672279,A73,A152,female,old,good
+A14,0.3571428571428571,0.30560140860570045,A73,A152,male,old,good
+A14,0.19642857142857142,0.03906679872345109,A74,A152,female,old,good
+A12,0.24999999999999997,0.044349070100143065,A75,A152,male,old,bad
+A13,0.6785714285714286,0.3322878837900297,A72,A152,male,old,good
+A14,0.08928571428571427,0.13772422141520854,A75,A152,male,old,good
+A14,0.10714285714285712,0.10784637394079453,A73,A151,male,young,bad
+A11,0.7857142857142857,0.5528227137669197,A74,A153,male,old,bad
+A12,0.5714285714285714,0.7740728513260702,A75,A153,male,old,bad
+A12,0.4642857142857143,0.2200396170353252,A71,A152,male,old,bad
+A14,0.3571428571428571,0.03736106525806096,A74,A152,male,old,good
+A12,0.08928571428571427,0.039011775063277215,A73,A152,female,old,bad
+A14,0.3035714285714286,0.13931990756025092,A74,A152,female,old,good
+A12,0.7857142857142857,0.44613183668977663,A72,A152,female,old,good
+A14,0.08928571428571427,0.11285352701661715,A72,A151,female,young,good
+A12,0.7857142857142857,0.18245845713656875,A74,A152,male,old,good
+A11,0.24999999999999997,0.20446792120611865,A75,A151,female,old,bad
+A14,0.03571428571428571,0.05430835259161439,A73,A152,female,old,good
+A11,0.24999999999999997,0.03994717728623308,A72,A152,female,young,bad
+A11,0.3571428571428571,0.03659073401562672,A75,A152,female,old,bad
+A13,0.3571428571428571,0.03835149114119071,A74,A153,male,old,bad
+A12,0.3571428571428571,0.14168592494772753,A74,A152,male,old,good
+A14,0.3035714285714286,0.11213821943435677,A72,A152,female,young,good
+A14,0.10714285714285712,0.145867723120942,A72,A151,female,old,good
+A12,0.7321428571428571,0.23803235391218225,A71,A152,male,old,good
+A14,0.7857142857142857,0.3845053372950369,A75,A152,male,old,good
+A11,0.5714285714285714,0.13733905579399142,A75,A152,male,old,bad
+A12,0.3571428571428571,0.08215032463959503,A75,A152,male,old,good
+A11,0.08928571428571427,0.06140640475404424,A72,A151,female,young,bad
+A14,0.24999999999999997,0.06905469351821283,A73,A153,male,old,good
+A11,0.03571428571428571,0.005117200396170352,A72,A152,female,old,good
+A14,0.3571428571428571,0.10856168152305491,A75,A152,male,old,good
+A14,0.14285714285714285,0.06850445691647408,A75,A152,male,old,good
+A13,0.19642857142857142,0.11428414218113787,A72,A152,female,young,bad
+A11,0.7321428571428571,0.6364036535710355,A75,A151,male,old,bad
+A12,0.14285714285714285,0.06789919665456146,A73,A152,male,young,bad
+A14,0.5714285714285714,0.4844833278309673,A73,A153,male,old,good
+A14,0.14285714285714285,0.07730824254429404,A75,A152,male,old,good
+A14,0.24999999999999997,0.06619346318917134,A72,A152,female,old,good
+A11,0.14285714285714285,0.053813139650049524,A73,A152,female,young,bad
+A12,0.03571428571428571,0.037526136238582586,A74,A152,female,old,good
+A12,0.5714285714285714,0.4311654011224827,A75,A152,male,old,bad
+A14,0.03571428571428571,0.06030593155056674,A75,A153,male,old,good
+A12,0.5714285714285714,0.1437218003741609,A75,A153,male,old,good
+A11,0.3571428571428571,0.15808297567954221,A72,A152,female,old,bad
+A12,1.0,0.3938593595245956,A72,A152,female,young,bad
+A12,0.19642857142857142,0.055793991416309016,A73,A151,male,young,bad
+A14,0.03571428571428571,0.07175085286673269,A74,A151,female,young,good
+A12,0.3035714285714286,0.18719049191152196,A74,A152,male,old,good
+A11,0.7857142857142857,0.20254209310003302,A74,A153,male,old,bad
+A11,0.03571428571428571,0.061131286453174866,A72,A152,male,old,good
+A11,0.03571428571428571,0.052437548145702645,A75,A152,male,old,good
+A14,0.14285714285714285,0.18031253438978762,A72,A152,male,old,good
+A14,0.5714285714285714,0.4184549356223176,A73,A152,female,young,bad
+A11,0.3035714285714286,0.17095851216022892,A72,A152,female,old,good
+A13,0.24999999999999997,0.10179377132166832,A73,A152,male,old,bad
+A14,0.24999999999999997,0.0505667436997909,A73,A152,male,old,good
+A14,0.5714285714285714,0.47854077253218885,A74,A152,male,old,good
+A11,0.24999999999999997,0.09276989105315285,A74,A151,male,young,good
+A12,0.3571428571428571,0.1794871794871795,A74,A152,male,old,good
+A12,0.3571428571428571,0.3695939253879168,A74,A151,male,old,good
+A14,0.4642857142857143,0.11461428414218112,A75,A152,male,old,good
+A11,0.24999999999999997,0.09843732805106195,A73,A151,female,young,bad
+A11,0.5714285714285714,0.49636843842852424,A74,A152,male,young,bad
+A14,0.3035714285714286,0.07274127874986244,A75,A152,female,old,good
+A12,0.03571428571428571,0.011224826675470454,A72,A152,male,young,good
+A14,0.5714285714285714,0.33311323869263787,A75,A152,male,old,good
+A14,0.3571428571428571,0.10713106635853417,A74,A152,male,old,good
+A14,0.7857142857142857,0.40601958842302194,A75,A152,male,old,good
+A13,0.10714285714285712,0.05639925167822163,A72,A152,female,young,good
+A11,0.7857142857142857,0.37795752173434577,A73,A153,male,old,bad
+A12,0.3571428571428571,0.6783867062837019,A75,A153,female,old,bad
+A12,0.24999999999999997,0.1391548365797293,A73,A151,male,young,good
+A14,0.0,0.06894464619786508,A74,A152,male,old,good
+A12,0.07142857142857142,0.05430835259161439,A73,A152,female,young,bad
+A14,0.7857142857142857,0.20160669087707717,A73,A152,male,old,bad
+A14,0.3571428571428571,0.15472653240893586,A75,A151,male,old,good
+A12,0.5714285714285714,0.09370529327610873,A75,A153,male,old,bad
+A12,0.4642857142857143,0.19709475074281943,A72,A152,male,young,good
+A14,0.03571428571428571,0.0708704743039507,A72,A152,female,old,good
+A14,0.03571428571428571,0.10223396060305931,A74,A151,male,old,good
+A11,0.03571428571428571,0.00979421151094971,A75,A152,female,old,good
+A12,0.5714285714285714,0.2313744910311434,A73,A152,male,old,bad
+A13,0.3035714285714286,0.1138439528997469,A72,A151,male,old,bad
+A14,0.14285714285714285,0.05023660173874765,A73,A152,male,old,good
+A14,0.08928571428571427,0.037746230879278087,A75,A152,male,old,good
+A11,0.24999999999999997,0.12231759656652359,A71,A152,male,young,bad
+A14,0.19642857142857142,0.07257620776934082,A75,A152,male,old,good
+A12,0.14285714285714285,0.1522504677011115,A72,A151,female,old,good
+A14,0.14285714285714285,0.11450423682183337,A75,A152,male,old,good
+A14,0.14285714285714285,0.10157367668097282,A74,A152,male,old,good
+A12,0.14285714285714285,0.04330362055683944,A74,A152,male,old,good
+A14,0.03571428571428571,0.05425332893144051,A73,A151,male,old,good
+A14,0.3571428571428571,0.21464729833828547,A73,A152,male,old,good
+A14,0.7857142857142857,0.18482447452404535,A73,A152,female,old,good
+A14,0.14285714285714285,0.06393749312204247,A73,A152,female,old,good
+A11,0.125,0.20298228238142402,A73,A152,male,old,good
+A11,0.14285714285714285,0.0702101903818642,A75,A153,male,old,good
+A14,0.3214285714285714,0.1334323759216463,A75,A152,male,old,good
+A12,0.4642857142857143,0.2184989545504567,A73,A152,female,old,good
+A11,0.3571428571428571,0.14713326730494114,A73,A152,male,old,good
+A14,0.24999999999999997,0.08622207549246176,A73,A152,female,old,good
+A13,0.10714285714285712,0.0536480686695279,A73,A152,male,old,good
+A11,0.19642857142857142,0.08776273797733025,A72,A151,female,old,good
+A14,0.03571428571428571,0.0,A73,A152,female,old,good
+A14,0.03571428571428571,0.022559700671288655,A74,A151,male,young,good
+A13,0.03571428571428571,0.025255860019808517,A72,A152,male,old,good
+A11,0.5714285714285714,0.8587542643336634,A71,A152,male,old,good
+A12,0.3571428571428571,0.08732254869593925,A74,A153,female,old,bad
+A11,0.1607142857142857,0.08512160228898426,A72,A152,male,old,good
+A11,0.03571428571428571,0.010894684714427203,A72,A152,female,young,bad
+A12,0.24999999999999997,0.16303510509519095,A72,A151,male,young,good
+A11,0.14285714285714285,0.12815010454495435,A72,A152,male,old,bad
+A11,0.03571428571428571,0.05216242984483327,A75,A153,female,old,bad
+A14,0.3035714285714286,0.26152745680642675,A73,A152,female,old,bad
+A12,0.3571428571428571,0.11461428414218112,A72,A152,male,old,good
+A14,0.19642857142857142,0.055683944095961266,A74,A152,male,old,good
+A12,0.14285714285714285,0.049796412457356665,A75,A152,male,old,good
+A14,0.10714285714285712,0.03543523715197534,A74,A152,female,old,good
+A14,0.14285714285714285,0.16952789699570817,A75,A152,male,old,good
+A12,0.24999999999999997,0.04748541873005392,A73,A152,female,old,good
+A14,0.14285714285714285,0.02580609662154727,A75,A152,male,old,bad
+A12,0.3571428571428571,0.09981291955540883,A71,A152,female,old,bad
+A13,0.4642857142857143,0.18741058655221746,A75,A152,male,old,good
+A11,0.3571428571428571,0.3502255970067129,A71,A153,male,old,good
+A14,0.24999999999999997,0.16391548365797293,A71,A152,male,old,good
+A14,0.19642857142857142,0.17018818091779467,A73,A153,male,old,good
+A11,0.4642857142857143,0.1572576207769341,A72,A152,male,old,bad
+A12,0.08928571428571427,0.011444921316165951,A73,A152,male,young,good
+A14,0.19642857142857142,0.06718388907230109,A73,A153,male,old,good
+A12,0.125,0.05898536370639375,A73,A152,female,old,good
+A11,0.08928571428571427,0.057114559260482006,A75,A152,male,old,good
+A14,0.14285714285714285,0.025145812699460767,A73,A152,male,old,good
+A12,0.19642857142857142,0.058215032463959496,A75,A152,male,old,good
+A14,0.08928571428571427,0.15538681633102236,A73,A152,male,old,good
+A11,0.19642857142857142,0.12440849565313085,A71,A151,female,young,good
+A14,0.03571428571428571,0.0819852536590734,A75,A151,male,old,good
+A14,0.14285714285714285,0.04803565533179267,A73,A151,female,old,bad
+A12,0.125,0.07301639705073182,A72,A152,female,young,good
+A11,0.7321428571428571,0.08776273797733025,A73,A153,male,young,bad
+A14,0.3571428571428571,0.09783206778914934,A74,A152,male,old,good
+A12,0.4107142857142857,0.4439308902828216,A75,A153,female,old,bad
+A12,0.14285714285714285,0.05749972488169913,A72,A151,female,young,bad
+A12,0.3571428571428571,0.6431165401122483,A73,A152,male,old,bad
+A14,0.5714285714285714,0.08633212281280951,A73,A153,male,old,bad
+A11,0.19642857142857142,0.0675690546935182,A75,A152,male,old,good
+A14,0.3571428571428571,0.06965995378012545,A73,A152,female,old,good
+A11,0.08928571428571427,0.1037746230879278,A73,A152,male,young,good
+A14,0.19642857142857142,0.19599427753934193,A72,A152,female,young,good
+A14,0.24999999999999997,0.21541762958071972,A73,A152,male,old,bad
+A12,0.3571428571428571,0.20947507428194126,A74,A152,male,old,bad
+A14,0.7857142857142857,0.543468691537361,A73,A153,male,old,bad
+A12,0.03571428571428571,0.012875536480686695,A74,A152,male,old,good
+A12,1.0,0.33223286012985587,A73,A153,male,old,bad
+A11,0.3571428571428571,0.36436667767139874,A72,A152,male,old,bad
+A14,0.19642857142857142,0.07054033234290745,A73,A152,female,old,good
+A14,0.08928571428571427,0.10366457576758006,A73,A152,male,old,good
+A11,0.14285714285714285,0.04715527676901067,A73,A151,male,young,good
+A11,0.2857142857142857,0.10795642126114229,A74,A152,male,old,good
+A14,0.19642857142857142,0.16154946627049632,A74,A151,female,young,good
+A11,0.7678571428571428,0.5762077693408165,A72,A152,female,old,good
+A14,0.053571428571428575,0.02641135688345989,A75,A151,male,old,good
+A14,0.14285714285714285,0.10030813249697369,A73,A152,female,old,good
+A11,0.3035714285714286,0.0176625949158138,A75,A152,male,old,good
+A13,0.19642857142857142,0.11609992296687575,A73,A152,male,old,good
+A11,0.14285714285714285,0.06019588423021899,A73,A152,male,old,good
+A11,0.6785714285714286,0.3809838230439089,A74,A152,female,old,bad
+A13,0.6785714285714286,0.2501375591504347,A75,A153,male,old,good
+A11,0.6785714285714286,0.20441289754594477,A72,A152,male,old,bad
+A14,0.125,0.38395510069329813,A73,A152,male,old,good
+A12,0.3571428571428571,0.05315285572796302,A72,A152,male,old,bad
+A14,0.14285714285714285,0.13744910311433917,A75,A153,female,old,good
+A12,0.6785714285714286,0.49702872235061074,A71,A153,male,old,good
+A14,0.4107142857142857,0.23528117090348852,A72,A152,male,old,good
+A14,0.3035714285714286,0.2750082535490261,A73,A152,male,old,good
+A14,0.3571428571428571,0.16127434796962695,A72,A152,female,old,good
+A14,0.3571428571428571,0.02569604930119952,A75,A152,male,old,good
+A14,0.14285714285714285,0.055683944095961266,A73,A152,male,young,good
+A14,0.14285714285714285,0.08990866072411137,A75,A152,male,old,good
+A12,0.24999999999999997,0.09210960713106636,A72,A151,female,old,bad
+A11,0.7857142857142857,0.39760096841641906,A74,A153,male,old,good
+A13,0.14285714285714285,0.03791130185979971,A74,A152,male,old,bad
+A12,0.14285714285714285,0.03251898316275999,A74,A151,female,young,good
+A12,0.3571428571428571,0.21255639925167824,A72,A151,female,old,bad
+A13,0.3571428571428571,0.2697259821723341,A74,A152,male,young,good
+A11,0.5178571428571429,0.22180037416088919,A73,A152,female,young,bad
+A12,0.14285714285714285,0.07246616044899307,A73,A152,female,young,good
+A11,0.14285714285714285,0.004897105755474855,A75,A152,male,old,good
+A11,0.19642857142857142,0.05639925167822163,A73,A151,female,young,bad
+A12,0.14285714285714285,0.021404203807637284,A73,A152,male,old,bad
+A14,0.5714285714285714,0.1127434796962694,A75,A152,male,old,good
+A11,0.3571428571428571,0.086552217453505,A71,A152,male,old,bad
+A11,0.7857142857142857,0.41339275888632115,A75,A153,male,old,bad
+A14,0.03571428571428571,0.07400682293386156,A74,A152,male,old,good
+A11,0.14285714285714285,0.007428194123473095,A74,A152,female,old,good
+A14,0.14285714285714285,0.028942445251458126,A73,A152,male,old,good
+A12,0.3571428571428571,0.15511169803015298,A75,A153,male,old,good
+A11,0.08928571428571427,0.06129635743369649,A74,A152,male,old,good
+A14,0.03571428571428571,0.0534279740288324,A73,A152,male,old,good
+A11,0.07142857142857142,0.17321448222735777,A74,A152,male,old,good
+A11,0.24999999999999997,0.21475734565863322,A73,A152,male,old,bad
+A12,0.19642857142857142,0.08407615274568064,A72,A151,female,old,bad
+A14,0.08928571428571427,0.05425332893144051,A72,A151,female,young,good
+A14,0.03571428571428571,0.08341586882359414,A73,A151,male,young,good
+A11,0.4642857142857143,0.43039506988004844,A72,A152,male,young,good
+A12,0.08928571428571427,0.268460437988335,A75,A153,female,old,bad
+A14,0.3571428571428571,0.4620886981402003,A72,A152,male,old,bad
+A11,0.3571428571428571,0.21563772422141522,A73,A152,male,old,good
+A14,0.14285714285714285,0.09271486739297898,A75,A152,male,old,good
+A14,0.3571428571428571,0.11813579839330911,A75,A152,male,old,bad
+A14,0.19642857142857142,0.06652360515021459,A73,A152,female,old,good
+A12,0.14285714285714285,0.023935292175635527,A74,A152,male,young,bad
+A11,0.7857142857142857,0.2396280400572246,A75,A153,male,young,bad
+A14,0.125,0.10410476504897105,A75,A152,male,old,good
+A11,0.19642857142857142,0.03851656212171234,A75,A151,male,old,bad
+A12,0.21428571428571427,0.050896885660834154,A71,A153,male,old,good
+A11,0.14285714285714285,0.06866952789699571,A73,A152,female,young,good
+A14,0.24999999999999997,0.09541102674149884,A73,A153,male,old,good
+A12,0.5714285714285714,0.19555408825795093,A73,A152,female,old,bad
+A11,0.3571428571428571,0.09557609772202047,A73,A151,male,young,bad
+A14,0.3571428571428571,0.21398701441619897,A73,A152,male,old,good
+A14,0.14285714285714285,0.024815670738417523,A73,A152,male,old,good
+A14,0.8928571428571428,0.5054473423572136,A73,A152,male,old,good
+A11,0.24999999999999997,0.16176956091119182,A73,A152,female,young,bad
+A12,0.1607142857142857,0.034774953229888855,A72,A152,male,young,good
+A12,0.3571428571428571,0.13810938703642567,A75,A153,male,old,good
+A14,0.19642857142857142,0.10845163420270716,A73,A151,female,young,good
+A11,0.14285714285714285,0.020688896225376913,A73,A152,female,young,bad
+A12,0.7857142857142857,1.0,A73,A152,female,old,bad
+A11,0.24999999999999997,0.0532078793881369,A73,A152,male,old,bad
+A12,0.053571428571428575,0.12798503356443272,A73,A152,male,old,good
+A11,0.03571428571428571,0.13189171343677783,A73,A151,male,old,good
+A14,0.4107142857142857,0.12765489160338947,A73,A151,female,young,bad
+A13,0.3571428571428571,0.1594585671838891,A73,A152,male,old,good
+A11,0.03571428571428571,0.022889842632331906,A74,A152,female,old,good
+A14,0.14285714285714285,0.051447122262572906,A73,A152,female,old,good
+A14,0.4642857142857143,0.3138549576317817,A74,A152,male,old,good
+A12,0.7321428571428571,0.24738637614174097,A72,A152,male,young,bad
+A14,0.19642857142857142,0.0529877847474414,A75,A152,male,old,good
+A14,0.03571428571428571,0.09067899196654561,A73,A152,male,old,good
+A14,0.7857142857142857,0.6239683063717398,A73,A151,female,young,bad
+A14,0.0,0.01931330472103004,A72,A151,female,young,good
+A12,0.24999999999999997,0.05474854187300539,A73,A152,male,old,bad
+A14,0.5714285714285714,0.30218994167492025,A74,A152,male,old,good
+A12,0.17857142857142858,0.06382744580169472,A75,A152,male,old,good
+A11,0.14285714285714285,0.035765379113018594,A73,A152,male,young,bad
+A14,0.03571428571428571,0.0061626499394739735,A73,A152,female,old,good
+A14,0.19642857142857142,0.09513590844062947,A75,A152,male,old,good
+A14,0.5714285714285714,0.17051832287883792,A73,A152,female,old,bad
+A14,0.14285714285714285,0.030097942115109497,A75,A152,female,old,bad
+A14,0.03571428571428571,0.014746340926598437,A73,A152,female,old,good
+A11,0.3571428571428571,0.1129635743369649,A75,A152,male,old,bad
+A11,0.5714285714285714,0.1354682513480797,A75,A152,male,old,bad
+A14,0.3571428571428571,0.06734896005282272,A72,A152,male,old,good
+A14,0.24999999999999997,0.08638714647298339,A73,A152,male,old,good
+A14,0.3571428571428571,0.09227467811158797,A73,A152,female,old,good
+A11,0.19642857142857142,0.06509298998569385,A73,A151,female,young,good
+A12,0.14285714285714285,0.03895675140310334,A74,A152,male,old,good
+A14,0.5714285714285714,0.19725982172334106,A75,A152,female,old,good
+A14,0.14285714285714285,0.028227137669197756,A73,A152,female,old,good
+A14,0.03571428571428571,0.15010454495433037,A73,A152,male,old,good
+A14,0.14285714285714285,0.05727963024100363,A73,A152,female,old,good
+A11,0.14285714285714285,0.0233300319137229,A74,A152,male,young,bad
+A14,1.0,0.3453835149114119,A73,A153,male,old,good
+A14,0.14285714285714285,0.03994717728623308,A75,A152,male,old,good
+A13,0.10714285714285712,0.20353251898316277,A72,A152,male,old,good
+A12,0.10714285714285712,0.043908880818752064,A73,A152,male,young,good
+A14,0.3571428571428571,0.13315725762077696,A75,A152,male,old,good
+A11,0.4642857142857143,0.23841751953339935,A73,A152,male,old,good
+A14,0.3571428571428571,0.11692527786948388,A73,A152,male,old,good
+A11,0.14285714285714285,0.056784417299438755,A73,A151,female,young,bad
+A14,0.3571428571428571,0.19296797622977882,A75,A153,female,old,good
+A12,0.3571428571428571,0.2195994277539342,A73,A152,male,old,bad
+A11,0.3571428571428571,0.3482447452404534,A71,A153,male,old,good
+A11,0.14285714285714285,0.10294926818531969,A73,A152,male,old,good
+A14,0.14285714285714285,0.10570045119401342,A72,A152,female,old,good
+A14,0.19642857142857142,0.18399911962143722,A72,A152,female,old,good
+A14,0.5714285714285714,0.24408495653130846,A73,A153,male,old,good
+A14,0.19642857142857142,0.2424892703862661,A73,A152,male,old,good
+A11,0.17857142857142858,0.48024650599757895,A75,A152,male,old,bad
+A14,0.24999999999999997,0.04429404643996919,A72,A152,female,old,good
+A11,0.7857142857142857,0.15412127214702323,A73,A152,male,old,bad
+A12,0.4642857142857143,0.17558049961483438,A73,A151,female,young,bad
+A14,0.3571428571428571,0.26961593485198637,A75,A152,male,old,good
+A14,0.14285714285714285,0.09425552987784747,A74,A151,male,old,good
+A12,0.24999999999999997,0.18377902498074172,A71,A152,male,old,good
+A14,0.5714285714285714,0.3076923076923077,A75,A152,male,old,good
+A14,0.3571428571428571,0.06707384175195334,A75,A151,male,old,good
+A14,0.3571428571428571,0.12947067238912735,A73,A151,female,old,good
+A11,0.14285714285714285,0.18713546825134808,A73,A152,male,old,good
+A11,0.3571428571428571,0.15951359084406297,A72,A153,male,young,good
+A11,0.3571428571428571,0.21239132827115662,A75,A151,male,young,bad
+A12,0.08928571428571427,0.06531308462638935,A74,A152,male,old,bad
+A13,0.3571428571428571,0.062011665015956854,A75,A153,female,old,good
+A11,0.08928571428571427,0.06448772972378122,A72,A153,male,old,bad
+A11,0.4642857142857143,0.18553978210630573,A75,A151,female,old,good
+A11,0.3571428571428571,0.17029822823814242,A75,A151,male,old,bad
+A12,0.14285714285714285,0.020633872565203038,A72,A152,male,old,good
+A12,0.14285714285714285,0.040332342907450205,A72,A151,female,young,good
+A12,0.4642857142857143,0.15081985253659075,A75,A152,female,young,good
+A14,0.24999999999999997,0.049466270496313414,A73,A152,male,old,good
+A12,0.08928571428571427,0.001430615164520744,A73,A151,male,young,good
+A11,0.3571428571428571,0.06168152305491362,A73,A151,female,young,bad
+A11,0.14285714285714285,0.029987894794761747,A72,A152,female,old,bad
+A11,0.3571428571428571,0.056784417299438755,A73,A152,female,old,bad
+A12,0.03571428571428571,0.0064927919005172245,A75,A152,male,old,good
+A14,0.3571428571428571,0.055463849455265765,A74,A152,male,young,good
+A14,0.19642857142857142,0.10652580609662154,A74,A151,female,old,good
+A14,0.3571428571428571,0.14223616154946628,A75,A152,male,old,good
+A11,0.3571428571428571,0.18609001870804448,A73,A151,female,young,good
+A14,0.3571428571428571,0.05838010344448112,A74,A152,male,old,good
+A11,0.5714285714285714,0.11543963904478925,A73,A152,male,old,good
+A12,0.03571428571428571,0.010069329811819083,A72,A151,female,young,bad
+A11,0.19642857142857142,0.20408275558490152,A73,A152,female,old,bad
+A12,0.19642857142857142,0.13101133487399583,A73,A152,female,young,good
+A14,0.14285714285714285,0.12077693408165512,A72,A151,male,old,good
+A11,0.10714285714285712,0.04335864421701331,A74,A152,male,old,good
+A14,0.7857142857142857,0.473643666776714,A74,A153,male,old,good
+A13,0.3571428571428571,0.05639925167822163,A73,A152,male,old,good
+A11,0.3571428571428571,0.3476394849785408,A73,A152,male,young,good
+A13,0.3571428571428571,0.06019588423021899,A74,A152,male,old,bad
+A12,0.4107142857142857,0.3144602178936943,A75,A152,male,old,good
+A14,0.14285714285714285,0.020248706943985912,A75,A152,male,old,good
+A14,0.10714285714285712,0.026246285902938263,A75,A153,male,old,good
+A12,0.19642857142857142,0.06569825024760646,A72,A152,male,young,good
+A11,0.5714285714285714,0.18542973478595798,A73,A152,male,old,good
+A11,0.10714285714285712,0.10955210740618465,A72,A151,male,old,good
+A12,1.0,0.799603829646748,A75,A153,female,old,bad
+A14,0.3571428571428571,0.09750192582810609,A73,A152,female,old,good
+A13,0.14285714285714285,0.019753494002421042,A72,A152,female,old,bad
+A11,0.3571428571428571,0.15247056234180698,A73,A151,male,young,good
+A11,0.24999999999999997,0.2272477165181028,A73,A152,male,old,good
+A14,0.10714285714285712,0.05827005612413337,A73,A152,male,old,bad
+A11,0.3571428571428571,0.14185099592824915,A73,A152,male,young,good
+A12,0.24999999999999997,0.08501155496863651,A75,A151,female,old,good
+A14,0.24999999999999997,0.07059535600308132,A72,A152,male,old,bad
+A14,0.3571428571428571,0.26702982282381427,A72,A153,male,old,good
+A11,0.4642857142857143,0.5707604269836029,A75,A153,male,old,good
+A14,0.14285714285714285,0.06250687795752173,A73,A152,female,old,bad
+A14,0.3571428571428571,0.23885770881479035,A73,A152,male,old,bad
+A13,0.14285714285714285,0.17326950588753165,A75,A152,male,old,good
+A13,0.03571428571428571,0.04385385715857819,A73,A152,female,old,good
+A11,0.7857142857142857,0.4091009133927589,A74,A151,female,old,bad
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-val.csv
new file mode 100644
index 0000000..757c07f
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB-val.csv
@@ -0,0 +1,301 @@
+status,month,credit_amount,employment,housing,sex,age,credit
+A11,0.24999999999999997,0.07169582920655881,A73,A152,male,old,bad
+A14,0.19642857142857142,0.06107626279300099,A73,A152,male,old,good
+A12,0.07142857142857142,0.02806206668867613,A74,A152,female,old,good
+A12,0.24999999999999997,0.14575767580059426,A75,A152,male,old,good
+A12,0.7321428571428571,0.15302079894354573,A73,A151,male,young,bad
+A12,0.3571428571428571,0.24551557169582922,A73,A152,male,old,good
+A12,0.24999999999999997,0.090073731704633,A73,A152,male,old,good
+A12,0.5714285714285714,0.11131286453174864,A74,A152,male,old,good
+A12,0.14285714285714285,0.035105095190932106,A75,A152,male,old,bad
+A11,0.14285714285714285,0.07538241443820842,A73,A152,female,old,good
+A14,0.19642857142857142,0.1352481567073842,A72,A152,male,old,good
+A12,0.24999999999999997,0.03389457466710685,A73,A152,male,young,good
+A12,0.07142857142857142,0.06404754044239022,A73,A152,male,old,good
+A14,0.08928571428571427,0.05849015076482887,A75,A152,male,young,good
+A14,0.3571428571428571,0.12809508088478047,A75,A152,male,old,good
+A14,0.19642857142857142,0.1825685044569165,A75,A151,female,old,good
+A14,0.3571428571428571,0.06289204357873886,A73,A152,male,old,good
+A13,0.03571428571428571,0.10267414988445031,A73,A152,male,old,good
+A14,0.14285714285714285,0.09695168922636734,A74,A152,female,old,good
+A14,0.3571428571428571,0.09298998569384835,A75,A152,male,old,good
+A13,0.03571428571428571,0.05904038736656762,A75,A152,male,old,good
+A11,0.5714285714285714,0.09502586112028172,A75,A152,male,old,bad
+A11,0.14285714285714285,0.047210300429184546,A74,A152,male,old,bad
+A14,0.19642857142857142,0.05777484318256851,A75,A153,male,old,good
+A11,0.03571428571428571,0.0505667436997909,A75,A152,male,old,good
+A11,0.5714285714285714,0.16501595686145043,A74,A153,male,old,good
+A14,0.625,0.1275998679432156,A73,A152,male,young,good
+A12,0.08928571428571427,0.07147573456586331,A72,A152,male,old,good
+A14,0.03571428571428571,0.024760647078243648,A75,A153,male,old,good
+A12,0.14285714285714285,0.32067789149334214,A74,A152,male,old,good
+A12,0.08928571428571427,0.05260261912622427,A75,A152,female,young,good
+A12,0.3571428571428571,0.060801144492131615,A72,A152,female,young,bad
+A14,0.3571428571428571,0.15010454495433037,A73,A152,male,old,good
+A14,0.3571428571428571,0.07015516672169032,A74,A152,female,old,good
+A14,0.03571428571428571,0.037526136238582586,A73,A152,female,young,good
+A12,0.4107142857142857,0.12534389787608674,A72,A152,female,old,good
+A11,0.24999999999999997,0.15758776273797734,A72,A151,female,old,bad
+A13,0.4642857142857143,0.09122922856828436,A75,A152,male,old,bad
+A14,0.14285714285714285,0.05447342357213601,A75,A152,female,old,good
+A14,0.14285714285714285,0.02569604930119952,A75,A152,male,old,good
+A12,0.19642857142857142,0.0695499064597777,A73,A152,male,young,good
+A14,0.19642857142857142,0.08688235941454825,A75,A152,male,old,good
+A14,0.14285714285714285,0.02145922746781116,A73,A152,male,old,good
+A12,0.5714285714285714,0.39517992736876856,A73,A151,female,old,good
+A14,0.14285714285714285,0.0710905689446462,A74,A152,male,old,good
+A11,0.5714285714285714,0.5160669087707714,A74,A152,male,young,bad
+A14,0.03571428571428571,0.00968416419060196,A75,A152,male,old,good
+A14,0.19642857142857142,0.24061846594035435,A73,A152,male,old,bad
+A11,0.24999999999999997,0.45234951028942444,A73,A151,female,young,good
+A12,0.5714285714285714,0.1904368878617806,A73,A152,male,old,good
+A14,0.7857142857142857,0.6877407285132606,A74,A152,male,old,good
+A11,0.6785714285714286,0.4199405744470122,A74,A153,male,old,good
+A14,1.0,0.5444040937603168,A74,A152,female,young,good
+A12,0.3571428571428571,0.05480356553317926,A72,A152,male,young,bad
+A14,0.5714285714285714,0.1840541432816111,A75,A152,male,old,good
+A12,0.14285714285714285,0.18526466380543635,A75,A151,male,old,good
+A14,0.14285714285714285,0.034995047870584356,A73,A152,female,young,good
+A11,0.3571428571428571,0.07131066358534169,A74,A151,male,young,bad
+A11,0.14285714285714285,0.10570045119401342,A73,A152,male,old,good
+A14,0.5714285714285714,0.13002090899086607,A73,A152,male,old,good
+A14,0.14285714285714285,0.04682513480796743,A73,A152,male,old,good
+A12,0.24999999999999997,0.1513700891383295,A74,A151,female,old,good
+A14,0.14285714285714285,0.1116430064927919,A73,A153,male,old,good
+A12,0.10714285714285712,0.06993507208099482,A73,A152,male,old,good
+A12,0.24999999999999997,0.04357873885770881,A73,A152,female,old,bad
+A12,0.7857142857142857,0.3538571585781886,A73,A153,male,old,good
+A11,0.5714285714285714,0.10426983602949268,A74,A152,male,young,bad
+A11,0.3571428571428571,0.2512930560140861,A74,A152,male,old,bad
+A12,0.7857142857142857,0.7797402883239792,A73,A152,male,young,bad
+A11,0.14285714285714285,0.02426543413667877,A75,A152,male,old,bad
+A12,0.24999999999999997,0.05441839991196214,A73,A153,male,old,good
+A12,0.3035714285714286,0.1384945526576428,A75,A151,male,old,bad
+A12,0.24999999999999997,0.33069219764498736,A74,A151,male,old,good
+A11,0.24999999999999997,0.027511830086937385,A71,A152,female,old,bad
+A14,0.14285714285714285,0.06894464619786508,A73,A151,male,old,good
+A11,0.14285714285714285,0.024815670738417523,A73,A152,male,old,good
+A14,0.125,0.04974138879718279,A71,A152,female,old,good
+A14,0.3035714285714286,0.16011885110597557,A75,A152,male,old,good
+A11,0.24999999999999997,0.10311433916584131,A73,A151,female,young,bad
+A12,0.14285714285714285,0.02839220864971938,A73,A152,male,old,bad
+A12,0.5714285714285714,0.3685484758446132,A73,A151,male,old,good
+A13,0.08928571428571427,0.0482007263123143,A75,A152,male,old,good
+A11,0.24999999999999997,0.05315285572796302,A72,A151,female,young,bad
+A11,0.14285714285714285,0.10806646858149004,A73,A152,male,young,good
+A14,0.3571428571428571,0.20864971937933313,A74,A152,male,old,good
+A11,0.19642857142857142,0.04418399911962144,A72,A152,male,old,good
+A12,0.3571428571428571,0.14559260482007264,A72,A152,male,old,good
+A11,0.14285714285714285,0.327335754374381,A73,A151,male,old,bad
+A12,0.3571428571428571,0.41311764058545175,A75,A151,female,old,good
+A11,0.14285714285714285,0.06173654671508748,A74,A152,male,old,bad
+A11,0.4642857142857143,0.33564432706063607,A75,A152,male,old,bad
+A12,0.24999999999999997,0.2304941124683614,A75,A152,male,old,good
+A12,0.7857142857142857,0.8485748872014967,A73,A152,male,young,bad
+A14,0.14285714285714285,0.030483107736326624,A75,A152,male,old,good
+A11,0.3035714285714286,0.07439198855507868,A75,A152,male,old,good
+A12,0.14285714285714285,0.3421371189611533,A71,A152,male,old,bad
+A14,0.24999999999999997,0.06729393639264884,A72,A152,male,old,good
+A14,0.3571428571428571,0.2851876306811929,A71,A151,female,old,good
+A14,0.03571428571428571,0.05436337625178826,A71,A152,male,old,good
+A14,0.5178571428571429,0.13832948167712117,A73,A152,female,old,good
+A12,0.14285714285714285,0.09199955981071861,A73,A152,male,old,bad
+A13,0.4642857142857143,0.1522504677011115,A75,A152,male,old,good
+A11,0.3571428571428571,0.052657642786398146,A72,A151,female,young,bad
+A11,0.21428571428571427,0.13068119291295258,A75,A151,male,old,bad
+A11,0.14285714285714285,0.023880268515461652,A73,A151,male,old,bad
+A12,0.08928571428571427,0.07813359744690217,A72,A152,female,young,bad
+A14,0.03571428571428571,0.02321998459337515,A71,A152,female,old,good
+A14,0.3035714285714286,0.12820512820512822,A72,A152,male,old,bad
+A14,0.3571428571428571,0.3336634752943766,A71,A152,male,old,good
+A11,0.3571428571428571,0.1641906019588423,A72,A151,female,young,bad
+A12,0.10714285714285712,0.3883569935072081,A71,A153,male,old,good
+A12,0.1607142857142857,0.10184879498184218,A72,A152,female,young,good
+A11,0.5714285714285714,0.32898646417959726,A72,A151,female,young,bad
+A13,0.14285714285714285,0.15219544404093763,A73,A152,male,young,good
+A14,0.14285714285714285,0.014966435567293938,A75,A152,male,old,good
+A14,0.14285714285714285,0.0926598437328051,A75,A152,male,old,good
+A14,0.7857142857142857,0.1376141740948608,A75,A152,male,old,good
+A12,0.19642857142857142,0.06943985913942995,A73,A152,male,old,bad
+A12,0.4642857142857143,0.12396830637173985,A75,A152,male,old,good
+A14,0.3571428571428571,0.05705953560030813,A75,A152,female,old,good
+A12,0.5714285714285714,0.6679322108506658,A73,A153,male,old,bad
+A13,0.08928571428571427,0.05981071860900186,A72,A152,male,old,bad
+A11,0.4107142857142857,0.12061186310113349,A75,A152,male,old,good
+A14,0.10714285714285712,0.06129635743369649,A73,A152,female,old,good
+A11,0.3571428571428571,0.11742049081104876,A75,A151,male,old,good
+A11,0.08928571428571427,0.1402553097832068,A73,A151,male,old,good
+A13,0.24999999999999997,0.06575327390778034,A74,A152,male,old,good
+A14,0.03571428571428571,0.08215032463959503,A73,A152,male,old,good
+A13,0.10714285714285712,0.06036095521074061,A74,A152,male,old,good
+A11,0.3571428571428571,0.25420931000330144,A73,A153,male,old,bad
+A14,0.14285714285714285,0.0817651590183779,A74,A152,female,old,good
+A12,0.14285714285714285,0.027731924727632886,A75,A152,male,old,good
+A14,0.5714285714285714,0.18075272367117862,A74,A152,male,old,good
+A14,0.19642857142857142,0.0675690546935182,A73,A152,male,old,good
+A14,0.4642857142857143,0.35721360184879497,A74,A152,male,old,good
+A14,0.5714285714285714,0.37839771101573677,A72,A151,female,young,bad
+A11,0.19642857142857142,0.030593155056674374,A73,A152,female,young,good
+A13,0.3571428571428571,0.14537251017937713,A75,A153,male,old,good
+A12,0.5714285714285714,0.30538131396500495,A73,A152,male,old,good
+A14,0.3571428571428571,0.15709254979641246,A72,A152,male,young,good
+A14,0.19642857142857142,0.06250687795752173,A73,A151,male,old,good
+A14,0.24999999999999997,0.11852096401452623,A73,A152,female,old,good
+A12,0.3571428571428571,0.19962583911081766,A72,A152,male,old,good
+A12,0.3571428571428571,0.15637724221415208,A72,A151,male,young,bad
+A14,0.4285714285714286,0.1371739848134698,A75,A152,male,old,good
+A14,0.14285714285714285,0.07285132607021019,A73,A152,male,old,good
+A11,0.3035714285714286,0.0872125013755915,A73,A152,female,young,bad
+A12,0.3571428571428571,0.13579839330912294,A73,A151,female,young,bad
+A14,0.7857142857142857,0.2527786948387807,A71,A151,male,old,bad
+A11,0.3571428571428571,0.052217453505007144,A75,A152,male,old,bad
+A11,0.17857142857142858,0.20485308682733577,A71,A153,male,young,good
+A14,0.3571428571428571,0.1336524705623418,A72,A152,female,old,good
+A14,0.14285714285714285,0.10360955210740619,A75,A153,female,old,good
+A11,0.3571428571428571,0.05397821063057115,A75,A151,female,old,good
+A14,0.3571428571428571,0.14427203697589966,A75,A152,male,old,good
+A12,0.24999999999999997,0.11978650819852536,A75,A152,male,old,good
+A11,1.0,0.3877517332452955,A75,A151,male,old,bad
+A14,0.4642857142857143,0.40420380763728403,A75,A152,male,old,good
+A11,0.625,0.7664245625619016,A74,A152,male,old,good
+A12,0.08928571428571427,0.06883459887751733,A75,A152,female,old,bad
+A12,0.7857142857142857,0.19775503466490593,A74,A153,male,old,bad
+A14,0.6785714285714286,0.20864971937933313,A73,A152,male,old,good
+A11,0.03571428571428571,0.06184659403543523,A71,A152,female,old,good
+A14,0.3571428571428571,0.41619896555518876,A74,A152,male,old,good
+A14,0.5714285714285714,0.03626059205458347,A75,A152,male,old,good
+A14,0.24999999999999997,0.34158688235941453,A75,A152,male,old,bad
+A14,0.3571428571428571,0.11962143721800374,A75,A152,male,old,good
+A14,0.19642857142857142,0.13965004952129417,A74,A152,female,young,good
+A11,0.14285714285714285,0.024595576097722022,A72,A152,male,old,bad
+A14,0.3035714285714286,0.12248266754704522,A75,A152,male,old,good
+A12,0.14285714285714285,0.0706503796632552,A72,A151,male,young,bad
+A12,0.03571428571428571,0.04500935402222955,A75,A152,male,old,good
+A12,0.14285714285714285,0.18284362275778587,A73,A152,female,young,good
+A12,0.24999999999999997,0.05782986684274238,A75,A152,male,old,good
+A14,0.6785714285714286,0.3805436337625179,A74,A151,male,old,good
+A12,0.3571428571428571,0.09843732805106195,A72,A152,male,young,bad
+A11,0.3571428571428571,0.12286783316826234,A73,A152,male,young,good
+A13,0.3571428571428571,0.18526466380543635,A75,A151,male,young,good
+A14,0.3571428571428571,0.09948277759436558,A73,A152,male,old,good
+A11,0.3571428571428571,0.08281060856168151,A75,A152,female,old,good
+A14,0.3571428571428571,0.07202597116760207,A74,A152,male,old,good
+A11,0.3571428571428571,0.16017387476614944,A73,A151,male,old,bad
+A14,0.7857142857142857,0.5486959392538792,A74,A152,male,old,good
+A11,0.3571428571428571,0.062231759656652355,A73,A152,female,old,bad
+A14,0.3035714285714286,0.10955210740618465,A75,A152,male,old,good
+A14,0.3571428571428571,0.12589413447782546,A75,A152,male,old,bad
+A11,0.5714285714285714,0.4300099042588313,A73,A152,female,young,bad
+A12,0.7857142857142857,0.33927588863211183,A75,A151,female,old,bad
+A12,0.3571428571428571,0.24683613954000222,A72,A152,female,young,bad
+A14,0.14285714285714285,0.037416088918234836,A75,A152,male,old,good
+A11,0.3571428571428571,0.1463629360625069,A74,A153,male,old,good
+A12,0.24999999999999997,0.11158798283261802,A72,A152,female,old,bad
+A14,0.5714285714285714,0.2318146803125344,A73,A152,male,old,bad
+A14,0.5714285714285714,0.11797072741278751,A73,A152,female,young,good
+A14,0.3571428571428571,0.1103224386486189,A72,A152,male,old,good
+A14,0.625,0.4587872785297678,A75,A152,male,old,good
+A12,0.3571428571428571,0.3022449653350941,A72,A153,female,young,good
+A11,0.14285714285714285,0.05227247716518102,A73,A151,female,young,good
+A14,0.4642857142857143,0.08897325850115549,A75,A152,male,old,good
+A11,0.03571428571428571,0.0335644327060636,A75,A152,female,old,good
+A11,0.3571428571428571,0.04258831297457907,A72,A152,male,old,bad
+A11,1.0,0.36238582590513924,A75,A152,male,old,bad
+A11,0.14285714285714285,0.07868383404864092,A75,A152,male,old,good
+A14,0.3571428571428571,0.06399251678221635,A73,A152,male,old,good
+A12,0.07142857142857142,0.03615054473423572,A72,A152,male,old,good
+A11,0.3571428571428571,0.05188731154396391,A71,A151,female,old,bad
+A11,0.14285714285714285,0.1280400572246066,A73,A152,male,old,good
+A14,0.3571428571428571,0.2047980631671619,A74,A151,female,young,good
+A12,0.3571428571428571,0.1930229998899527,A71,A151,female,young,good
+A14,0.24999999999999997,0.11389897655992078,A71,A152,male,old,good
+A14,0.3571428571428571,0.12176735996478487,A74,A152,male,old,good
+A13,0.03571428571428571,0.06014086057004511,A75,A152,male,old,good
+A14,0.4285714285714286,0.4167492021569275,A72,A151,male,old,good
+A14,0.3035714285714286,0.16644657202597118,A75,A152,male,old,good
+A12,0.24999999999999997,0.16474083856058105,A73,A152,female,old,good
+A14,0.4107142857142857,0.46016287003411466,A73,A152,male,old,good
+A11,0.5714285714285714,0.08759766699680863,A72,A152,female,old,bad
+A14,0.3571428571428571,0.12991086167051832,A75,A152,male,old,good
+A12,0.08928571428571427,0.09794211510949709,A74,A152,male,young,good
+A14,0.3571428571428571,0.49669858038956755,A73,A153,male,old,good
+A14,0.14285714285714285,0.045449543303620554,A73,A152,male,old,good
+A14,0.10714285714285712,0.07131066358534169,A73,A152,male,old,good
+A11,0.14285714285714285,0.09040387366567623,A73,A152,female,old,good
+A14,0.14285714285714285,0.10404974138879718,A74,A152,male,old,good
+A14,0.5714285714285714,0.425332893144052,A72,A151,male,old,bad
+A14,0.14285714285714285,0.055793991416309016,A75,A151,male,old,good
+A11,0.24999999999999997,0.10999229668757565,A74,A152,male,old,good
+A11,0.7857142857142857,0.324254429404644,A75,A153,female,old,bad
+A11,0.3035714285714286,0.1842742379223066,A74,A151,female,old,good
+A12,0.19642857142857142,0.07081545064377683,A75,A152,male,old,good
+A14,0.5714285714285714,0.590073731704633,A71,A152,female,old,bad
+A12,0.24999999999999997,0.2226807527236712,A75,A152,male,old,bad
+A11,0.24999999999999997,0.05172224056344228,A71,A153,female,old,bad
+A11,0.14285714285714285,0.056344228018047754,A72,A152,female,old,bad
+A14,0.08928571428571427,0.11863101133487398,A71,A152,male,old,good
+A12,0.3571428571428571,0.33856058104985143,A72,A152,male,old,good
+A14,0.10714285714285712,0.05705953560030813,A75,A152,male,old,good
+A12,0.6785714285714286,0.3138549576317817,A74,A152,female,old,good
+A11,0.03571428571428571,0.09392538791680423,A74,A152,female,old,good
+A14,0.3571428571428571,0.07164080554638494,A74,A152,male,old,good
+A14,0.3571428571428571,0.03780125453945196,A72,A152,male,old,good
+A14,0.3571428571428571,0.09805216242984482,A75,A153,male,old,good
+A14,0.5714285714285714,0.5846263893474194,A75,A152,male,old,good
+A12,0.3571428571428571,0.2659843732805106,A75,A152,female,old,good
+A11,0.5714285714285714,0.2712116209970287,A74,A152,male,old,bad
+A12,0.03571428571428571,0.044018928139099814,A71,A152,male,old,good
+A13,0.5714285714285714,0.2015516672169033,A73,A152,male,young,good
+A13,0.24999999999999997,0.15401122482667548,A72,A152,female,old,good
+A11,0.5714285714285714,0.44486629250577747,A75,A153,male,old,bad
+A11,0.14285714285714285,0.08847804555959062,A72,A151,female,young,good
+A11,0.14285714285714285,0.25272367117860683,A75,A151,male,old,bad
+A14,0.24999999999999997,0.19423352041377795,A72,A152,male,old,good
+A12,0.14285714285714285,0.09783206778914934,A73,A152,male,old,good
+A14,0.3571428571428571,0.11191812479366127,A74,A152,male,old,good
+A12,0.19642857142857142,0.1363486299108617,A74,A152,male,old,good
+A11,0.4107142857142857,0.27748431825685044,A71,A152,male,old,bad
+A12,1.0,0.4900957411687025,A73,A153,male,old,good
+A14,0.24999999999999997,0.09315505667436998,A72,A152,female,young,bad
+A11,0.3571428571428571,0.148949048090679,A75,A152,male,old,good
+A11,0.24999999999999997,0.11769560911191813,A72,A152,female,old,good
+A11,0.24999999999999997,0.08979861340376362,A73,A151,female,young,bad
+A12,0.4642857142857143,0.08060966215472654,A73,A152,female,old,good
+A11,0.7857142857142857,0.33459887751733247,A75,A153,male,old,bad
+A14,0.19642857142857142,0.0586552217453505,A73,A152,male,old,good
+A11,0.03571428571428571,0.8058765269065697,A75,A152,male,old,bad
+A14,0.4642857142857143,0.15555188731154398,A75,A152,male,old,good
+A14,0.0,0.07120061626499395,A74,A152,male,old,good
+A12,0.14285714285714285,0.041652910751623196,A73,A152,male,young,good
+A14,0.10714285714285712,0.10008803785627819,A73,A152,male,old,good
+A12,0.24999999999999997,0.7002310993727302,A71,A153,female,old,bad
+A11,0.24999999999999997,0.1570375261362386,A74,A152,male,old,good
+A12,0.5714285714285714,0.20578848905029165,A72,A152,female,old,good
+A11,0.14285714285714285,0.17877187190491914,A73,A152,female,old,bad
+A12,0.08928571428571427,0.048750962914053037,A75,A153,male,old,bad
+A13,0.19642857142857142,0.007813359744690218,A72,A151,female,young,good
+A11,0.24999999999999997,0.048475844613183675,A71,A152,female,old,bad
+A12,0.14285714285714285,0.06459777704412897,A74,A152,male,old,good
+A11,0.6785714285714286,0.17299438758666227,A71,A152,male,old,good
+A11,0.7857142857142857,0.3713546825134808,A74,A152,male,old,bad
+A11,0.4642857142857143,0.6464179597226807,A72,A152,male,old,bad
+A14,0.4642857142857143,0.14201606690877078,A73,A152,female,old,good
+A12,0.3571428571428571,0.09436557719819522,A73,A151,female,old,good
+A14,0.3571428571428571,0.07411687025420931,A75,A153,male,old,good
+A12,0.03571428571428571,0.09975789589523495,A72,A151,male,old,good
+A12,0.3928571428571428,0.42456256190161773,A72,A152,male,old,good
+A12,0.7857142857142857,0.4034334763948498,A71,A153,male,old,good
+A14,0.3571428571428571,0.24369979091009136,A74,A152,male,old,good
+A14,0.14285714285714285,0.30543633762517886,A75,A151,male,old,good
+A14,0.3571428571428571,0.1990756025090789,A75,A151,female,old,good
+A14,0.14285714285714285,0.11775063277209201,A75,A152,male,old,good
+A12,0.19642857142857142,0.3631561571475735,A71,A152,male,old,bad
+A14,0.08928571428571427,0.0675690546935182,A74,A152,male,young,bad
+A14,0.03571428571428571,0.10069329811819082,A73,A152,male,young,good
+A12,0.03571428571428571,0.03747111257840871,A72,A152,female,old,bad
+A14,0.08928571428571427,0.05469351821283151,A75,A151,female,old,good
+A11,0.10714285714285712,0.10355452844723231,A72,A151,female,old,good
+A14,0.24999999999999997,0.0935402222955871,A74,A152,male,old,good
+A11,0.03571428571428571,0.06184659403543523,A73,A152,male,old,good
+A12,0.03571428571428571,0.06597336854847584,A75,A152,male,old,good
+A14,0.24999999999999997,0.07329151535160118,A75,A152,male,old,good
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-test.csv
new file mode 100644
index 0000000..4fbf0ba
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-test.csv
@@ -0,0 +1,201 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.5714285714285714,0.30802244965335096,male,young,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11290855067679102,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.04693518212831517,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.0704302850225597,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08093980411576977,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.1389347419390338,male,old,bad,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.08974358974358974,female,old,good,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.15291075162319798,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.07741828986464179,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.11571475734565863,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.13255199735886433,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.07384175195333993,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09216463079124024,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.594475624518543,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09447562451854297,female,young,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.13321228128095083,female,old,bad,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.17497523935292178,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.23236491691427316,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.24347969626939586,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.041212721470232194,male,young,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.5893584241223726,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.05876526906569825,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.019753494002421042,female,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.1509298998569385,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+1.0,0.847529437658193,male,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.17156377242214152,male,young,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.7857142857142857,0.24969736986904373,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.10982722570705403,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.08237041928029053,male,young,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.2666446572025971,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.034885000550236606,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.32023770221195114,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06228678331682623,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.02767690106745901,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.13464289644547156,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.01870804445911742,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.4642857142857143,0.09183448883019699,male,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.3035714285714286,0.07686805326290305,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.3090128755364807,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3132496973698691,female,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.026246285902938263,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.14097061736546715,female,old,good,1,0,0,0,0,0,0,0,1,1,0,0
+0.7857142857142857,0.22328601298558381,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.3266754704522945,male,young,good,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.05749972488169913,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.3004842082095301,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.32243864861890614,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.44767249917464513,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09414548255749973,female,young,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.04996148343787829,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06382744580169472,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.5178571428571429,0.385330692197645,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.023825244855287777,female,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.06399251678221635,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.2817761637504127,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.2217453505007153,female,old,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06096621547265324,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.0292175635523275,male,old,good,0,0,1,0,0,0,0,0,1,0,0,1
+0.0,0.17222405634422802,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.08928571428571427,0.045779685264663805,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.11483437878287663,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.053571428571428575,0.03279410146362936,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.7857142857142857,0.45322988885220644,female,young,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.09106415758776273,male,old,good,0,0,1,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.08550676791020138,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.022119511389897654,female,young,good,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.059480576647958625,male,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.15566193463189174,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.1340926598437328,male,old,good,0,0,1,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.05942555298778475,male,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05051172003961703,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.17585561791570375,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.052217453505007144,female,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.02701661714537251,female,young,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.39198855507868385,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.10559040387366568,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.05986574226917574,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.33944095961263343,female,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.039892153626059204,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13392758886321118,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.23528117090348852,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.038571585781886214,female,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.8928571428571428,0.8635963464289644,male,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.18190822053483,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.0,0.06630351050951909,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09271486739297898,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.1581379993397161,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.07114559260482008,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14933421371189612,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.24999999999999997,0.06558820292725871,male,old,bad,1,0,0,0,0,0,0,1,0,0,0,1
+1.0,0.7431495543083525,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.2857142857142857,0.3245295477055134,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3336084516342027,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.41900517222405637,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05760977220204688,male,young,good,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.121712336304611,male,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.3582590513920986,male,old,bad,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.5129305601408606,male,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.39391438318476946,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.36519203257400684,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18548475844613185,male,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.06063607351160999,female,young,good,1,0,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.06767910201386595,male,old,good,0,0,1,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.042808407615274574,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.035875426433366345,male,young,bad,1,0,0,0,0,0,0,1,0,1,0,0
+0.10714285714285712,0.1136238582590514,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.09519093210080334,female,young,bad,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07169582920655881,female,young,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.10714285714285712,0.05447342357213601,female,old,bad,0,0,1,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05353802134918015,male,old,bad,0,1,0,0,0,0,0,0,1,1,0,0
+0.3035714285714286,0.12963574336964895,female,old,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.06377242214152085,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.07252118410916694,female,young,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.01785714285714285,0.1759656652360515,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.1992956971497744,female,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.116430064927919,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.4087157477715418,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.5033564432706064,male,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.5714285714285714,0.5126004181798173,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.0928799383735006,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.10448993067018818,male,old,bad,1,0,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.350170573346539,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.14355672939363928,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.2857142857142857,0.13090128755364808,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.32761087267525035,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.18851105975569496,male,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.114229118520964,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.17244415098492352,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.1546164850885881,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05771981952239463,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06426763508308572,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.04104765048971058,female,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.2857142857142857,0.1780015406624849,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.17211400902388027,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.05359304500935402,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.02723671178606801,female,old,bad,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08858809287993837,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.2857142857142857,0.2213051612193243,female,young,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.053571428571428575,0.11912622427643886,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19406844943325632,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.10625068779575217,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.239022779795312,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.028007043028502255,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.23550126554418402,female,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.10714285714285712,0.09210960713106636,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.17591064157587763,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+1.0,0.7580609662154726,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.14383184769450866,male,old,bad,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05529877847474414,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.6785714285714286,0.22669747991636405,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.03719599427753935,female,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.030373060415978873,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.04632992186640256,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.07378672829316606,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.23131946737096953,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.023770221195113902,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.3362495873225487,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.055463849455265765,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.4107142857142857,0.12490370859469573,male,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.00484208209530098,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.2857142857142857,0.37454605480356556,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.2857142857142857,0.10922196544514141,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06459777704412897,female,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.6095521074061846,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05496863651370089,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.08928571428571427,0.10278419720479806,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05694948827996038,female,old,bad,1,0,0,0,0,0,0,1,0,1,0,0
+0.7857142857142857,0.18311874105865522,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.14295146913172666,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.17932210850665786,female,young,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.07142857142857142,0.05029162539892153,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.2050181578078574,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.18498954550456698,female,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+1.2142857142857142,0.2941014636293606,male,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.025255860019808517,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.43903378452734676,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.17409486079013978,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.6802024870694399,male,old,bad,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.08253549026081215,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.15555188731154398,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.2791900517222406,female,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.0,0.06844943325630021,male,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.12809508088478047,female,old,good,1,0,0,0,1,0,0,0,0,0,0,1
+0.053571428571428575,0.11439418950148562,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.022229558710245404,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.4337515131506548,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19880048420820953,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.39738087377572356,female,young,good,0,1,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.07180587652690656,male,old,bad,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.055683944095961266,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.02971277649389237,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.27423792230659183,male,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.11863101133487398,female,young,bad,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.1496093320127655,female,young,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.3035714285714286,0.6839440959612633,male,old,bad,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.045779685264663805,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.20166171453725104,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.19709475074281943,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09232970177176185,male,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.5286123032904149,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-train.csv
new file mode 100644
index 0000000..7fa0cc9
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-train.csv
@@ -0,0 +1,501 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.5714285714285714,0.20408275558490152,male,old,good,1,0,0,0,1,0,0,0,0,0,1,0
+0.08928571428571427,0.18306371739848135,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.12462859029382635,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09601628700341147,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+1.0,0.5566193463189171,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.39303400462198745,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.008748761967646089,female,old,good,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.12418840101243535,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.4642857142857143,0.25096291405304283,female,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.7857142857142857,0.3471992956971498,male,young,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.48332783096731596,male,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1816881258941345,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.2857142857142857,0.3421371189611533,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.1250137559150435,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.14141080664685815,male,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.029327610872675252,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.19329811819082207,male,young,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.39809618135798397,female,old,bad,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.06344228018047761,female,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.125,0.24876196764608782,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.20743919885550788,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.052327500825354895,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.28232640035215145,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.28947947617475517,male,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.04666006382744581,female,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.7857142857142857,0.32871134587872786,male,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.125,0.06289204357873886,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.08928571428571427,0.09183448883019699,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.19642857142857142,0.2638384505337295,female,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.3136898866512601,female,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.07142857142857142,0.026466380543633764,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.02085396720589854,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.08572686255089688,female,young,bad,0,0,0,1,0,0,0,1,0,0,1,0
+0.3035714285714286,0.05161219324309453,female,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.16424562561901618,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.6223175965665235,female,young,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.04974138879718279,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.19808517662594918,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.11604489930670188,male,old,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.07059535600308132,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.21789369428854408,male,old,bad,0,0,1,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.19847034224716628,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06289204357873886,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.08528667326950588,male,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1101023440079234,female,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03224386486189061,female,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.06025090789039286,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.2500825354902608,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.025200836359634642,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05716958292065588,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08880818752063387,female,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.5342797402883239,female,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.27797953119841534,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.03571428571428571,0.08924837680202487,male,old,good,1,0,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.12044679212061186,female,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.06261692527786948,female,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.6558270056124132,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.13271706833938596,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.1607142857142857,0.06377242214152085,female,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.1866952789699571,female,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.358093980411577,female,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.0880928799383735,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.4107142857142857,0.17420490811048753,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.2678001540662485,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.10714285714285712,0.07411687025420931,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.06228678331682623,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.05397821063057115,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.13101133487399583,female,old,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.10454495433036205,female,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1474634092659844,female,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.3582590513920986,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3035714285714286,0.13728403213381754,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.11417409486079015,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05249257180587652,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.04396390447892594,female,young,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11235831407505227,male,old,bad,0,0,0,1,1,0,0,0,0,0,1,0
+0.08928571428571427,0.06597336854847584,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.39952679652250467,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.03675580499614835,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1153295917244415,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.07444701221525256,female,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.125,0.2011114779355123,male,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0482007263123143,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.3035714285714286,0.14707824364476726,female,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1620446792120612,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.24573566633652472,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.7871134587872785,male,young,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.2884890502916254,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.0734565863321228,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.18377902498074172,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1380543633762518,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.10300429184549356,male,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.09640145262462858,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.09084406294706723,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06839440959612633,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.03571428571428571,0.052767690106745896,male,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.20518322878837902,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.4110817651590184,female,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.1870804445911742,female,young,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.08968856608341587,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09304500935402223,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.034334763948497854,female,young,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.4107142857142857,0.2718168812589414,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.5627269725982172,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.22565203037306042,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06195664135578298,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.4642857142857143,0.1786068009243975,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.6428571428571428,0.3162759986794322,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.4285714285714286,0.20666886761307363,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.1701331572576208,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.13051612193243095,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11742049081104876,male,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07752833718498954,female,old,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1781666116430065,female,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+1.0,0.39440959612633436,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.10938703642566304,female,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.059590623968306375,male,old,bad,1,0,0,0,1,0,0,0,0,0,0,1
+0.3214285714285714,0.05683944095961263,female,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.17981732144822274,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.018432926158248045,male,young,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.625,0.25767580059425554,male,young,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.056564322658743255,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.08352591614394189,male,young,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.17051832287883792,male,old,bad,1,0,0,0,0,1,0,0,0,0,0,1
+0.08928571428571427,0.0627819962583911,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.17453505007153078,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.10432485968966655,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.06988004842082095,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3627159678661825,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.09909761197314845,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15109497083746012,male,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18763068119291296,female,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06558820292725871,female,young,bad,1,0,0,0,0,0,0,1,0,1,0,0
+0.08928571428571427,0.04886101023440079,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.3037856278199626,female,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.12930560140860572,male,old,bad,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19252778694838782,female,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.15456146142841423,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.022669747991636405,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.09298998569384835,male,old,good,1,0,0,0,0,1,0,0,0,0,0,1
+0.5714285714285714,0.1140640475404424,male,young,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.7857142857142857,0.6577528337184989,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.28925938153405967,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.24999999999999997,0.010069329811819083,female,young,bad,0,0,0,1,1,0,0,0,0,1,0,0
+0.03571428571428571,0.23995818201826785,female,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.13821943435677342,female,young,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.4642857142857143,0.2213051612193243,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.625,0.6333223286012986,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.5714285714285714,0.10867172884340266,male,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.06734896005282272,female,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.06338725652030373,female,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.24999999999999997,0.09150434686915374,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.17857142857142858,0.030373060415978873,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.05617915703752613,male,old,bad,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.055463849455265765,female,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.18240343347639487,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.17255419830527127,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.1098822493672279,female,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.30560140860570045,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.03906679872345109,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.044349070100143065,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3322878837900297,male,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.13772422141520854,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.10784637394079453,male,young,bad,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.5528227137669197,male,old,bad,1,0,0,0,0,0,0,1,0,0,0,1
+0.5714285714285714,0.7740728513260702,male,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.4642857142857143,0.2200396170353252,male,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.03736106525806096,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.08928571428571427,0.039011775063277215,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.13931990756025092,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.7857142857142857,0.44613183668977663,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.11285352701661715,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.7857142857142857,0.18245845713656875,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.20446792120611865,female,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.03571428571428571,0.05430835259161439,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.03994717728623308,female,young,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.03659073401562672,female,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.03835149114119071,male,old,bad,0,0,1,0,0,0,0,1,0,0,0,1
+0.3571428571428571,0.14168592494772753,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.11213821943435677,female,young,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.10714285714285712,0.145867723120942,female,old,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.7321428571428571,0.23803235391218225,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.3845053372950369,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.13733905579399142,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.08215032463959503,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.06140640475404424,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.06905469351821283,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.005117200396170352,female,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.10856168152305491,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.06850445691647408,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.11428414218113787,female,young,bad,0,0,1,0,0,1,0,0,0,0,1,0
+0.7321428571428571,0.6364036535710355,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.06789919665456146,male,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.4844833278309673,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.07730824254429404,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.06619346318917134,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.053813139650049524,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.037526136238582586,female,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.4311654011224827,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.06030593155056674,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.5714285714285714,0.1437218003741609,male,old,good,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.15808297567954221,female,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+1.0,0.3938593595245956,female,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.055793991416309016,male,young,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.07175085286673269,female,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.3035714285714286,0.18719049191152196,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.20254209310003302,male,old,bad,1,0,0,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.061131286453174866,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.052437548145702645,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18031253438978762,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.4184549356223176,female,young,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.17095851216022892,female,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.10179377132166832,male,old,bad,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0505667436997909,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.47854077253218885,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09276989105315285,male,young,good,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.1794871794871795,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.3695939253879168,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.4642857142857143,0.11461428414218112,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09843732805106195,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.49636843842852424,male,young,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.07274127874986244,female,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.011224826675470454,male,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.33311323869263787,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.10713106635853417,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.7857142857142857,0.40601958842302194,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.05639925167822163,female,young,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.37795752173434577,male,old,bad,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.6783867062837019,female,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.24999999999999997,0.1391548365797293,male,young,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.0,0.06894464619786508,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.07142857142857142,0.05430835259161439,female,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.20160669087707717,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15472653240893586,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.5714285714285714,0.09370529327610873,male,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.4642857142857143,0.19709475074281943,male,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.0708704743039507,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.03571428571428571,0.10223396060305931,male,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.00979421151094971,female,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.2313744910311434,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.1138439528997469,male,old,bad,0,0,1,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.05023660173874765,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.037746230879278087,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.12231759656652359,male,young,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.19642857142857142,0.07257620776934082,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.1522504677011115,female,old,good,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11450423682183337,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.10157367668097282,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.04330362055683944,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05425332893144051,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.21464729833828547,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.18482447452404535,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06393749312204247,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.125,0.20298228238142402,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.0702101903818642,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.3214285714285714,0.1334323759216463,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.4642857142857143,0.2184989545504567,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14713326730494114,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08622207549246176,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.0536480686695279,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08776273797733025,female,old,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.0,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022559700671288655,male,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.025255860019808517,male,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.8587542643336634,male,old,good,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.08732254869593925,female,old,bad,0,1,0,0,0,0,0,1,0,0,0,1
+0.1607142857142857,0.08512160228898426,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.010894684714427203,female,young,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.16303510509519095,male,young,good,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.12815010454495435,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.05216242984483327,female,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.26152745680642675,female,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11461428414218112,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.055683944095961266,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.049796412457356665,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.03543523715197534,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.16952789699570817,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.04748541873005392,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.02580609662154727,male,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.09981291955540883,female,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.18741058655221746,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3502255970067129,male,old,good,1,0,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.16391548365797293,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.17018818091779467,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.1572576207769341,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.011444921316165951,male,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06718388907230109,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.125,0.05898536370639375,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.057114559260482006,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.025145812699460767,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.058215032463959496,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.15538681633102236,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.12440849565313085,female,young,good,1,0,0,0,1,0,0,0,0,1,0,0
+0.03571428571428571,0.0819852536590734,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.04803565533179267,female,old,bad,0,0,0,1,0,0,1,0,0,1,0,0
+0.125,0.07301639705073182,female,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.7321428571428571,0.08776273797733025,male,young,bad,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09783206778914934,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.4107142857142857,0.4439308902828216,female,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05749972488169913,female,young,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.6431165401122483,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.08633212281280951,male,old,bad,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.0675690546935182,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06965995378012545,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1037746230879278,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.19599427753934193,female,young,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.21541762958071972,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.20947507428194126,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.543468691537361,male,old,bad,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.012875536480686695,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+1.0,0.33223286012985587,male,old,bad,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.36436667767139874,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.07054033234290745,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.10366457576758006,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.04715527676901067,male,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.2857142857142857,0.10795642126114229,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.16154946627049632,female,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.7678571428571428,0.5762077693408165,female,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.053571428571428575,0.02641135688345989,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.10030813249697369,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.0176625949158138,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.19642857142857142,0.11609992296687575,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06019588423021899,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.6785714285714286,0.3809838230439089,female,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.6785714285714286,0.2501375591504347,male,old,good,0,0,1,0,0,0,0,0,1,0,0,1
+0.6785714285714286,0.20441289754594477,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.125,0.38395510069329813,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05315285572796302,male,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.13744910311433917,female,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.6785714285714286,0.49702872235061074,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.4107142857142857,0.23528117090348852,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.3035714285714286,0.2750082535490261,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.16127434796962695,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.02569604930119952,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.055683944095961266,male,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08990866072411137,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09210960713106636,female,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.39760096841641906,male,old,good,1,0,0,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.03791130185979971,male,old,bad,0,0,1,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.03251898316275999,female,young,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.21255639925167824,female,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.2697259821723341,male,young,good,0,0,1,0,0,0,0,1,0,0,1,0
+0.5178571428571429,0.22180037416088919,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.07246616044899307,female,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.004897105755474855,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.19642857142857142,0.05639925167822163,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.021404203807637284,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.1127434796962694,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.086552217453505,male,old,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.41339275888632115,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.03571428571428571,0.07400682293386156,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.007428194123473095,female,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.028942445251458126,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15511169803015298,male,old,good,0,1,0,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.06129635743369649,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.0534279740288324,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.17321448222735777,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.21475734565863322,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08407615274568064,female,old,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.05425332893144051,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.03571428571428571,0.08341586882359414,male,young,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.4642857142857143,0.43039506988004844,male,young,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.268460437988335,female,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.4620886981402003,male,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.21563772422141522,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09271486739297898,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.11813579839330911,male,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.06652360515021459,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.023935292175635527,male,young,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.2396280400572246,male,young,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.125,0.10410476504897105,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.03851656212171234,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.21428571428571427,0.050896885660834154,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.06866952789699571,female,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.09541102674149884,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.19555408825795093,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09557609772202047,male,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.21398701441619897,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.024815670738417523,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.8928571428571428,0.5054473423572136,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.16176956091119182,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.1607142857142857,0.034774953229888855,male,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.13810938703642567,male,old,good,0,1,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.10845163420270716,female,young,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.020688896225376913,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,1.0,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0532078793881369,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.053571428571428575,0.12798503356443272,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.13189171343677783,male,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.4107142857142857,0.12765489160338947,female,young,bad,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.1594585671838891,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022889842632331906,female,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.051447122262572906,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.3138549576317817,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.7321428571428571,0.24738637614174097,male,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.0529877847474414,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.09067899196654561,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.6239683063717398,female,young,bad,0,0,0,1,0,0,1,0,0,1,0,0
+0.0,0.01931330472103004,female,young,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.24999999999999997,0.05474854187300539,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.30218994167492025,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.17857142857142858,0.06382744580169472,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.035765379113018594,male,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.0061626499394739735,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.09513590844062947,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.17051832287883792,female,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.030097942115109497,female,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.014746340926598437,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1129635743369649,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.1354682513480797,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06734896005282272,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.08638714647298339,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09227467811158797,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06509298998569385,female,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.03895675140310334,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.19725982172334106,female,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.028227137669197756,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.15010454495433037,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05727963024100363,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.0233300319137229,male,young,bad,1,0,0,0,0,0,0,1,0,0,1,0
+1.0,0.3453835149114119,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.03994717728623308,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.20353251898316277,male,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.10714285714285712,0.043908880818752064,male,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13315725762077696,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.4642857142857143,0.23841751953339935,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11692527786948388,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.056784417299438755,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.19296797622977882,female,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.2195994277539342,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3482447452404534,male,old,good,1,0,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.10294926818531969,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.10570045119401342,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.19642857142857142,0.18399911962143722,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.24408495653130846,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.2424892703862661,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.17857142857142858,0.48024650599757895,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.04429404643996919,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.7857142857142857,0.15412127214702323,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.17558049961483438,female,young,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.26961593485198637,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.09425552987784747,male,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.24999999999999997,0.18377902498074172,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.3076923076923077,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06707384175195334,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.12947067238912735,female,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.18713546825134808,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15951359084406297,male,young,good,1,0,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.21239132827115662,male,young,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.08928571428571427,0.06531308462638935,male,old,bad,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.062011665015956854,female,old,good,0,0,1,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.06448772972378122,male,old,bad,1,0,0,0,0,1,0,0,0,0,0,1
+0.4642857142857143,0.18553978210630573,female,old,good,1,0,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.17029822823814242,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.020633872565203038,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.040332342907450205,female,young,good,0,1,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.15081985253659075,female,young,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.049466270496313414,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.001430615164520744,male,young,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06168152305491362,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.029987894794761747,female,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.056784417299438755,female,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.0064927919005172245,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.055463849455265765,male,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.10652580609662154,female,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.14223616154946628,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.18609001870804448,female,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05838010344448112,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.11543963904478925,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.010069329811819083,female,young,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.20408275558490152,female,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.13101133487399583,female,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.12077693408165512,male,old,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.10714285714285712,0.04335864421701331,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.473643666776714,male,old,good,0,0,0,1,0,0,0,1,0,0,0,1
+0.3571428571428571,0.05639925167822163,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3476394849785408,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06019588423021899,male,old,bad,0,0,1,0,0,0,0,1,0,0,1,0
+0.4107142857142857,0.3144602178936943,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.020248706943985912,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.026246285902938263,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.19642857142857142,0.06569825024760646,male,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.18542973478595798,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.10955210740618465,male,old,good,1,0,0,0,0,1,0,0,0,1,0,0
+1.0,0.799603829646748,female,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.09750192582810609,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.019753494002421042,female,old,bad,0,0,1,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15247056234180698,male,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.2272477165181028,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.05827005612413337,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14185099592824915,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08501155496863651,female,old,good,0,1,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.07059535600308132,male,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.26702982282381427,male,old,good,0,0,0,1,0,1,0,0,0,0,0,1
+0.4642857142857143,0.5707604269836029,male,old,good,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.06250687795752173,female,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.23885770881479035,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.17326950588753165,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.04385385715857819,female,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.4091009133927589,female,old,bad,1,0,0,0,0,0,0,1,0,1,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-val.csv
new file mode 100644
index 0000000..3907229
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE-val.csv
@@ -0,0 +1,301 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.24999999999999997,0.07169582920655881,male,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06107626279300099,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.02806206668867613,female,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.14575767580059426,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.7321428571428571,0.15302079894354573,male,young,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.24551557169582922,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.090073731704633,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11131286453174864,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.035105095190932106,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07538241443820842,female,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.1352481567073842,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.03389457466710685,male,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.07142857142857142,0.06404754044239022,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.05849015076482887,male,young,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.12809508088478047,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.1825685044569165,female,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.06289204357873886,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.10267414988445031,male,old,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09695168922636734,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09298998569384835,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.05904038736656762,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.09502586112028172,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.047210300429184546,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.05777484318256851,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.03571428571428571,0.0505667436997909,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.16501595686145043,male,old,good,1,0,0,0,0,0,0,1,0,0,0,1
+0.625,0.1275998679432156,male,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.07147573456586331,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.024760647078243648,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.32067789149334214,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.08928571428571427,0.05260261912622427,female,young,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.060801144492131615,female,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15010454495433037,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.07015516672169032,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.037526136238582586,female,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.4107142857142857,0.12534389787608674,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.15758776273797734,female,old,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.09122922856828436,male,old,bad,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.05447342357213601,female,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.02569604930119952,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.0695499064597777,male,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08688235941454825,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.02145922746781116,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.39517992736876856,female,old,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.0710905689446462,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.5160669087707714,male,young,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.00968416419060196,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.24061846594035435,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.45234951028942444,female,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1904368878617806,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.6877407285132606,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.6785714285714286,0.4199405744470122,male,old,good,1,0,0,0,0,0,0,1,0,0,0,1
+1.0,0.5444040937603168,female,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.05480356553317926,male,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.1840541432816111,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18526466380543635,male,old,good,0,1,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.034995047870584356,female,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.07131066358534169,male,young,bad,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.10570045119401342,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.13002090899086607,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.04682513480796743,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.1513700891383295,female,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.1116430064927919,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.10714285714285712,0.06993507208099482,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.04357873885770881,female,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.3538571585781886,male,old,good,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.10426983602949268,male,young,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.2512930560140861,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.7797402883239792,male,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.02426543413667877,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05441839991196214,male,old,good,0,1,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.1384945526576428,male,old,bad,0,1,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.33069219764498736,male,old,good,0,1,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.027511830086937385,female,old,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.06894464619786508,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.024815670738417523,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.125,0.04974138879718279,female,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.3035714285714286,0.16011885110597557,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.10311433916584131,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02839220864971938,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.3685484758446132,male,old,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.0482007263123143,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05315285572796302,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.10806646858149004,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.20864971937933313,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.04418399911962144,male,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.14559260482007264,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.327335754374381,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.41311764058545175,female,old,good,0,1,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.06173654671508748,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.33564432706063607,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.2304941124683614,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.8485748872014967,male,young,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.030483107736326624,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3035714285714286,0.07439198855507868,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.3421371189611533,male,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.06729393639264884,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.2851876306811929,female,old,good,0,0,0,1,1,0,0,0,0,1,0,0
+0.03571428571428571,0.05436337625178826,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.5178571428571429,0.13832948167712117,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09199955981071861,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.1522504677011115,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.052657642786398146,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.21428571428571427,0.13068119291295258,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.023880268515461652,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.07813359744690217,female,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.02321998459337515,female,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.3035714285714286,0.12820512820512822,male,old,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.3336634752943766,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.1641906019588423,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.10714285714285712,0.3883569935072081,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.1607142857142857,0.10184879498184218,female,young,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.32898646417959726,female,young,bad,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.15219544404093763,male,young,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.014966435567293938,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.0926598437328051,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.1376141740948608,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.06943985913942995,male,old,bad,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.12396830637173985,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.05705953560030813,female,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.6679322108506658,male,old,bad,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.05981071860900186,male,old,bad,0,0,1,0,0,1,0,0,0,0,1,0
+0.4107142857142857,0.12061186310113349,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.06129635743369649,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11742049081104876,male,old,good,1,0,0,0,0,0,0,0,1,1,0,0
+0.08928571428571427,0.1402553097832068,male,old,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06575327390778034,male,old,good,0,0,1,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.08215032463959503,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06036095521074061,male,old,good,0,0,1,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.25420931000330144,male,old,bad,1,0,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.0817651590183779,female,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.027731924727632886,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.18075272367117862,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.0675690546935182,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.35721360184879497,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.37839771101573677,female,young,bad,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.030593155056674374,female,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14537251017937713,male,old,good,0,0,1,0,0,0,0,0,1,0,0,1
+0.5714285714285714,0.30538131396500495,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15709254979641246,male,young,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.19642857142857142,0.06250687795752173,male,old,good,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.11852096401452623,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19962583911081766,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15637724221415208,male,young,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.4285714285714286,0.1371739848134698,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07285132607021019,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.0872125013755915,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13579839330912294,female,young,bad,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.2527786948387807,male,old,bad,0,0,0,1,1,0,0,0,0,1,0,0
+0.3571428571428571,0.052217453505007144,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.17857142857142858,0.20485308682733577,male,young,good,1,0,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.1336524705623418,female,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.10360955210740619,female,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.05397821063057115,female,old,good,1,0,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.14427203697589966,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.11978650819852536,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+1.0,0.3877517332452955,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.4642857142857143,0.40420380763728403,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.625,0.7664245625619016,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.08928571428571427,0.06883459887751733,female,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.19775503466490593,male,old,bad,0,1,0,0,0,0,0,1,0,0,0,1
+0.6785714285714286,0.20864971937933313,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.06184659403543523,female,old,good,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.41619896555518876,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.03626059205458347,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.34158688235941453,male,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.11962143721800374,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.13965004952129417,female,young,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.024595576097722022,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.3035714285714286,0.12248266754704522,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.0706503796632552,male,young,bad,0,1,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.04500935402222955,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18284362275778587,female,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.05782986684274238,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3805436337625179,male,old,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.09843732805106195,male,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.12286783316826234,male,young,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18526466380543635,male,young,good,0,0,1,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.09948277759436558,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.08281060856168151,female,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.07202597116760207,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.16017387476614944,male,old,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.5486959392538792,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.062231759656652355,female,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.10955210740618465,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.12589413447782546,male,old,bad,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.4300099042588313,female,young,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.33927588863211183,female,old,bad,0,1,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.24683613954000222,female,young,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.037416088918234836,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.1463629360625069,male,old,good,1,0,0,0,0,0,0,1,0,0,0,1
+0.24999999999999997,0.11158798283261802,female,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.2318146803125344,male,old,bad,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11797072741278751,female,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1103224386486189,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.625,0.4587872785297678,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3022449653350941,female,young,good,0,1,0,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.05227247716518102,female,young,good,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.08897325850115549,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.0335644327060636,female,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.04258831297457907,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+1.0,0.36238582590513924,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07868383404864092,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06399251678221635,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.03615054473423572,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.05188731154396391,female,old,bad,1,0,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.1280400572246066,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.2047980631671619,female,young,good,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.1930229998899527,female,young,good,0,1,0,0,1,0,0,0,0,1,0,0
+0.24999999999999997,0.11389897655992078,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.12176735996478487,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.06014086057004511,male,old,good,0,0,1,0,0,0,0,0,1,0,1,0
+0.4285714285714286,0.4167492021569275,male,old,good,0,0,0,1,0,1,0,0,0,1,0,0
+0.3035714285714286,0.16644657202597118,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.16474083856058105,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.46016287003411466,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.08759766699680863,female,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.12991086167051832,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.08928571428571427,0.09794211510949709,male,young,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.49669858038956755,male,old,good,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.045449543303620554,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.07131066358534169,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09040387366567623,female,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.10404974138879718,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.425332893144052,male,old,bad,0,0,0,1,0,1,0,0,0,1,0,0
+0.14285714285714285,0.055793991416309016,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.24999999999999997,0.10999229668757565,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.324254429404644,female,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.1842742379223066,female,old,good,1,0,0,0,0,0,0,1,0,1,0,0
+0.19642857142857142,0.07081545064377683,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.590073731704633,female,old,bad,0,0,0,1,1,0,0,0,0,0,1,0
+0.24999999999999997,0.2226807527236712,male,old,bad,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05172224056344228,female,old,bad,1,0,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.056344228018047754,female,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.11863101133487398,male,old,good,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.33856058104985143,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.10714285714285712,0.05705953560030813,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3138549576317817,female,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.09392538791680423,female,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.07164080554638494,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.03780125453945196,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.09805216242984482,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.5714285714285714,0.5846263893474194,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.2659843732805106,female,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.2712116209970287,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.044018928139099814,male,old,good,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.2015516672169033,male,young,good,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.15401122482667548,female,old,good,0,0,1,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.44486629250577747,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.08847804555959062,female,young,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.25272367117860683,male,old,bad,1,0,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.19423352041377795,male,old,good,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.09783206778914934,male,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11191812479366127,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.1363486299108617,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.4107142857142857,0.27748431825685044,male,old,bad,1,0,0,0,1,0,0,0,0,0,1,0
+1.0,0.4900957411687025,male,old,good,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09315505667436998,female,young,bad,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.148949048090679,male,old,good,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.11769560911191813,female,old,good,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.08979861340376362,female,young,bad,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.08060966215472654,female,old,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.33459887751733247,male,old,bad,1,0,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.0586552217453505,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.8058765269065697,male,old,bad,1,0,0,0,0,0,0,0,1,0,1,0
+0.4642857142857143,0.15555188731154398,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.0,0.07120061626499395,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.041652910751623196,male,young,good,0,1,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.10008803785627819,male,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.7002310993727302,female,old,bad,0,1,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.1570375261362386,male,old,good,1,0,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.20578848905029165,female,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.17877187190491914,female,old,bad,1,0,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.048750962914053037,male,old,bad,0,1,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.007813359744690218,female,young,good,0,0,1,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.048475844613183675,female,old,bad,1,0,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.06459777704412897,male,old,good,0,1,0,0,0,0,0,1,0,0,1,0
+0.6785714285714286,0.17299438758666227,male,old,good,1,0,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.3713546825134808,male,old,bad,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.6464179597226807,male,old,bad,1,0,0,0,0,1,0,0,0,0,1,0
+0.4642857142857143,0.14201606690877078,female,old,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09436557719819522,female,old,good,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.07411687025420931,male,old,good,0,0,0,1,0,0,0,0,1,0,0,1
+0.03571428571428571,0.09975789589523495,male,old,good,0,1,0,0,0,1,0,0,0,1,0,0
+0.3928571428571428,0.42456256190161773,male,old,good,0,1,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.4034334763948498,male,old,good,0,1,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.24369979091009136,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.30543633762517886,male,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.1990756025090789,female,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.11775063277209201,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.3631561571475735,male,old,bad,0,1,0,0,1,0,0,0,0,0,1,0
+0.08928571428571427,0.0675690546935182,male,young,bad,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.10069329811819082,male,young,good,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.03747111257840871,female,old,bad,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.05469351821283151,female,old,good,0,0,0,1,0,0,0,0,1,1,0,0
+0.10714285714285712,0.10355452844723231,female,old,good,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.0935402222955871,male,old,good,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.06184659403543523,male,old,good,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.06597336854847584,male,old,good,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.07329151535160118,male,old,good,0,0,0,1,0,0,0,0,1,0,1,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-test.csv
new file mode 100644
index 0000000..912f248
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-test.csv
@@ -0,0 +1,201 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.5714285714285714,0.30802244965335096,1,0,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11290855067679102,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.04693518212831517,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.0704302850225597,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08093980411576977,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.1389347419390338,1,1,0,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.08974358974358974,0,1,1,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.15291075162319798,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.07741828986464179,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.11571475734565863,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.13255199735886433,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.07384175195333993,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09216463079124024,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.594475624518543,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09447562451854297,0,0,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.13321228128095083,0,1,0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.17497523935292178,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.23236491691427316,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.24347969626939586,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.041212721470232194,1,0,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.5893584241223726,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.05876526906569825,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.019753494002421042,0,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.1509298998569385,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+1.0,0.847529437658193,1,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.17156377242214152,1,0,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.7857142857142857,0.24969736986904373,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.10982722570705403,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.08237041928029053,1,0,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.2666446572025971,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.034885000550236606,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.32023770221195114,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06228678331682623,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.02767690106745901,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.13464289644547156,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.01870804445911742,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.4642857142857143,0.09183448883019699,1,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.3035714285714286,0.07686805326290305,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.3090128755364807,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3132496973698691,0,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.026246285902938263,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.14097061736546715,0,1,1,1,0,0,0,0,0,0,0,1,1,0,0
+0.7857142857142857,0.22328601298558381,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.3266754704522945,1,0,1,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.05749972488169913,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.3004842082095301,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.32243864861890614,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.44767249917464513,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09414548255749973,0,0,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.04996148343787829,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06382744580169472,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.5178571428571429,0.385330692197645,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.023825244855287777,0,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.06399251678221635,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.2817761637504127,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.2217453505007153,0,1,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06096621547265324,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.0292175635523275,1,1,1,0,0,1,0,0,0,0,0,1,0,0,1
+0.0,0.17222405634422802,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.08928571428571427,0.045779685264663805,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.11483437878287663,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.053571428571428575,0.03279410146362936,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.7857142857142857,0.45322988885220644,0,0,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.09106415758776273,1,1,1,0,0,1,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.08550676791020138,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.022119511389897654,0,0,1,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.059480576647958625,1,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.15566193463189174,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.1340926598437328,1,1,1,0,0,1,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.05942555298778475,1,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05051172003961703,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.17585561791570375,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.052217453505007144,0,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.02701661714537251,0,0,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.39198855507868385,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.10559040387366568,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.05986574226917574,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.33944095961263343,0,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.039892153626059204,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13392758886321118,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.23528117090348852,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.038571585781886214,0,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.8928571428571428,0.8635963464289644,1,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.18190822053483,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.0,0.06630351050951909,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09271486739297898,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.1581379993397161,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.07114559260482008,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14933421371189612,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.24999999999999997,0.06558820292725871,1,1,0,1,0,0,0,0,0,0,1,0,0,0,1
+1.0,0.7431495543083525,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.2857142857142857,0.3245295477055134,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3336084516342027,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.41900517222405637,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05760977220204688,1,0,1,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.121712336304611,1,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.3582590513920986,1,1,0,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.5129305601408606,1,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.39391438318476946,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.36519203257400684,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18548475844613185,1,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.06063607351160999,0,0,1,1,0,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.06767910201386595,1,1,1,0,0,1,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.042808407615274574,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.035875426433366345,1,0,0,1,0,0,0,0,0,0,1,0,1,0,0
+0.10714285714285712,0.1136238582590514,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.09519093210080334,0,0,0,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07169582920655881,0,0,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.10714285714285712,0.05447342357213601,0,1,0,0,0,1,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05353802134918015,1,1,0,0,1,0,0,0,0,0,0,1,1,0,0
+0.3035714285714286,0.12963574336964895,0,1,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.06377242214152085,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.07252118410916694,0,0,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.01785714285714285,0.1759656652360515,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.1992956971497744,0,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.116430064927919,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.4087157477715418,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.5033564432706064,1,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.5714285714285714,0.5126004181798173,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.0928799383735006,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.10448993067018818,1,1,0,1,0,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.350170573346539,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.14355672939363928,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.2857142857142857,0.13090128755364808,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.32761087267525035,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.18851105975569496,1,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.114229118520964,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.17244415098492352,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.1546164850885881,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05771981952239463,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06426763508308572,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.04104765048971058,0,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.2857142857142857,0.1780015406624849,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.17211400902388027,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.05359304500935402,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.02723671178606801,0,1,0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08858809287993837,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.2857142857142857,0.2213051612193243,0,0,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.053571428571428575,0.11912622427643886,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19406844943325632,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.10625068779575217,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.239022779795312,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.028007043028502255,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.23550126554418402,0,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.10714285714285712,0.09210960713106636,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.17591064157587763,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+1.0,0.7580609662154726,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.14383184769450866,1,1,0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05529877847474414,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.6785714285714286,0.22669747991636405,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.03719599427753935,0,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.030373060415978873,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.04632992186640256,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.07378672829316606,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.23131946737096953,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.023770221195113902,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.3362495873225487,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.055463849455265765,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.4107142857142857,0.12490370859469573,1,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.00484208209530098,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.2857142857142857,0.37454605480356556,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.2857142857142857,0.10922196544514141,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06459777704412897,0,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.6095521074061846,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05496863651370089,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.08928571428571427,0.10278419720479806,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05694948827996038,0,1,0,1,0,0,0,0,0,0,1,0,1,0,0
+0.7857142857142857,0.18311874105865522,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.14295146913172666,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.17932210850665786,0,0,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.07142857142857142,0.05029162539892153,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.2050181578078574,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.18498954550456698,0,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+1.2142857142857142,0.2941014636293606,1,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.025255860019808517,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.43903378452734676,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.17409486079013978,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.6802024870694399,1,1,0,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.08253549026081215,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.15555188731154398,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.2791900517222406,0,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.0,0.06844943325630021,1,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.12809508088478047,0,1,1,1,0,0,0,1,0,0,0,0,0,0,1
+0.053571428571428575,0.11439418950148562,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.022229558710245404,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.4337515131506548,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19880048420820953,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.39738087377572356,0,0,1,0,1,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.07180587652690656,1,1,0,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.055683944095961266,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.02971277649389237,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.27423792230659183,1,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.11863101133487398,0,0,0,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.1496093320127655,0,0,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3035714285714286,0.6839440959612633,1,1,0,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.045779685264663805,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.20166171453725104,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.19709475074281943,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09232970177176185,1,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.5286123032904149,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-train.csv
new file mode 100644
index 0000000..cd248b0
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-train.csv
@@ -0,0 +1,501 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.5714285714285714,0.20408275558490152,1,1,1,1,0,0,0,1,0,0,0,0,0,1,0
+0.08928571428571427,0.18306371739848135,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.12462859029382635,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09601628700341147,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+1.0,0.5566193463189171,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.39303400462198745,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.008748761967646089,0,1,1,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.12418840101243535,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.4642857142857143,0.25096291405304283,0,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.7857142857142857,0.3471992956971498,1,0,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.48332783096731596,1,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1816881258941345,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.2857142857142857,0.3421371189611533,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.1250137559150435,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.14141080664685815,1,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.029327610872675252,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.19329811819082207,1,0,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.39809618135798397,0,1,0,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.06344228018047761,0,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.125,0.24876196764608782,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.20743919885550788,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.052327500825354895,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.28232640035215145,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.28947947617475517,1,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.04666006382744581,0,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.7857142857142857,0.32871134587872786,1,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.125,0.06289204357873886,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.08928571428571427,0.09183448883019699,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.19642857142857142,0.2638384505337295,0,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.3136898866512601,0,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.07142857142857142,0.026466380543633764,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.02085396720589854,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.08572686255089688,0,0,0,0,0,0,1,0,0,0,1,0,0,1,0
+0.3035714285714286,0.05161219324309453,0,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.16424562561901618,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.6223175965665235,0,0,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.04974138879718279,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.19808517662594918,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.11604489930670188,1,1,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.07059535600308132,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.21789369428854408,1,1,0,0,0,1,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.19847034224716628,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06289204357873886,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.08528667326950588,1,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1101023440079234,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03224386486189061,0,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.06025090789039286,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.2500825354902608,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.025200836359634642,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05716958292065588,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08880818752063387,0,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.5342797402883239,0,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.27797953119841534,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.03571428571428571,0.08924837680202487,1,1,1,1,0,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.12044679212061186,0,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.06261692527786948,0,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.6558270056124132,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.13271706833938596,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.1607142857142857,0.06377242214152085,0,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.1866952789699571,0,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.358093980411577,0,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.0880928799383735,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.4107142857142857,0.17420490811048753,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.2678001540662485,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.10714285714285712,0.07411687025420931,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.06228678331682623,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.05397821063057115,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.13101133487399583,0,1,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.10454495433036205,0,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1474634092659844,0,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.3582590513920986,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3035714285714286,0.13728403213381754,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.11417409486079015,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05249257180587652,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.04396390447892594,0,0,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11235831407505227,1,1,0,0,0,0,1,1,0,0,0,0,0,1,0
+0.08928571428571427,0.06597336854847584,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.39952679652250467,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.03675580499614835,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1153295917244415,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.07444701221525256,0,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.125,0.2011114779355123,1,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0482007263123143,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.3035714285714286,0.14707824364476726,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1620446792120612,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.24573566633652472,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.7871134587872785,1,0,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.2884890502916254,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.0734565863321228,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.18377902498074172,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.1380543633762518,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.10300429184549356,1,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.09640145262462858,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.09084406294706723,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06839440959612633,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.03571428571428571,0.052767690106745896,1,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.20518322878837902,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.4110817651590184,0,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.1870804445911742,0,0,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.08968856608341587,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09304500935402223,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.034334763948497854,0,0,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.4107142857142857,0.2718168812589414,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.5627269725982172,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.22565203037306042,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06195664135578298,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.4642857142857143,0.1786068009243975,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.6428571428571428,0.3162759986794322,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.4285714285714286,0.20666886761307363,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.1701331572576208,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.13051612193243095,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11742049081104876,1,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07752833718498954,0,1,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1781666116430065,0,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+1.0,0.39440959612633436,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.10938703642566304,0,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.059590623968306375,1,1,0,1,0,0,0,1,0,0,0,0,0,0,1
+0.3214285714285714,0.05683944095961263,0,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.17981732144822274,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.018432926158248045,1,0,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.625,0.25767580059425554,1,0,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.056564322658743255,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.08352591614394189,1,0,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.17051832287883792,1,1,0,1,0,0,0,0,1,0,0,0,0,0,1
+0.08928571428571427,0.0627819962583911,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.17453505007153078,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.10432485968966655,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.06988004842082095,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3627159678661825,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.09909761197314845,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15109497083746012,1,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18763068119291296,0,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06558820292725871,0,0,0,1,0,0,0,0,0,0,1,0,1,0,0
+0.08928571428571427,0.04886101023440079,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.3037856278199626,0,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.12930560140860572,1,1,0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19252778694838782,0,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.15456146142841423,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.022669747991636405,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.09298998569384835,1,1,1,1,0,0,0,0,1,0,0,0,0,0,1
+0.5714285714285714,0.1140640475404424,1,0,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.7857142857142857,0.6577528337184989,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.28925938153405967,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.24999999999999997,0.010069329811819083,0,0,0,0,0,0,1,1,0,0,0,0,1,0,0
+0.03571428571428571,0.23995818201826785,0,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.13821943435677342,0,0,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.4642857142857143,0.2213051612193243,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.625,0.6333223286012986,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.5714285714285714,0.10867172884340266,1,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.06734896005282272,0,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.06338725652030373,0,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.24999999999999997,0.09150434686915374,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.17857142857142858,0.030373060415978873,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.05617915703752613,1,1,0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.055463849455265765,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.18240343347639487,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.17255419830527127,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.1098822493672279,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.30560140860570045,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.03906679872345109,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.044349070100143065,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3322878837900297,1,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.13772422141520854,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.10784637394079453,1,0,0,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.5528227137669197,1,1,0,1,0,0,0,0,0,0,1,0,0,0,1
+0.5714285714285714,0.7740728513260702,1,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.4642857142857143,0.2200396170353252,1,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.03736106525806096,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.08928571428571427,0.039011775063277215,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.13931990756025092,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.7857142857142857,0.44613183668977663,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.11285352701661715,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.7857142857142857,0.18245845713656875,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.20446792120611865,0,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.03571428571428571,0.05430835259161439,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.03994717728623308,0,0,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.03659073401562672,0,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.03835149114119071,1,1,0,0,0,1,0,0,0,0,1,0,0,0,1
+0.3571428571428571,0.14168592494772753,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.11213821943435677,0,0,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.10714285714285712,0.145867723120942,0,1,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.7321428571428571,0.23803235391218225,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.3845053372950369,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.13733905579399142,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.08215032463959503,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.06140640475404424,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.06905469351821283,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.005117200396170352,0,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.10856168152305491,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.06850445691647408,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.11428414218113787,0,0,0,0,0,1,0,0,1,0,0,0,0,1,0
+0.7321428571428571,0.6364036535710355,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.06789919665456146,1,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.4844833278309673,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.07730824254429404,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.06619346318917134,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.053813139650049524,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.037526136238582586,0,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.4311654011224827,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.06030593155056674,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.5714285714285714,0.1437218003741609,1,1,1,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.15808297567954221,0,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+1.0,0.3938593595245956,0,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.055793991416309016,1,0,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.07175085286673269,0,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.3035714285714286,0.18719049191152196,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.20254209310003302,1,1,0,1,0,0,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.061131286453174866,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.052437548145702645,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18031253438978762,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.4184549356223176,0,0,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.17095851216022892,0,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.10179377132166832,1,1,0,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0505667436997909,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.47854077253218885,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.24999999999999997,0.09276989105315285,1,0,1,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.1794871794871795,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.3695939253879168,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.4642857142857143,0.11461428414218112,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09843732805106195,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.49636843842852424,1,0,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.3035714285714286,0.07274127874986244,0,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.011224826675470454,1,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.33311323869263787,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.10713106635853417,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.7857142857142857,0.40601958842302194,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.05639925167822163,0,0,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.37795752173434577,1,1,0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.6783867062837019,0,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.24999999999999997,0.1391548365797293,1,0,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.0,0.06894464619786508,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.07142857142857142,0.05430835259161439,0,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.20160669087707717,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15472653240893586,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.5714285714285714,0.09370529327610873,1,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.4642857142857143,0.19709475074281943,1,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.0708704743039507,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.03571428571428571,0.10223396060305931,1,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.00979421151094971,0,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.2313744910311434,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.1138439528997469,1,1,0,0,0,1,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.05023660173874765,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.037746230879278087,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.12231759656652359,1,0,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.19642857142857142,0.07257620776934082,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.1522504677011115,0,1,1,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11450423682183337,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.10157367668097282,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.04330362055683944,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.05425332893144051,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.21464729833828547,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.18482447452404535,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06393749312204247,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.125,0.20298228238142402,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.0702101903818642,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.3214285714285714,0.1334323759216463,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.4642857142857143,0.2184989545504567,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14713326730494114,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08622207549246176,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.0536480686695279,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08776273797733025,0,1,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.0,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022559700671288655,1,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.03571428571428571,0.025255860019808517,1,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.8587542643336634,1,1,1,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.08732254869593925,0,1,0,0,1,0,0,0,0,0,1,0,0,0,1
+0.1607142857142857,0.08512160228898426,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.010894684714427203,0,0,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.16303510509519095,1,0,1,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.12815010454495435,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.05216242984483327,0,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.26152745680642675,0,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11461428414218112,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.055683944095961266,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.049796412457356665,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.03543523715197534,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.16952789699570817,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.04748541873005392,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.02580609662154727,1,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.09981291955540883,0,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.4642857142857143,0.18741058655221746,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3502255970067129,1,1,1,1,0,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.16391548365797293,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.19642857142857142,0.17018818091779467,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.1572576207769341,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.011444921316165951,1,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06718388907230109,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.125,0.05898536370639375,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.057114559260482006,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.025145812699460767,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.058215032463959496,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.08928571428571427,0.15538681633102236,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.12440849565313085,0,0,1,1,0,0,0,1,0,0,0,0,1,0,0
+0.03571428571428571,0.0819852536590734,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.04803565533179267,0,1,0,0,0,0,1,0,0,1,0,0,1,0,0
+0.125,0.07301639705073182,0,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.7321428571428571,0.08776273797733025,1,0,0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09783206778914934,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.4107142857142857,0.4439308902828216,0,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.05749972488169913,0,0,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.6431165401122483,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.08633212281280951,1,1,0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.0675690546935182,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06965995378012545,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1037746230879278,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.19599427753934193,0,0,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.21541762958071972,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.20947507428194126,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.543468691537361,1,1,0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.012875536480686695,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+1.0,0.33223286012985587,1,1,0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.36436667767139874,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.07054033234290745,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.10366457576758006,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.04715527676901067,1,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.2857142857142857,0.10795642126114229,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.16154946627049632,0,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.7678571428571428,0.5762077693408165,0,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.053571428571428575,0.02641135688345989,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.10030813249697369,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.0176625949158138,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.19642857142857142,0.11609992296687575,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06019588423021899,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.6785714285714286,0.3809838230439089,0,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.6785714285714286,0.2501375591504347,1,1,1,0,0,1,0,0,0,0,0,1,0,0,1
+0.6785714285714286,0.20441289754594477,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.125,0.38395510069329813,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.05315285572796302,1,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.13744910311433917,0,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.6785714285714286,0.49702872235061074,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.4107142857142857,0.23528117090348852,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.3035714285714286,0.2750082535490261,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.16127434796962695,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.02569604930119952,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.055683944095961266,1,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08990866072411137,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.09210960713106636,0,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.39760096841641906,1,1,1,1,0,0,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.03791130185979971,1,1,0,0,0,1,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.03251898316275999,0,0,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.21255639925167824,0,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.2697259821723341,1,0,1,0,0,1,0,0,0,0,1,0,0,1,0
+0.5178571428571429,0.22180037416088919,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.07246616044899307,0,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.004897105755474855,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.19642857142857142,0.05639925167822163,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.021404203807637284,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.1127434796962694,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.086552217453505,1,1,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.41339275888632115,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.03571428571428571,0.07400682293386156,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.007428194123473095,0,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.028942445251458126,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15511169803015298,1,1,1,0,1,0,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.06129635743369649,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.0534279740288324,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.17321448222735777,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.21475734565863322,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08407615274568064,0,1,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.05425332893144051,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.03571428571428571,0.08341586882359414,1,0,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.4642857142857143,0.43039506988004844,1,0,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.268460437988335,0,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.4620886981402003,1,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.21563772422141522,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09271486739297898,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.11813579839330911,1,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.06652360515021459,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.023935292175635527,1,0,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.2396280400572246,1,0,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.125,0.10410476504897105,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.03851656212171234,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.21428571428571427,0.050896885660834154,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.06866952789699571,0,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.09541102674149884,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.19555408825795093,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09557609772202047,1,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.21398701441619897,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.024815670738417523,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.8928571428571428,0.5054473423572136,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.16176956091119182,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.1607142857142857,0.034774953229888855,1,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.13810938703642567,1,1,1,0,1,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.10845163420270716,0,0,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.020688896225376913,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,1.0,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0532078793881369,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.053571428571428575,0.12798503356443272,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.13189171343677783,1,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.4107142857142857,0.12765489160338947,0,0,0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.1594585671838891,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022889842632331906,0,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.051447122262572906,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.3138549576317817,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.7321428571428571,0.24738637614174097,1,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.0529877847474414,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.09067899196654561,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.7857142857142857,0.6239683063717398,0,0,0,0,0,0,1,0,0,1,0,0,1,0,0
+0.0,0.01931330472103004,0,0,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.24999999999999997,0.05474854187300539,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.30218994167492025,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.17857142857142858,0.06382744580169472,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.035765379113018594,1,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.0061626499394739735,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.09513590844062947,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.17051832287883792,0,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.030097942115109497,0,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.014746340926598437,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1129635743369649,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.1354682513480797,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06734896005282272,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.08638714647298339,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09227467811158797,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06509298998569385,0,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.03895675140310334,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.19725982172334106,0,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.028227137669197756,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.15010454495433037,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05727963024100363,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.0233300319137229,1,0,0,1,0,0,0,0,0,0,1,0,0,1,0
+1.0,0.3453835149114119,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.03994717728623308,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.20353251898316277,1,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.10714285714285712,0.043908880818752064,1,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13315725762077696,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.4642857142857143,0.23841751953339935,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11692527786948388,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.056784417299438755,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.19296797622977882,0,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.2195994277539342,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3482447452404534,1,1,1,1,0,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.10294926818531969,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.10570045119401342,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.19642857142857142,0.18399911962143722,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.5714285714285714,0.24408495653130846,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.2424892703862661,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.17857142857142858,0.48024650599757895,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.04429404643996919,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.7857142857142857,0.15412127214702323,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.17558049961483438,0,0,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.26961593485198637,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.09425552987784747,1,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.24999999999999997,0.18377902498074172,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.3076923076923077,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06707384175195334,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.12947067238912735,0,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.18713546825134808,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15951359084406297,1,0,1,1,0,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.21239132827115662,1,0,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.08928571428571427,0.06531308462638935,1,1,0,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.062011665015956854,0,1,1,0,0,1,0,0,0,0,0,1,0,0,1
+0.08928571428571427,0.06448772972378122,1,1,0,1,0,0,0,0,1,0,0,0,0,0,1
+0.4642857142857143,0.18553978210630573,0,1,1,1,0,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.17029822823814242,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.020633872565203038,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.040332342907450205,0,0,1,0,1,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.15081985253659075,0,0,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.049466270496313414,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.001430615164520744,1,0,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06168152305491362,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.029987894794761747,0,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.056784417299438755,0,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.0064927919005172245,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.055463849455265765,1,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.10652580609662154,0,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.14223616154946628,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.18609001870804448,0,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05838010344448112,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.11543963904478925,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.010069329811819083,0,0,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.20408275558490152,0,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.13101133487399583,0,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.12077693408165512,1,1,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.10714285714285712,0.04335864421701331,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.473643666776714,1,1,1,0,0,0,1,0,0,0,1,0,0,0,1
+0.3571428571428571,0.05639925167822163,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3476394849785408,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.06019588423021899,1,1,0,0,0,1,0,0,0,0,1,0,0,1,0
+0.4107142857142857,0.3144602178936943,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.020248706943985912,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.10714285714285712,0.026246285902938263,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.19642857142857142,0.06569825024760646,1,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.18542973478595798,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.10955210740618465,1,1,1,1,0,0,0,0,1,0,0,0,1,0,0
+1.0,0.799603829646748,0,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.3571428571428571,0.09750192582810609,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.019753494002421042,0,1,0,0,0,1,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15247056234180698,1,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.2272477165181028,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.05827005612413337,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14185099592824915,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.08501155496863651,0,1,1,0,1,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.07059535600308132,1,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.26702982282381427,1,1,1,0,0,0,1,0,1,0,0,0,0,0,1
+0.4642857142857143,0.5707604269836029,1,1,1,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.06250687795752173,0,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.23885770881479035,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.17326950588753165,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.03571428571428571,0.04385385715857819,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.4091009133927589,0,1,0,1,0,0,0,0,0,0,1,0,1,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-val.csv
new file mode 100644
index 0000000..ac5e47f
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME-val.csv
@@ -0,0 +1,301 @@
+month,credit_amount,sex,age,credit,status=A11,status=A12,status=A13,status=A14,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153
+0.24999999999999997,0.07169582920655881,1,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.06107626279300099,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.02806206668867613,0,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.14575767580059426,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.7321428571428571,0.15302079894354573,1,0,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.24551557169582922,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.090073731704633,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11131286453174864,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.14285714285714285,0.035105095190932106,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07538241443820842,0,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.1352481567073842,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.24999999999999997,0.03389457466710685,1,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.07142857142857142,0.06404754044239022,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.05849015076482887,1,0,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.12809508088478047,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.1825685044569165,0,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.06289204357873886,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.10267414988445031,1,1,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09695168922636734,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.09298998569384835,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.05904038736656762,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.09502586112028172,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.047210300429184546,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.19642857142857142,0.05777484318256851,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.03571428571428571,0.0505667436997909,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.16501595686145043,1,1,1,1,0,0,0,0,0,0,1,0,0,0,1
+0.625,0.1275998679432156,1,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.08928571428571427,0.07147573456586331,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.024760647078243648,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.14285714285714285,0.32067789149334214,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.08928571428571427,0.05260261912622427,0,0,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.060801144492131615,0,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15010454495433037,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.07015516672169032,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.037526136238582586,0,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.4107142857142857,0.12534389787608674,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.15758776273797734,0,1,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.09122922856828436,1,1,0,0,0,1,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.05447342357213601,0,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.02569604930119952,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.0695499064597777,1,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08688235941454825,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.02145922746781116,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.39517992736876856,0,1,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.0710905689446462,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.5160669087707714,1,0,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.00968416419060196,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.24061846594035435,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.45234951028942444,0,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1904368878617806,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.6877407285132606,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.6785714285714286,0.4199405744470122,1,1,1,1,0,0,0,0,0,0,1,0,0,0,1
+1.0,0.5444040937603168,0,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.05480356553317926,1,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.1840541432816111,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18526466380543635,1,1,1,0,1,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.034995047870584356,0,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.07131066358534169,1,0,0,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.10570045119401342,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.13002090899086607,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.04682513480796743,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.1513700891383295,0,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.1116430064927919,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.10714285714285712,0.06993507208099482,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.04357873885770881,0,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.3538571585781886,1,1,1,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.10426983602949268,1,0,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.2512930560140861,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.7797402883239792,1,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.02426543413667877,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05441839991196214,1,1,1,0,1,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.1384945526576428,1,1,0,0,1,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.33069219764498736,1,1,1,0,1,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.027511830086937385,0,1,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.06894464619786508,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.024815670738417523,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.125,0.04974138879718279,0,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.3035714285714286,0.16011885110597557,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.10311433916584131,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02839220864971938,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.3685484758446132,1,1,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.0482007263123143,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05315285572796302,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.10806646858149004,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.20864971937933313,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.04418399911962144,1,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.14559260482007264,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.327335754374381,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.41311764058545175,0,1,1,0,1,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.06173654671508748,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.33564432706063607,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.2304941124683614,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.8485748872014967,1,0,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.030483107736326624,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3035714285714286,0.07439198855507868,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.3421371189611533,1,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.24999999999999997,0.06729393639264884,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.2851876306811929,0,1,1,0,0,0,1,1,0,0,0,0,1,0,0
+0.03571428571428571,0.05436337625178826,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.5178571428571429,0.13832948167712117,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09199955981071861,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.1522504677011115,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.052657642786398146,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.21428571428571427,0.13068119291295258,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.14285714285714285,0.023880268515461652,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.07813359744690217,0,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.03571428571428571,0.02321998459337515,0,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.3035714285714286,0.12820512820512822,1,1,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.3336634752943766,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.1641906019588423,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.10714285714285712,0.3883569935072081,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.1607142857142857,0.10184879498184218,0,0,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.32898646417959726,0,0,0,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.15219544404093763,1,0,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.014966435567293938,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.0926598437328051,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.7857142857142857,0.1376141740948608,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.06943985913942995,1,1,0,0,1,0,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.12396830637173985,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.05705953560030813,0,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.6679322108506658,1,1,0,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.05981071860900186,1,1,0,0,0,1,0,0,1,0,0,0,0,1,0
+0.4107142857142857,0.12061186310113349,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.10714285714285712,0.06129635743369649,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11742049081104876,1,1,1,1,0,0,0,0,0,0,0,1,1,0,0
+0.08928571428571427,0.1402553097832068,1,1,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06575327390778034,1,1,1,0,0,1,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.08215032463959503,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06036095521074061,1,1,1,0,0,1,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.25420931000330144,1,1,0,1,0,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.0817651590183779,0,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.027731924727632886,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.18075272367117862,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.0675690546935182,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.4642857142857143,0.35721360184879497,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.37839771101573677,0,0,0,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.030593155056674374,0,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.14537251017937713,1,1,1,0,0,1,0,0,0,0,0,1,0,0,1
+0.5714285714285714,0.30538131396500495,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15709254979641246,1,0,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.19642857142857142,0.06250687795752173,1,1,1,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.11852096401452623,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.19962583911081766,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.15637724221415208,1,0,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.4285714285714286,0.1371739848134698,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07285132607021019,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3035714285714286,0.0872125013755915,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.13579839330912294,0,0,0,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.2527786948387807,1,1,0,0,0,0,1,1,0,0,0,0,1,0,0
+0.3571428571428571,0.052217453505007144,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.17857142857142858,0.20485308682733577,1,0,1,1,0,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.1336524705623418,0,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.10360955210740619,0,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.3571428571428571,0.05397821063057115,0,1,1,1,0,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.14427203697589966,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.11978650819852536,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+1.0,0.3877517332452955,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.4642857142857143,0.40420380763728403,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.625,0.7664245625619016,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.08928571428571427,0.06883459887751733,0,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.7857142857142857,0.19775503466490593,1,1,0,0,1,0,0,0,0,0,1,0,0,0,1
+0.6785714285714286,0.20864971937933313,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.06184659403543523,0,1,1,1,0,0,0,1,0,0,0,0,0,1,0
+0.3571428571428571,0.41619896555518876,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.03626059205458347,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.34158688235941453,1,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.11962143721800374,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.13965004952129417,0,0,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.024595576097722022,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.3035714285714286,0.12248266754704522,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.14285714285714285,0.0706503796632552,1,0,0,0,1,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.04500935402222955,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.18284362275778587,0,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.05782986684274238,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3805436337625179,1,1,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.09843732805106195,1,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.12286783316826234,1,0,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.18526466380543635,1,0,1,0,0,1,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.09948277759436558,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.08281060856168151,0,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.07202597116760207,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.16017387476614944,1,1,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.5486959392538792,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.062231759656652355,0,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.3035714285714286,0.10955210740618465,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.12589413447782546,1,1,0,0,0,0,1,0,0,0,0,1,0,1,0
+0.5714285714285714,0.4300099042588313,0,0,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.33927588863211183,0,1,0,0,1,0,0,0,0,0,0,1,1,0,0
+0.3571428571428571,0.24683613954000222,0,0,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.037416088918234836,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.1463629360625069,1,1,1,1,0,0,0,0,0,0,1,0,0,0,1
+0.24999999999999997,0.11158798283261802,0,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.2318146803125344,1,1,0,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11797072741278751,0,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.1103224386486189,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.625,0.4587872785297678,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.3022449653350941,0,0,1,0,1,0,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.05227247716518102,0,0,1,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.08897325850115549,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.03571428571428571,0.0335644327060636,0,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.04258831297457907,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+1.0,0.36238582590513924,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.14285714285714285,0.07868383404864092,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.3571428571428571,0.06399251678221635,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.07142857142857142,0.03615054473423572,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.05188731154396391,0,1,0,1,0,0,0,1,0,0,0,0,1,0,0
+0.14285714285714285,0.1280400572246066,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.2047980631671619,0,0,1,0,0,0,1,0,0,0,1,0,1,0,0
+0.3571428571428571,0.1930229998899527,0,0,1,0,1,0,0,1,0,0,0,0,1,0,0
+0.24999999999999997,0.11389897655992078,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.12176735996478487,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.06014086057004511,1,1,1,0,0,1,0,0,0,0,0,1,0,1,0
+0.4285714285714286,0.4167492021569275,1,1,1,0,0,0,1,0,1,0,0,0,1,0,0
+0.3035714285714286,0.16644657202597118,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.24999999999999997,0.16474083856058105,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.46016287003411466,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.5714285714285714,0.08759766699680863,0,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.3571428571428571,0.12991086167051832,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.08928571428571427,0.09794211510949709,1,0,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.49669858038956755,1,1,1,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.045449543303620554,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.10714285714285712,0.07131066358534169,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09040387366567623,0,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.10404974138879718,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.5714285714285714,0.425332893144052,1,1,0,0,0,0,1,0,1,0,0,0,1,0,0
+0.14285714285714285,0.055793991416309016,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.24999999999999997,0.10999229668757565,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.7857142857142857,0.324254429404644,0,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.3035714285714286,0.1842742379223066,0,1,1,1,0,0,0,0,0,0,1,0,1,0,0
+0.19642857142857142,0.07081545064377683,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.590073731704633,0,1,0,0,0,0,1,1,0,0,0,0,0,1,0
+0.24999999999999997,0.2226807527236712,1,1,0,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.05172224056344228,0,1,0,1,0,0,0,1,0,0,0,0,0,0,1
+0.14285714285714285,0.056344228018047754,0,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.11863101133487398,1,1,1,0,0,0,1,1,0,0,0,0,0,1,0
+0.3571428571428571,0.33856058104985143,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.10714285714285712,0.05705953560030813,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.6785714285714286,0.3138549576317817,0,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.09392538791680423,0,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.07164080554638494,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.03780125453945196,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.09805216242984482,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.5714285714285714,0.5846263893474194,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.3571428571428571,0.2659843732805106,0,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.5714285714285714,0.2712116209970287,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.03571428571428571,0.044018928139099814,1,1,1,0,1,0,0,1,0,0,0,0,0,1,0
+0.5714285714285714,0.2015516672169033,1,0,1,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.15401122482667548,0,1,1,0,0,1,0,0,1,0,0,0,0,1,0
+0.5714285714285714,0.44486629250577747,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.14285714285714285,0.08847804555959062,0,0,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.25272367117860683,1,1,0,1,0,0,0,0,0,0,0,1,1,0,0
+0.24999999999999997,0.19423352041377795,1,1,1,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.09783206778914934,1,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.11191812479366127,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.19642857142857142,0.1363486299108617,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.4107142857142857,0.27748431825685044,1,1,0,1,0,0,0,1,0,0,0,0,0,1,0
+1.0,0.4900957411687025,1,1,1,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09315505667436998,0,0,0,0,0,0,1,0,1,0,0,0,0,1,0
+0.3571428571428571,0.148949048090679,1,1,1,1,0,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.11769560911191813,0,1,1,1,0,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.08979861340376362,0,0,0,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.08060966215472654,0,1,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.33459887751733247,1,1,0,1,0,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.0586552217453505,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.8058765269065697,1,1,0,1,0,0,0,0,0,0,0,1,0,1,0
+0.4642857142857143,0.15555188731154398,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.0,0.07120061626499395,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.041652910751623196,1,0,1,0,1,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.10008803785627819,1,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.24999999999999997,0.7002310993727302,0,1,0,0,1,0,0,1,0,0,0,0,0,0,1
+0.24999999999999997,0.1570375261362386,1,1,1,1,0,0,0,0,0,0,1,0,0,1,0
+0.5714285714285714,0.20578848905029165,0,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.17877187190491914,0,1,0,1,0,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.048750962914053037,1,1,0,0,1,0,0,0,0,0,0,1,0,0,1
+0.19642857142857142,0.007813359744690218,0,0,1,0,0,1,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.048475844613183675,0,1,0,1,0,0,0,1,0,0,0,0,0,1,0
+0.14285714285714285,0.06459777704412897,1,1,1,0,1,0,0,0,0,0,1,0,0,1,0
+0.6785714285714286,0.17299438758666227,1,1,1,1,0,0,0,1,0,0,0,0,0,1,0
+0.7857142857142857,0.3713546825134808,1,1,0,1,0,0,0,0,0,0,1,0,0,1,0
+0.4642857142857143,0.6464179597226807,1,1,0,1,0,0,0,0,1,0,0,0,0,1,0
+0.4642857142857143,0.14201606690877078,0,1,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.09436557719819522,0,1,1,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.07411687025420931,1,1,1,0,0,0,1,0,0,0,0,1,0,0,1
+0.03571428571428571,0.09975789589523495,1,1,1,0,1,0,0,0,1,0,0,0,1,0,0
+0.3928571428571428,0.42456256190161773,1,1,1,0,1,0,0,0,1,0,0,0,0,1,0
+0.7857142857142857,0.4034334763948498,1,1,1,0,1,0,0,1,0,0,0,0,0,0,1
+0.3571428571428571,0.24369979091009136,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.14285714285714285,0.30543633762517886,1,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.3571428571428571,0.1990756025090789,0,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.14285714285714285,0.11775063277209201,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
+0.19642857142857142,0.3631561571475735,1,1,0,0,1,0,0,1,0,0,0,0,0,1,0
+0.08928571428571427,0.0675690546935182,1,0,0,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.10069329811819082,1,0,1,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.03747111257840871,0,1,0,0,1,0,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.05469351821283151,0,1,1,0,0,0,1,0,0,0,0,1,1,0,0
+0.10714285714285712,0.10355452844723231,0,1,1,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.0935402222955871,1,1,1,0,0,0,1,0,0,0,1,0,0,1,0
+0.03571428571428571,0.06184659403543523,1,1,1,1,0,0,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.06597336854847584,1,1,1,0,1,0,0,0,0,0,0,1,0,1,0
+0.24999999999999997,0.07329151535160118,1,1,1,0,0,0,1,0,0,0,0,1,0,1,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-test.csv
new file mode 100644
index 0000000..2714c97
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-test.csv
@@ -0,0 +1,201 @@
+month,credit_amount,sex,age,credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.30802244965335096,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11290855067679102,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.04693518212831517,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0704302850225597,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.08093980411576977,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1389347419390338,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08974358974358974,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.15291075162319798,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.07741828986464179,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.11571475734565863,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.13255199735886433,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07384175195333993,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09216463079124024,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.594475624518543,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09447562451854297,0.0,0.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.13321228128095083,0.0,1.0,0.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.17497523935292178,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.23236491691427316,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.24347969626939586,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.041212721470232194,1.0,0.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.5893584241223726,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.05876526906569825,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.019753494002421042,0.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.1509298998569385,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.847529437658193,1.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17156377242214152,1.0,0.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.7857142857142857,0.24969736986904373,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10982722570705403,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08237041928029053,1.0,0.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.2666446572025971,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.034885000550236606,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.32023770221195114,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06228678331682623,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.02767690106745901,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.13464289644547156,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.01870804445911742,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.09183448883019699,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07686805326290305,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3090128755364807,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3132496973698691,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.026246285902938263,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14097061736546715,0.0,1.0,1.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.7857142857142857,0.22328601298558381,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.3266754704522945,1.0,0.0,1.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.24999999999999997,0.05749972488169913,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3004842082095301,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.32243864861890614,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.44767249917464513,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.09414548255749973,0.0,0.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.04996148343787829,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06382744580169472,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5178571428571429,0.385330692197645,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.023825244855287777,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06399251678221635,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.2817761637504127,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.2217453505007153,0.0,1.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06096621547265324,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.0292175635523275,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,1,0
+0.0,0.17222405634422802,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.045779685264663805,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.11483437878287663,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.053571428571428575,0.03279410146362936,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.7857142857142857,0.45322988885220644,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.19642857142857142,0.09106415758776273,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,1,0
+0.14285714285714285,0.08550676791020138,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.022119511389897654,0.0,0.0,1.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.059480576647958625,1.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.15566193463189174,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1340926598437328,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05942555298778475,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05051172003961703,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.17585561791570375,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.052217453505007144,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02701661714537251,0.0,0.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.39198855507868385,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10559040387366568,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.05986574226917574,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33944095961263343,0.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.039892153626059204,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13392758886321118,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23528117090348852,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.038571585781886214,0.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.8928571428571428,0.8635963464289644,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.18190822053483,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.0,0.06630351050951909,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09271486739297898,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.1581379993397161,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.07114559260482008,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14933421371189612,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.24999999999999997,0.06558820292725871,1.0,1.0,0.0,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.7431495543083525,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.2857142857142857,0.3245295477055134,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.3336084516342027,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.41900517222405637,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.05760977220204688,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.121712336304611,1.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3582590513920986,1.0,1.0,0.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.5714285714285714,0.5129305601408606,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39391438318476946,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.36519203257400684,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18548475844613185,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.06063607351160999,0.0,0.0,1.0,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06767910201386595,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.042808407615274574,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.035875426433366345,1.0,0.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.10714285714285712,0.1136238582590514,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.09519093210080334,0.0,0.0,0.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.07169582920655881,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.10714285714285712,0.05447342357213601,0.0,1.0,0.0,0,0,0,0,1,0,0,1,0,0,1,0
+0.14285714285714285,0.05353802134918015,1.0,1.0,0.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.3035714285714286,0.12963574336964895,0.0,1.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06377242214152085,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.07252118410916694,0.0,0.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.01785714285714285,0.1759656652360515,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1992956971497744,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.116430064927919,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4087157477715418,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5033564432706064,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.5714285714285714,0.5126004181798173,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.0928799383735006,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10448993067018818,1.0,1.0,0.0,0,0,1,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.350170573346539,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.14355672939363928,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.2857142857142857,0.13090128755364808,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.32761087267525035,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.18851105975569496,1.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.19642857142857142,0.114229118520964,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.17244415098492352,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.1546164850885881,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05771981952239463,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06426763508308572,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.04104765048971058,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.2857142857142857,0.1780015406624849,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17211400902388027,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.05359304500935402,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.02723671178606801,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08858809287993837,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.2857142857142857,0.2213051612193243,0.0,0.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.053571428571428575,0.11912622427643886,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.19406844943325632,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.10625068779575217,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.239022779795312,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.028007043028502255,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.23550126554418402,0.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.10714285714285712,0.09210960713106636,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17591064157587763,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.7580609662154726,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14383184769450866,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05529877847474414,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.22669747991636405,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.03719599427753935,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.030373060415978873,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.04632992186640256,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07378672829316606,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.23131946737096953,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023770221195113902,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.3362495873225487,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.055463849455265765,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.4107142857142857,0.12490370859469573,1.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.00484208209530098,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.2857142857142857,0.37454605480356556,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.2857142857142857,0.10922196544514141,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.6095521074061846,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05496863651370089,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10278419720479806,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05694948827996038,0.0,1.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.7857142857142857,0.18311874105865522,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.14295146913172666,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17932210850665786,0.0,0.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.07142857142857142,0.05029162539892153,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.2050181578078574,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.18498954550456698,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+1.2142857142857142,0.2941014636293606,1.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025255860019808517,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.43903378452734676,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.17409486079013978,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.6802024870694399,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.03571428571428571,0.08253549026081215,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.15555188731154398,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.2791900517222406,0.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.0,0.06844943325630021,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.12809508088478047,0.0,1.0,1.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.053571428571428575,0.11439418950148562,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.022229558710245404,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.4337515131506548,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19880048420820953,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.39738087377572356,0.0,0.0,1.0,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.07180587652690656,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.02971277649389237,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.27423792230659183,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.11863101133487398,0.0,0.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1496093320127655,0.0,0.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3035714285714286,0.6839440959612633,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.045779685264663805,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.20166171453725104,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.19709475074281943,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09232970177176185,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.5286123032904149,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-train.csv
new file mode 100644
index 0000000..9e3b099
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-train.csv
@@ -0,0 +1,501 @@
+month,credit_amount,sex,age,credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.20408275558490152,1.0,1.0,1.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.18306371739848135,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.12462859029382635,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09601628700341147,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+1.0,0.5566193463189171,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.39303400462198745,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.008748761967646089,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.12418840101243535,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.4642857142857143,0.25096291405304283,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.7857142857142857,0.3471992956971498,1.0,0.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.48332783096731596,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1816881258941345,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.2857142857142857,0.3421371189611533,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1250137559150435,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.14141080664685815,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.029327610872675252,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.19329811819082207,1.0,0.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.39809618135798397,0.0,1.0,0.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06344228018047761,0.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.125,0.24876196764608782,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.20743919885550788,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.052327500825354895,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.28232640035215145,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.28947947617475517,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.04666006382744581,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.32871134587872786,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.125,0.06289204357873886,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.09183448883019699,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.2638384505337295,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.3136898866512601,0.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.026466380543633764,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.02085396720589854,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08572686255089688,0.0,0.0,0.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.05161219324309453,0.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.16424562561901618,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.6223175965665235,0.0,0.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.04974138879718279,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.19808517662594918,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11604489930670188,1.0,1.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07059535600308132,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.21789369428854408,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.19847034224716628,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06289204357873886,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08528667326950588,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.1101023440079234,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03224386486189061,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06025090789039286,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.2500825354902608,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025200836359634642,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.05716958292065588,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08880818752063387,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.5342797402883239,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.27797953119841534,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.08924837680202487,1.0,1.0,1.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.12044679212061186,0.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06261692527786948,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.6558270056124132,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.13271706833938596,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.1607142857142857,0.06377242214152085,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1866952789699571,0.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.7857142857142857,0.358093980411577,0.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.0880928799383735,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.17420490811048753,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.2678001540662485,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07411687025420931,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.06228678331682623,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.05397821063057115,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13101133487399583,0.0,1.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.10454495433036205,0.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.1474634092659844,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.3582590513920986,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.13728403213381754,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11417409486079015,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05249257180587652,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.04396390447892594,0.0,0.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.11235831407505227,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.06597336854847584,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.39952679652250467,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.08928571428571427,0.03675580499614835,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.1153295917244415,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07444701221525256,0.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.125,0.2011114779355123,1.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0482007263123143,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.14707824364476726,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1620446792120612,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.24573566633652472,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.7871134587872785,1.0,0.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2884890502916254,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.3571428571428571,0.0734565863321228,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.18377902498074172,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.1380543633762518,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10300429184549356,1.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.09640145262462858,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.09084406294706723,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.06839440959612633,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.052767690106745896,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20518322878837902,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.4110817651590184,0.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1870804445911742,0.0,0.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.08968856608341587,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09304500935402223,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.034334763948497854,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.2718168812589414,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.5627269725982172,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.22565203037306042,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06195664135578298,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.1786068009243975,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.6428571428571428,0.3162759986794322,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4285714285714286,0.20666886761307363,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.1701331572576208,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.13051612193243095,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11742049081104876,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07752833718498954,0.0,1.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1781666116430065,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+1.0,0.39440959612633436,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10938703642566304,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.059590623968306375,1.0,1.0,0.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.3214285714285714,0.05683944095961263,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.17981732144822274,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.018432926158248045,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.625,0.25767580059425554,1.0,0.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.056564322658743255,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.08352591614394189,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.17051832287883792,1.0,1.0,0.0,0,1,0,0,0,0,0,1,1,0,0,0
+0.08928571428571427,0.0627819962583911,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.17453505007153078,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10432485968966655,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.06988004842082095,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3627159678661825,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09909761197314845,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15109497083746012,1.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18763068119291296,0.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06558820292725871,0.0,0.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.08928571428571427,0.04886101023440079,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.3037856278199626,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.12930560140860572,1.0,1.0,0.0,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.19252778694838782,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.15456146142841423,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022669747991636405,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09298998569384835,1.0,1.0,1.0,0,1,0,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.1140640475404424,1.0,0.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.7857142857142857,0.6577528337184989,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.28925938153405967,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.24999999999999997,0.010069329811819083,0.0,0.0,0.0,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.23995818201826785,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.13821943435677342,0.0,0.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.4642857142857143,0.2213051612193243,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.625,0.6333223286012986,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.5714285714285714,0.10867172884340266,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.06734896005282272,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06338725652030373,0.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.09150434686915374,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.17857142857142858,0.030373060415978873,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.05617915703752613,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.055463849455265765,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.18240343347639487,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17255419830527127,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.1098822493672279,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.30560140860570045,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.03906679872345109,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.044349070100143065,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3322878837900297,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.13772422141520854,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10784637394079453,1.0,0.0,0.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.5528227137669197,1.0,1.0,0.0,0,0,0,1,0,0,0,1,1,0,0,0
+0.5714285714285714,0.7740728513260702,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.2200396170353252,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.03736106525806096,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.039011775063277215,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.13931990756025092,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.44613183668977663,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.11285352701661715,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.18245845713656875,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.20446792120611865,0.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.03571428571428571,0.05430835259161439,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03994717728623308,0.0,0.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.03659073401562672,0.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.03835149114119071,1.0,1.0,0.0,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.14168592494772753,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3035714285714286,0.11213821943435677,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.145867723120942,0.0,1.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.7321428571428571,0.23803235391218225,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3845053372950369,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.13733905579399142,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08215032463959503,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.06140640475404424,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06905469351821283,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.005117200396170352,0.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.10856168152305491,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.06850445691647408,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.11428414218113787,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.7321428571428571,0.6364036535710355,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.06789919665456146,1.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4844833278309673,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.07730824254429404,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.06619346318917134,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.053813139650049524,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.037526136238582586,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4311654011224827,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.06030593155056674,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.1437218003741609,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.15808297567954221,0.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.3938593595245956,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055793991416309016,1.0,0.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.07175085286673269,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.3035714285714286,0.18719049191152196,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20254209310003302,1.0,1.0,0.0,0,0,0,1,0,0,0,1,1,0,0,0
+0.03571428571428571,0.061131286453174866,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.052437548145702645,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.18031253438978762,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.4184549356223176,0.0,0.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.17095851216022892,0.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.10179377132166832,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0505667436997909,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.47854077253218885,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09276989105315285,1.0,0.0,1.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1794871794871795,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3695939253879168,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.4642857142857143,0.11461428414218112,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09843732805106195,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.49636843842852424,1.0,0.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.07274127874986244,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.011224826675470454,1.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.33311323869263787,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.10713106635853417,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.40601958842302194,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.05639925167822163,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.37795752173434577,1.0,1.0,0.0,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.6783867062837019,0.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.24999999999999997,0.1391548365797293,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.0,0.06894464619786508,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.07142857142857142,0.05430835259161439,0.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20160669087707717,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15472653240893586,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.5714285714285714,0.09370529327610873,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.19709475074281943,1.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.0708704743039507,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10223396060305931,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.00979421151094971,0.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.2313744910311434,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.1138439528997469,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05023660173874765,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.037746230879278087,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.12231759656652359,1.0,0.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07257620776934082,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.1522504677011115,0.0,1.0,1.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11450423682183337,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10157367668097282,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04330362055683944,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05425332893144051,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.21464729833828547,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.18482447452404535,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.06393749312204247,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.125,0.20298228238142402,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.0702101903818642,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.3214285714285714,0.1334323759216463,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.2184989545504567,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.14713326730494114,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08622207549246176,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.0536480686695279,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08776273797733025,0.0,1.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022559700671288655,1.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.025255860019808517,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.8587542643336634,1.0,1.0,1.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.08732254869593925,0.0,1.0,0.0,0,0,0,1,0,0,0,1,0,1,0,0
+0.1607142857142857,0.08512160228898426,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010894684714427203,0.0,0.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.16303510509519095,1.0,0.0,1.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.12815010454495435,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05216242984483327,0.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.26152745680642675,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11461428414218112,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055683944095961266,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.049796412457356665,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.10714285714285712,0.03543523715197534,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.16952789699570817,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.04748541873005392,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02580609662154727,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.09981291955540883,0.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.18741058655221746,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.3502255970067129,1.0,1.0,1.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.24999999999999997,0.16391548365797293,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17018818091779467,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.1572576207769341,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.011444921316165951,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.06718388907230109,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.125,0.05898536370639375,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.057114559260482006,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.025145812699460767,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.058215032463959496,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.15538681633102236,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.12440849565313085,0.0,0.0,1.0,1,0,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0819852536590734,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.04803565533179267,0.0,1.0,0.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.125,0.07301639705073182,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.7321428571428571,0.08776273797733025,1.0,0.0,0.0,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.09783206778914934,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.4439308902828216,0.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.05749972488169913,0.0,0.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.6431165401122483,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.08633212281280951,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06965995378012545,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.1037746230879278,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.19599427753934193,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.21541762958071972,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.20947507428194126,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.543468691537361,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.012875536480686695,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+1.0,0.33223286012985587,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.36436667767139874,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07054033234290745,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10366457576758006,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04715527676901067,1.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.2857142857142857,0.10795642126114229,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.16154946627049632,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.7678571428571428,0.5762077693408165,0.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.02641135688345989,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.10030813249697369,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0176625949158138,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.11609992296687575,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06019588423021899,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.6785714285714286,0.3809838230439089,0.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.6785714285714286,0.2501375591504347,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,1,0
+0.6785714285714286,0.20441289754594477,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.125,0.38395510069329813,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05315285572796302,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.13744910311433917,0.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.6785714285714286,0.49702872235061074,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.4107142857142857,0.23528117090348852,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.2750082535490261,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16127434796962695,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.02569604930119952,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08990866072411137,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09210960713106636,0.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.39760096841641906,1.0,1.0,1.0,0,0,0,1,0,0,0,1,1,0,0,0
+0.14285714285714285,0.03791130185979971,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03251898316275999,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.21255639925167824,0.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.2697259821723341,1.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,1,0
+0.5178571428571429,0.22180037416088919,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07246616044899307,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.004897105755474855,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.05639925167822163,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.021404203807637284,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1127434796962694,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.086552217453505,1.0,1.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.41339275888632115,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.07400682293386156,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.007428194123473095,0.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.028942445251458126,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15511169803015298,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.08928571428571427,0.06129635743369649,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0534279740288324,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.17321448222735777,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.21475734565863322,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.08407615274568064,0.0,1.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.05425332893144051,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.08341586882359414,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.4642857142857143,0.43039506988004844,1.0,0.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.268460437988335,0.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.4620886981402003,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.21563772422141522,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.09271486739297898,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11813579839330911,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06652360515021459,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023935292175635527,1.0,0.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.2396280400572246,1.0,0.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.125,0.10410476504897105,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.03851656212171234,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.21428571428571427,0.050896885660834154,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.06866952789699571,0.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09541102674149884,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.5714285714285714,0.19555408825795093,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.09557609772202047,1.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.21398701441619897,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.8928571428571428,0.5054473423572136,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.16176956091119182,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.1607142857142857,0.034774953229888855,1.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13810938703642567,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.10845163420270716,0.0,0.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.020688896225376913,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,1.0,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0532078793881369,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.12798503356443272,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.13189171343677783,1.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.4107142857142857,0.12765489160338947,0.0,0.0,0.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1594585671838891,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022889842632331906,0.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.051447122262572906,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.3138549576317817,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.7321428571428571,0.24738637614174097,1.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.0529877847474414,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.09067899196654561,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.6239683063717398,0.0,0.0,0.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.0,0.01931330472103004,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.05474854187300539,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.30218994167492025,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.17857142857142858,0.06382744580169472,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.035765379113018594,1.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0061626499394739735,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.09513590844062947,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.17051832287883792,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.030097942115109497,0.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.014746340926598437,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1129635743369649,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.1354682513480797,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06734896005282272,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.08638714647298339,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09227467811158797,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06509298998569385,0.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.03895675140310334,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.19725982172334106,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.028227137669197756,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.15010454495433037,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.05727963024100363,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.0233300319137229,1.0,0.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+1.0,0.3453835149114119,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.03994717728623308,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.20353251898316277,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.043908880818752064,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13315725762077696,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.23841751953339935,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.11692527786948388,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.056784417299438755,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.19296797622977882,0.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.2195994277539342,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3482447452404534,1.0,1.0,1.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.10294926818531969,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.18399911962143722,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.24408495653130846,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.2424892703862661,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.17857142857142858,0.48024650599757895,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.04429404643996919,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.15412127214702323,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.17558049961483438,0.0,0.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.26961593485198637,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.09425552987784747,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.18377902498074172,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3076923076923077,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06707384175195334,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.12947067238912735,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.18713546825134808,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.15951359084406297,1.0,0.0,1.0,0,1,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.21239132827115662,1.0,0.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.06531308462638935,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.062011665015956854,0.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,1,0
+0.08928571428571427,0.06448772972378122,1.0,1.0,0.0,0,1,0,0,0,0,0,1,1,0,0,0
+0.4642857142857143,0.18553978210630573,0.0,1.0,1.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.17029822823814242,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.020633872565203038,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.040332342907450205,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.15081985253659075,0.0,0.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.049466270496313414,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.001430615164520744,1.0,0.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06168152305491362,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.029987894794761747,0.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.056784417299438755,0.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0064927919005172245,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.055463849455265765,1.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.10652580609662154,0.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.14223616154946628,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.18609001870804448,0.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.05838010344448112,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11543963904478925,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010069329811819083,0.0,0.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.20408275558490152,0.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.13101133487399583,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.12077693408165512,1.0,1.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.10714285714285712,0.04335864421701331,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.473643666776714,1.0,1.0,1.0,0,0,0,1,0,0,0,1,0,0,0,1
+0.3571428571428571,0.05639925167822163,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3476394849785408,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06019588423021899,1.0,1.0,0.0,0,0,0,1,0,0,1,0,0,0,1,0
+0.4107142857142857,0.3144602178936943,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.020248706943985912,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.026246285902938263,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.19642857142857142,0.06569825024760646,1.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.18542973478595798,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.10955210740618465,1.0,1.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+1.0,0.799603829646748,0.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.09750192582810609,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.019753494002421042,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15247056234180698,1.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.2272477165181028,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.05827005612413337,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14185099592824915,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08501155496863651,0.0,1.0,1.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.07059535600308132,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.26702982282381427,1.0,1.0,1.0,0,1,0,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.5707604269836029,1.0,1.0,1.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.06250687795752173,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23885770881479035,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17326950588753165,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.03571428571428571,0.04385385715857819,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.4091009133927589,0.0,1.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-val.csv
new file mode 100644
index 0000000..d26a4e0
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW-val.csv
@@ -0,0 +1,301 @@
+month,credit_amount,sex,age,credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.24999999999999997,0.07169582920655881,1.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.06107626279300099,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.02806206668867613,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14575767580059426,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.7321428571428571,0.15302079894354573,1.0,0.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.24551557169582922,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.090073731704633,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11131286453174864,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.035105095190932106,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.07538241443820842,0.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.1352481567073842,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03389457466710685,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.06404754044239022,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05849015076482887,1.0,0.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12809508088478047,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.1825685044569165,0.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.06289204357873886,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10267414988445031,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09695168922636734,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09298998569384835,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.05904038736656762,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.5714285714285714,0.09502586112028172,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.047210300429184546,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.05777484318256851,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.0505667436997909,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.16501595686145043,1.0,1.0,1.0,0,0,0,1,0,0,0,1,1,0,0,0
+0.625,0.1275998679432156,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.07147573456586331,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.024760647078243648,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.32067789149334214,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05260261912622427,0.0,0.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.060801144492131615,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15010454495433037,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07015516672169032,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.037526136238582586,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.12534389787608674,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.15758776273797734,0.0,1.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.09122922856828436,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.05447342357213601,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02569604930119952,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.0695499064597777,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.08688235941454825,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02145922746781116,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39517992736876856,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.0710905689446462,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5160669087707714,1.0,0.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.00968416419060196,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.24061846594035435,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.45234951028942444,0.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.1904368878617806,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.6877407285132606,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.6785714285714286,0.4199405744470122,1.0,1.0,1.0,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.5444040937603168,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05480356553317926,1.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1840541432816111,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.18526466380543635,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.034995047870584356,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07131066358534169,1.0,0.0,0.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.13002090899086607,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04682513480796743,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1513700891383295,0.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.14285714285714285,0.1116430064927919,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.10714285714285712,0.06993507208099482,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.04357873885770881,0.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3538571585781886,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.5714285714285714,0.10426983602949268,1.0,0.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2512930560140861,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.7797402883239792,1.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02426543413667877,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.05441839991196214,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.3035714285714286,0.1384945526576428,1.0,1.0,0.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.33069219764498736,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,1,0,0
+0.24999999999999997,0.027511830086937385,0.0,1.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06894464619786508,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.125,0.04974138879718279,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.16011885110597557,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.10311433916584131,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.02839220864971938,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3685484758446132,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.0482007263123143,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.24999999999999997,0.05315285572796302,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10806646858149004,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.20864971937933313,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.04418399911962144,1.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14559260482007264,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.327335754374381,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.41311764058545175,0.0,1.0,1.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.06173654671508748,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.33564432706063607,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.2304941124683614,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.8485748872014967,1.0,0.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.030483107736326624,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3035714285714286,0.07439198855507868,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.3421371189611533,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06729393639264884,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.2851876306811929,0.0,1.0,1.0,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.05436337625178826,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.5178571428571429,0.13832948167712117,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09199955981071861,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1522504677011115,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.052657642786398146,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.21428571428571427,0.13068119291295258,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.023880268515461652,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.08928571428571427,0.07813359744690217,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.02321998459337515,0.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.12820512820512822,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.3336634752943766,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1641906019588423,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.10714285714285712,0.3883569935072081,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.1607142857142857,0.10184879498184218,0.0,0.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.32898646417959726,0.0,0.0,0.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.15219544404093763,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.014966435567293938,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0926598437328051,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.1376141740948608,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06943985913942995,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.12396830637173985,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.05705953560030813,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.6679322108506658,1.0,1.0,0.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.08928571428571427,0.05981071860900186,1.0,1.0,0.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.12061186310113349,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.06129635743369649,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11742049081104876,1.0,1.0,1.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.1402553097832068,1.0,1.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06575327390778034,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.08215032463959503,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.06036095521074061,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.25420931000330144,1.0,1.0,0.0,0,0,1,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.0817651590183779,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.027731924727632886,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.18075272367117862,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.35721360184879497,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.37839771101573677,0.0,0.0,0.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.030593155056674374,0.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14537251017937713,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,1,0
+0.5714285714285714,0.30538131396500495,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15709254979641246,1.0,0.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06250687795752173,1.0,1.0,1.0,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.11852096401452623,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19962583911081766,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15637724221415208,1.0,0.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.4285714285714286,0.1371739848134698,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.07285132607021019,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0872125013755915,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13579839330912294,0.0,0.0,0.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.2527786948387807,1.0,1.0,0.0,1,0,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.052217453505007144,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.17857142857142858,0.20485308682733577,1.0,0.0,1.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.1336524705623418,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.10360955210740619,0.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.05397821063057115,0.0,1.0,1.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.14427203697589966,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.11978650819852536,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+1.0,0.3877517332452955,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.4642857142857143,0.40420380763728403,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.625,0.7664245625619016,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06883459887751733,0.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.19775503466490593,1.0,1.0,0.0,0,0,0,1,0,0,0,1,0,1,0,0
+0.6785714285714286,0.20864971937933313,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,0.0,1.0,1.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.41619896555518876,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.03626059205458347,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.34158688235941453,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11962143721800374,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13965004952129417,0.0,0.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024595576097722022,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.12248266754704522,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0706503796632552,1.0,0.0,0.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.04500935402222955,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.18284362275778587,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.05782986684274238,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3805436337625179,1.0,1.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.09843732805106195,1.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.12286783316826234,1.0,0.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18526466380543635,1.0,0.0,1.0,0,0,0,0,1,1,0,0,0,0,1,0
+0.3571428571428571,0.09948277759436558,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.08281060856168151,0.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.07202597116760207,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16017387476614944,1.0,1.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.5486959392538792,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.062231759656652355,0.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.10955210740618465,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12589413447782546,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.4300099042588313,0.0,0.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.33927588863211183,0.0,1.0,0.0,0,0,0,0,1,1,0,0,0,1,0,0
+0.3571428571428571,0.24683613954000222,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.037416088918234836,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.1463629360625069,1.0,1.0,1.0,0,0,0,1,0,0,0,1,1,0,0,0
+0.24999999999999997,0.11158798283261802,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2318146803125344,1.0,1.0,0.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11797072741278751,0.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1103224386486189,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.625,0.4587872785297678,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3022449653350941,0.0,0.0,1.0,0,1,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.05227247716518102,0.0,0.0,1.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08897325850115549,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.0335644327060636,0.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.04258831297457907,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.36238582590513924,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.07868383404864092,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06399251678221635,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.03615054473423572,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05188731154396391,0.0,1.0,0.0,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1280400572246066,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2047980631671619,0.0,0.0,1.0,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1930229998899527,0.0,0.0,1.0,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.11389897655992078,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.12176735996478487,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06014086057004511,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,1,0
+0.4285714285714286,0.4167492021569275,1.0,1.0,1.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.16644657202597118,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.16474083856058105,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.4107142857142857,0.46016287003411466,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.08759766699680863,0.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.12991086167051832,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.08928571428571427,0.09794211510949709,1.0,0.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.49669858038956755,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.045449543303620554,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07131066358534169,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09040387366567623,0.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10404974138879718,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.425332893144052,1.0,1.0,0.0,0,1,0,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.055793991416309016,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.24999999999999997,0.10999229668757565,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.324254429404644,0.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.1842742379223066,0.0,1.0,1.0,0,0,0,1,0,1,0,0,1,0,0,0
+0.19642857142857142,0.07081545064377683,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.590073731704633,0.0,1.0,0.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.2226807527236712,1.0,1.0,0.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.05172224056344228,0.0,1.0,0.0,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.056344228018047754,0.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.11863101133487398,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33856058104985143,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.05705953560030813,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.3138549576317817,0.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.09392538791680423,0.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07164080554638494,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.03780125453945196,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09805216242984482,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.5846263893474194,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.2659843732805106,0.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.2712116209970287,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.044018928139099814,1.0,1.0,1.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2015516672169033,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.15401122482667548,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.44486629250577747,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.08847804555959062,0.0,0.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.25272367117860683,1.0,1.0,0.0,0,0,0,0,1,1,0,0,1,0,0,0
+0.24999999999999997,0.19423352041377795,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09783206778914934,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11191812479366127,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1363486299108617,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.4107142857142857,0.27748431825685044,1.0,1.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+1.0,0.4900957411687025,1.0,1.0,1.0,0,0,1,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.09315505667436998,0.0,0.0,0.0,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.148949048090679,1.0,1.0,1.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.11769560911191813,0.0,1.0,1.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08979861340376362,0.0,0.0,0.0,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08060966215472654,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.33459887751733247,1.0,1.0,0.0,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.0586552217453505,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.8058765269065697,1.0,1.0,0.0,0,0,0,0,1,0,1,0,1,0,0,0
+0.4642857142857143,0.15555188731154398,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.0,0.07120061626499395,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.041652910751623196,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.10008803785627819,1.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.7002310993727302,0.0,1.0,0.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.1570375261362386,1.0,1.0,1.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.5714285714285714,0.20578848905029165,0.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.17877187190491914,0.0,1.0,0.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.048750962914053037,1.0,1.0,0.0,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.007813359744690218,0.0,0.0,1.0,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.048475844613183675,0.0,1.0,0.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,1,0,0
+0.6785714285714286,0.17299438758666227,1.0,1.0,1.0,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.3713546825134808,1.0,1.0,0.0,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.6464179597226807,1.0,1.0,0.0,0,1,0,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.14201606690877078,0.0,1.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09436557719819522,0.0,1.0,1.0,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07411687025420931,1.0,1.0,1.0,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.09975789589523495,1.0,1.0,1.0,0,1,0,0,0,1,0,0,0,1,0,0
+0.3928571428571428,0.42456256190161773,1.0,1.0,1.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.4034334763948498,1.0,1.0,1.0,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.24369979091009136,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.30543633762517886,1.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.1990756025090789,0.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.11775063277209201,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.3631561571475735,1.0,1.0,0.0,1,0,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.0675690546935182,1.0,0.0,0.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10069329811819082,1.0,0.0,1.0,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.03747111257840871,0.0,1.0,0.0,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05469351821283151,0.0,1.0,1.0,0,0,0,0,1,1,0,0,0,0,0,1
+0.10714285714285712,0.10355452844723231,0.0,1.0,1.0,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0935402222955871,1.0,1.0,1.0,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,1.0,1.0,1.0,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.06597336854847584,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.07329151535160118,1.0,1.0,1.0,0,0,0,0,1,0,1,0,0,0,0,1
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-test.csv
new file mode 100644
index 0000000..5e83fb1
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-test.csv
@@ -0,0 +1,201 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.30802244965335096,1.0,0.0,1.0,0.31967729971945585,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11290855067679102,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.04693518212831517,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0704302850225597,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.08093980411576977,1.0,1.0,1.0,0.4768786443351435,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1389347419390338,1.0,1.0,0.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08974358974358974,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.15291075162319798,1.0,1.0,1.0,0.9318352763505735,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.07741828986464179,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.11571475734565863,1.0,1.0,1.0,0.5855097720668199,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.13255199735886433,1.0,1.0,1.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07384175195333993,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09216463079124024,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.594475624518543,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09447562451854297,0.0,0.0,1.0,0.3676314741249905,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.13321228128095083,0.0,1.0,0.0,0.2581995981513079,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.17497523935292178,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.23236491691427316,1.0,1.0,1.0,0.48667382453412633,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.24347969626939586,0.0,0.0,1.0,0.8290700279000807,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.041212721470232194,1.0,0.0,1.0,0.8205202256524424,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.5893584241223726,1.0,1.0,0.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.05876526906569825,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.019753494002421042,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.1509298998569385,1.0,1.0,1.0,0.8769881149919502,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.847529437658193,1.0,0.0,1.0,0.5547639995361563,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17156377242214152,1.0,0.0,1.0,0.8695372347753674,0,0,0,0,1,1,0,0,0,0,0,1
+0.7857142857142857,0.24969736986904373,1.0,1.0,1.0,0.21980151898449196,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10982722570705403,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08237041928029053,1.0,0.0,1.0,0.2729209608367981,0,1,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.2666446572025971,0.0,1.0,0.0,0.22634832395437568,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.034885000550236606,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.32023770221195114,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06228678331682623,1.0,1.0,1.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.02767690106745901,0.0,1.0,1.0,0.6295469835412176,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.13464289644547156,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.01870804445911742,1.0,1.0,1.0,0.7623117828116139,0,1,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.09183448883019699,1.0,1.0,0.0,0.539932226665866,0,1,0,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07686805326290305,1.0,1.0,0.0,0.3692311661992841,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3090128755364807,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3132496973698691,0.0,1.0,0.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.026246285902938263,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14097061736546715,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,1,0,0,1,0,0,0
+0.7857142857142857,0.22328601298558381,0.0,0.0,0.0,0.0810126268631418,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.3266754704522945,1.0,0.0,1.0,0.26441754960248154,0,0,0,1,0,1,0,0,1,0,0,0
+0.24999999999999997,0.05749972488169913,0.0,1.0,1.0,0.5545926067069147,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3004842082095301,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.32243864861890614,1.0,1.0,1.0,0.6234592356361524,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.44767249917464513,1.0,1.0,0.0,0.591004897644727,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.09414548255749973,0.0,0.0,1.0,0.43705682731209555,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.04996148343787829,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06382744580169472,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5178571428571429,0.385330692197645,1.0,1.0,1.0,0.8654407578963966,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.023825244855287777,0.0,1.0,1.0,0.681321561513423,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06399251678221635,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.2817761637504127,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.2217453505007153,0.0,1.0,0.0,0.383421164823506,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06096621547265324,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.0292175635523275,1.0,1.0,1.0,0.6696986103919891,0,0,0,0,1,0,0,1,0,0,1,0
+0.0,0.17222405634422802,0.0,1.0,1.0,0.9329250944686631,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.045779685264663805,1.0,1.0,1.0,0.7387979215486421,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.11483437878287663,1.0,1.0,1.0,0.47720039923434754,0,0,0,0,1,0,1,0,0,1,0,0
+0.053571428571428575,0.03279410146362936,1.0,1.0,1.0,0.9178375942401232,0,0,0,0,1,0,0,1,0,0,0,1
+0.7857142857142857,0.45322988885220644,0.0,0.0,1.0,0.20756498401823725,0,0,0,1,0,0,1,0,0,1,0,0
+0.19642857142857142,0.09106415758776273,1.0,1.0,1.0,0.6955811084767616,0,0,0,0,1,1,0,0,0,0,1,0
+0.14285714285714285,0.08550676791020138,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.022119511389897654,0.0,0.0,1.0,0.2848197012851972,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.059480576647958625,1.0,0.0,0.0,0.5995907180616916,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.15566193463189174,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1340926598437328,1.0,1.0,1.0,0.7377669250251128,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05942555298778475,1.0,1.0,1.0,0.7215189623993398,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05051172003961703,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.17585561791570375,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.052217453505007144,0.0,1.0,1.0,0.6908920489455583,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02701661714537251,0.0,0.0,0.0,0.2844193852648953,1,0,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.39198855507868385,1.0,1.0,1.0,0.6594326055741871,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10559040387366568,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.05986574226917574,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33944095961263343,0.0,1.0,1.0,0.22408792958769466,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.039892153626059204,1.0,0.0,1.0,0.311205543841052,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13392758886321118,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23528117090348852,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.038571585781886214,0.0,1.0,0.0,0.6155200677110325,0,1,0,0,0,1,0,0,0,1,0,0
+0.8928571428571428,0.8635963464289644,1.0,1.0,0.0,0.3004289224837474,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.18190822053483,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.0,0.06630351050951909,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09271486739297898,1.0,1.0,0.0,0.6014238410536501,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.1581379993397161,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.07114559260482008,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14933421371189612,1.0,1.0,1.0,0.3103438400976657,0,0,0,0,1,0,0,1,1,0,0,0
+0.24999999999999997,0.06558820292725871,1.0,1.0,0.0,0.41602094847665755,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.7431495543083525,1.0,1.0,1.0,0.5481571705989746,0,0,0,0,1,0,0,1,0,0,0,1
+0.2857142857142857,0.3245295477055134,1.0,1.0,1.0,0.6408270671508309,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.3336084516342027,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.41900517222405637,1.0,1.0,0.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.05760977220204688,1.0,0.0,1.0,0.5621939327570208,0,0,1,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.121712336304611,1.0,0.0,0.0,0.2849263027670955,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3582590513920986,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,1,0,0,0,0,0,1
+0.5714285714285714,0.5129305601408606,1.0,1.0,0.0,0.8216427748498599,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39391438318476946,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.36519203257400684,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18548475844613185,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.06063607351160999,0.0,0.0,1.0,0.3382029230298863,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06767910201386595,1.0,1.0,1.0,0.650463907744803,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.042808407615274574,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.035875426433366345,1.0,0.0,0.0,0.43311856998291953,0,0,0,1,0,1,0,0,1,0,0,0
+0.10714285714285712,0.1136238582590514,1.0,1.0,1.0,0.5291896437274354,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.09519093210080334,0.0,0.0,0.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.07169582920655881,0.0,0.0,1.0,0.41718500620577564,0,0,0,1,0,1,0,0,0,1,0,0
+0.10714285714285712,0.05447342357213601,0.0,1.0,0.0,0.5654556515140055,0,0,0,0,1,0,0,1,0,0,1,0
+0.14285714285714285,0.05353802134918015,1.0,1.0,0.0,0.7138311294496358,0,0,0,0,1,1,0,0,0,1,0,0
+0.3035714285714286,0.12963574336964895,0.0,1.0,1.0,0.31273337858185585,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06377242214152085,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.07252118410916694,0.0,0.0,1.0,0.7569602202607998,0,0,1,0,0,1,0,0,0,0,0,1
+0.01785714285714285,0.1759656652360515,1.0,1.0,1.0,0.9540917374927895,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1992956971497744,0.0,1.0,1.0,0.5541068942854244,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.116430064927919,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4087157477715418,0.0,1.0,1.0,0.7844990820387668,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5033564432706064,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.5714285714285714,0.5126004181798173,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.0928799383735006,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10448993067018818,1.0,1.0,0.0,0.38048671184038707,0,0,1,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.350170573346539,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.14355672939363928,1.0,1.0,1.0,0.9262782795387031,1,0,0,0,0,0,1,0,0,0,0,1
+0.2857142857142857,0.13090128755364808,1.0,1.0,1.0,0.5956378735010521,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.32761087267525035,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.18851105975569496,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.19642857142857142,0.114229118520964,1.0,1.0,1.0,0.6449158361571704,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.17244415098492352,1.0,1.0,0.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.1546164850885881,1.0,1.0,0.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05771981952239463,1.0,1.0,1.0,0.7333540005028889,0,0,1,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06426763508308572,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.04104765048971058,0.0,0.0,1.0,0.8565716032957927,0,0,0,1,0,0,1,0,0,0,0,1
+0.2857142857142857,0.1780015406624849,1.0,1.0,1.0,0.9000458966532298,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17211400902388027,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.05359304500935402,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.02723671178606801,0.0,1.0,0.0,0.6088670652522917,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08858809287993837,1.0,1.0,1.0,0.7134293371692706,1,0,0,0,0,0,1,0,0,1,0,0
+0.2857142857142857,0.2213051612193243,0.0,0.0,1.0,0.22169998970541557,0,0,0,0,1,0,1,0,1,0,0,0
+0.053571428571428575,0.11912622427643886,1.0,1.0,1.0,0.7174552542410282,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.19406844943325632,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.10625068779575217,1.0,1.0,1.0,0.539932226665866,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.239022779795312,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.028007043028502255,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.23550126554418402,0.0,1.0,1.0,0.8239601442576517,0,0,0,1,0,1,0,0,0,0,0,1
+0.10714285714285712,0.09210960713106636,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17591064157587763,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.7580609662154726,1.0,1.0,0.0,0.29141195522842805,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14383184769450866,1.0,1.0,0.0,0.624582242671011,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05529877847474414,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.22669747991636405,1.0,1.0,0.0,0.26590645487362585,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.03719599427753935,0.0,1.0,1.0,0.8896713869108571,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.030373060415978873,1.0,1.0,0.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.04632992186640256,0.0,1.0,1.0,0.5692869095347953,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07378672829316606,1.0,1.0,1.0,0.9140325439012884,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.23131946737096953,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023770221195113902,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.3362495873225487,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.055463849455265765,0.0,0.0,1.0,0.8290700279000807,0,1,0,0,0,1,0,0,0,0,0,1
+0.4107142857142857,0.12490370859469573,1.0,0.0,0.0,0.39742586223642357,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.00484208209530098,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.2857142857142857,0.37454605480356556,1.0,1.0,1.0,0.6871883067965017,0,0,0,1,0,1,0,0,0,1,0,0
+0.2857142857142857,0.10922196544514141,1.0,1.0,0.0,0.3790392890575462,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,0.0,1.0,1.0,0.6244595378782976,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.6095521074061846,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05496863651370089,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10278419720479806,1.0,1.0,1.0,0.7001676841419655,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05694948827996038,0.0,1.0,0.0,0.3307077506353231,0,0,0,1,0,1,0,0,1,0,0,0
+0.7857142857142857,0.18311874105865522,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.14295146913172666,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17932210850665786,0.0,0.0,1.0,0.276086333823588,0,0,0,1,0,0,1,0,1,0,0,0
+0.07142857142857142,0.05029162539892153,1.0,1.0,1.0,0.4679701801537487,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.2050181578078574,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.18498954550456698,0.0,1.0,1.0,0.5545926067069147,0,0,0,0,1,0,1,0,0,1,0,0
+1.2142857142857142,0.2941014636293606,1.0,0.0,0.0,0.09102457054520846,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025255860019808517,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.43903378452734676,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.17409486079013978,1.0,1.0,0.0,0.4148665750517137,0,1,0,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.6802024870694399,1.0,1.0,0.0,0.3516352721472709,0,0,1,0,0,0,0,1,0,1,0,0
+0.03571428571428571,0.08253549026081215,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.15555188731154398,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.2791900517222406,0.0,1.0,1.0,0.8368159662459218,0,0,0,0,1,0,0,1,0,0,0,1
+0.0,0.06844943325630021,1.0,1.0,1.0,0.7836588733899705,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.12809508088478047,0.0,1.0,1.0,0.32271911415921684,1,0,0,0,0,0,0,1,1,0,0,0
+0.053571428571428575,0.11439418950148562,0.0,1.0,1.0,0.6637429098283977,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.022229558710245404,1.0,1.0,0.0,0.4917873581105606,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.4337515131506548,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19880048420820953,1.0,1.0,1.0,0.8189994124469444,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.39738087377572356,0.0,0.0,1.0,0.48957791029126796,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.07180587652690656,1.0,1.0,0.0,0.9005973681731101,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.02971277649389237,0.0,1.0,1.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.27423792230659183,1.0,1.0,0.0,0.5394438034973912,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.11863101133487398,0.0,0.0,0.0,0.18745666815478274,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1496093320127655,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.3035714285714286,0.6839440959612633,1.0,1.0,0.0,0.8613899174477632,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.045779685264663805,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.20166171453725104,1.0,1.0,0.0,0.5235081475203799,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.19709475074281943,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09232970177176185,1.0,1.0,0.0,0.6598740118142182,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.5286123032904149,1.0,1.0,1.0,0.5291642625196091,0,0,0,1,0,0,1,0,0,1,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-train.csv
new file mode 100644
index 0000000..1627cd1
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-train.csv
@@ -0,0 +1,501 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.20408275558490152,1.0,1.0,1.0,0.2740485661497681,1,0,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.18306371739848135,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.12462859029382635,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09601628700341147,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+1.0,0.5566193463189171,1.0,1.0,1.0,0.6276600176708876,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.39303400462198745,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.008748761967646089,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.12418840101243535,1.0,1.0,1.0,0.9112945990675128,0,0,0,0,1,0,0,1,0,0,0,1
+0.4642857142857143,0.25096291405304283,0.0,0.0,1.0,0.737517899393192,0,0,0,1,0,1,0,0,0,0,0,1
+0.7857142857142857,0.3471992956971498,1.0,0.0,0.0,0.28983662188342285,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.48332783096731596,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1816881258941345,1.0,1.0,0.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.2857142857142857,0.3421371189611533,1.0,1.0,1.0,0.6403744137297571,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1250137559150435,1.0,1.0,1.0,0.32719918796985015,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.14141080664685815,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.029327610872675252,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.19329811819082207,1.0,0.0,1.0,0.34386083170097576,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.39809618135798397,0.0,1.0,0.0,0.7914015454381981,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06344228018047761,0.0,1.0,1.0,0.3257054908520209,0,0,1,0,0,1,0,0,1,0,0,0
+0.125,0.24876196764608782,1.0,1.0,1.0,0.7620595412443067,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.20743919885550788,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.052327500825354895,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.28232640035215145,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.28947947617475517,1.0,0.0,1.0,0.7905497651969635,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.04666006382744581,0.0,1.0,1.0,0.862481750443303,1,0,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.32871134587872786,1.0,1.0,0.0,0.28436974428843736,0,0,0,0,1,0,0,1,0,1,0,0
+0.125,0.06289204357873886,0.0,1.0,1.0,0.8939030026966095,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.09183448883019699,1.0,1.0,1.0,0.7769155728885463,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.2638384505337295,0.0,1.0,1.0,0.8769316531253295,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.3136898866512601,0.0,0.0,0.0,0.1493953955906259,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.026466380543633764,1.0,1.0,1.0,0.5499994958063575,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.02085396720589854,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08572686255089688,0.0,0.0,0.0,0.8228490355418768,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.05161219324309453,0.0,1.0,0.0,0.523377616709452,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.16424562561901618,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.6223175965665235,0.0,0.0,0.0,0.3243121077691882,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.04974138879718279,1.0,1.0,1.0,0.7387979215486421,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.19808517662594918,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11604489930670188,1.0,1.0,0.0,0.3842577810457014,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07059535600308132,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.21789369428854408,1.0,1.0,0.0,0.4390666360957615,0,0,1,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.19847034224716628,1.0,1.0,1.0,0.2864857137967765,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06289204357873886,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08528667326950588,1.0,0.0,1.0,0.8291441697520612,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.1101023440079234,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03224386486189061,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06025090789039286,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.2500825354902608,0.0,1.0,1.0,0.3694096935983138,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025200836359634642,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.05716958292065588,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08880818752063387,0.0,1.0,0.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.5342797402883239,0.0,1.0,0.0,0.26164945103668286,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.27797953119841534,1.0,1.0,1.0,0.36651873073537256,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.08924837680202487,1.0,1.0,1.0,0.48838112309751835,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.12044679212061186,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06261692527786948,0.0,1.0,1.0,0.8832307883744815,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.6558270056124132,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.13271706833938596,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.1607142857142857,0.06377242214152085,0.0,1.0,1.0,0.8854921566472022,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1866952789699571,0.0,1.0,1.0,0.3691093043872932,0,0,0,0,1,0,1,0,1,0,0,0
+0.7857142857142857,0.358093980411577,0.0,1.0,0.0,0.10812898709422081,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.0880928799383735,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.17420490811048753,1.0,1.0,1.0,0.3128458574189485,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.2678001540662485,1.0,1.0,1.0,0.8921185284956621,0,0,0,1,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07411687025420931,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.06228678331682623,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.05397821063057115,1.0,1.0,1.0,0.9319236160108605,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13101133487399583,0.0,1.0,0.0,0.5382443052921533,0,0,1,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.10454495433036205,0.0,0.0,0.0,0.27183024502308256,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.1474634092659844,0.0,1.0,1.0,0.6634322478338307,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.3582590513920986,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.13728403213381754,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11417409486079015,1.0,1.0,1.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05249257180587652,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.04396390447892594,0.0,0.0,1.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.11235831407505227,1.0,1.0,0.0,0.9262782795387031,1,0,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.06597336854847584,0.0,1.0,1.0,0.9185698239945864,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.39952679652250467,1.0,1.0,0.0,0.36651873073537256,0,0,0,0,1,0,0,1,1,0,0,0
+0.08928571428571427,0.03675580499614835,0.0,1.0,0.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.1153295917244415,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07444701221525256,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.125,0.2011114779355123,1.0,1.0,1.0,0.4708753344299414,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0482007263123143,0.0,0.0,1.0,0.7904631079537381,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.14707824364476726,0.0,1.0,1.0,0.4849768887446002,0,0,1,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1620446792120612,0.0,1.0,1.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.24573566633652472,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.7871134587872785,1.0,0.0,0.0,0.6577144367789873,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2884890502916254,1.0,1.0,1.0,0.21396727904669746,0,0,0,0,1,0,0,1,1,0,0,0
+0.3571428571428571,0.0734565863321228,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.18377902498074172,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.1380543633762518,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10300429184549356,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.09640145262462858,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.09084406294706723,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.06839440959612633,0.0,1.0,1.0,0.8898642220971337,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.052767690106745896,1.0,1.0,0.0,0.7619553600617908,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20518322878837902,1.0,1.0,1.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.4110817651590184,0.0,1.0,1.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1870804445911742,0.0,0.0,1.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.08968856608341587,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09304500935402223,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.034334763948497854,0.0,0.0,1.0,0.8105188297717469,0,1,0,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.2718168812589414,1.0,1.0,1.0,0.8704016314829519,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.5627269725982172,1.0,1.0,1.0,0.7682673905629588,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.22565203037306042,0.0,1.0,1.0,0.4443019547874373,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06195664135578298,0.0,1.0,1.0,0.8575078279322482,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.1786068009243975,1.0,1.0,1.0,0.49210970933040776,0,0,1,0,0,0,1,0,0,1,0,0
+0.6428571428571428,0.3162759986794322,1.0,1.0,0.0,0.20892794417015212,0,0,1,0,0,0,1,0,1,0,0,0
+0.4285714285714286,0.20666886761307363,1.0,1.0,0.0,0.30391227130738524,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.1701331572576208,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.13051612193243095,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11742049081104876,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07752833718498954,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1781666116430065,0.0,0.0,1.0,0.7832044401604136,0,0,0,1,0,0,1,0,0,0,0,1
+1.0,0.39440959612633436,1.0,1.0,1.0,0.21615402523359528,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10938703642566304,0.0,0.0,1.0,0.7569602202607998,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.059590623968306375,1.0,1.0,0.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.3214285714285714,0.05683944095961263,0.0,0.0,1.0,0.7970917912232778,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.17981732144822274,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.018432926158248045,1.0,0.0,1.0,0.5528336477984951,0,0,1,0,0,1,0,0,0,1,0,0
+0.625,0.25767580059425554,1.0,0.0,0.0,0.3730439023791205,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.056564322658743255,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.08352591614394189,1.0,0.0,1.0,0.8618695563470995,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.17051832287883792,1.0,1.0,0.0,0.3103438400976657,0,1,0,0,0,0,0,1,1,0,0,0
+0.08928571428571427,0.0627819962583911,1.0,1.0,1.0,0.7001676841419655,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.17453505007153078,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10432485968966655,0.0,0.0,1.0,0.8406172417730042,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.06988004842082095,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3627159678661825,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09909761197314845,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15109497083746012,1.0,0.0,0.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18763068119291296,0.0,1.0,1.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06558820292725871,0.0,0.0,0.0,0.2287663808190244,0,0,0,1,0,1,0,0,1,0,0,0
+0.08928571428571427,0.04886101023440079,1.0,0.0,1.0,0.36745330533116055,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.3037856278199626,0.0,0.0,1.0,0.737517899393192,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.12930560140860572,1.0,1.0,0.0,0.3232634558617386,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.19252778694838782,0.0,1.0,1.0,0.501464322292528,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.15456146142841423,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022669747991636405,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09298998569384835,1.0,1.0,1.0,0.36651873073537256,0,1,0,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.1140640475404424,1.0,0.0,1.0,0.4028715067363472,0,0,0,1,0,1,0,0,0,1,0,0
+0.7857142857142857,0.6577528337184989,1.0,1.0,1.0,0.3131231838824953,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.28925938153405967,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.24999999999999997,0.010069329811819083,0.0,0.0,0.0,0.7901372777915644,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.23995818201826785,0.0,1.0,0.0,0.9121910841333913,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.13821943435677342,0.0,0.0,0.0,0.49006922670215497,0,0,0,0,1,0,1,0,0,1,0,0
+0.4642857142857143,0.2213051612193243,1.0,1.0,1.0,0.8302239413120919,0,0,1,0,0,0,1,0,0,0,0,1
+0.625,0.6333223286012986,1.0,1.0,1.0,0.49778098104998286,0,0,0,1,0,1,0,0,0,1,0,0
+0.5714285714285714,0.10867172884340266,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.06734896005282272,0.0,1.0,1.0,0.6244595378782976,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06338725652030373,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.09150434686915374,1.0,1.0,1.0,0.6598740118142182,0,1,0,0,0,0,1,0,0,1,0,0
+0.17857142857142858,0.030373060415978873,1.0,1.0,1.0,0.9052897264392974,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.05617915703752613,1.0,1.0,0.0,0.5758465012701344,0,0,1,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.055463849455265765,0.0,1.0,1.0,0.45368904432201734,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.18240343347639487,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17255419830527127,1.0,1.0,0.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.1098822493672279,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.30560140860570045,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.03906679872345109,0.0,1.0,1.0,0.8976829391418685,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.044349070100143065,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3322878837900297,1.0,1.0,1.0,0.42442143235694413,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.13772422141520854,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10784637394079453,1.0,0.0,0.0,0.8715443022869169,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.5528227137669197,1.0,1.0,0.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.5714285714285714,0.7740728513260702,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.2200396170353252,1.0,1.0,0.0,0.5394438034973912,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.03736106525806096,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.039011775063277215,0.0,1.0,0.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.13931990756025092,0.0,1.0,1.0,0.8721840349758069,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.44613183668977663,0.0,1.0,1.0,0.26164945103668286,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.11285352701661715,0.0,0.0,1.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.18245845713656875,1.0,1.0,1.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.20446792120611865,0.0,1.0,0.0,0.3403565857713484,0,0,0,0,1,1,0,0,1,0,0,0
+0.03571428571428571,0.05430835259161439,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03994717728623308,0.0,0.0,0.0,0.2364918904353026,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.03659073401562672,0.0,1.0,0.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.03835149114119071,1.0,1.0,0.0,0.5813781424290714,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.14168592494772753,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.3035714285714286,0.11213821943435677,0.0,0.0,1.0,0.7688898730617176,0,1,0,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.145867723120942,0.0,1.0,1.0,0.8978106855080499,0,1,0,0,0,1,0,0,0,0,0,1
+0.7321428571428571,0.23803235391218225,1.0,1.0,1.0,0.38456298371763664,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3845053372950369,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.13733905579399142,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08215032463959503,1.0,1.0,1.0,0.6014238410536501,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.06140640475404424,0.0,0.0,0.0,0.31109338775758766,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06905469351821283,1.0,1.0,1.0,0.8533241696172105,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.005117200396170352,0.0,1.0,1.0,0.4603229620470974,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.10856168152305491,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.06850445691647408,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.11428414218113787,0.0,0.0,0.0,0.4681806327003752,0,1,0,0,0,0,1,0,0,0,1,0
+0.7321428571428571,0.6364036535710355,1.0,1.0,0.0,0.20599972812749537,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.06789919665456146,1.0,0.0,0.0,0.5528336477984951,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4844833278309673,1.0,1.0,1.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.07730824254429404,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.06619346318917134,0.0,1.0,1.0,0.8627147751385414,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.053813139650049524,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.037526136238582586,0.0,1.0,1.0,0.7170669202424681,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4311654011224827,1.0,1.0,0.0,0.47720039923434754,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.06030593155056674,1.0,1.0,1.0,0.9209417015586914,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.1437218003741609,1.0,1.0,1.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.15808297567954221,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.3938593595245956,0.0,0.0,0.0,0.1140139540448384,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055793991416309016,1.0,0.0,0.0,0.5216017218323117,0,0,1,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.07175085286673269,0.0,0.0,1.0,0.8847736592891744,0,0,0,1,0,1,0,0,0,0,0,1
+0.3035714285714286,0.18719049191152196,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20254209310003302,1.0,1.0,0.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.03571428571428571,0.061131286453174866,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.052437548145702645,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.18031253438978762,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.4184549356223176,0.0,0.0,0.0,0.594377341023918,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.17095851216022892,0.0,1.0,1.0,0.31273337858185585,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.10179377132166832,1.0,1.0,0.0,0.624582242671011,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0505667436997909,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.47854077253218885,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09276989105315285,1.0,0.0,1.0,0.3727423047912524,0,0,0,1,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1794871794871795,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3695939253879168,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,1,0,0,0,1,0,0
+0.4642857142857143,0.11461428414218112,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09843732805106195,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.49636843842852424,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.07274127874986244,0.0,1.0,1.0,0.8471418368374785,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.011224826675470454,1.0,0.0,1.0,0.6581569155182169,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.33311323869263787,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.10713106635853417,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.40601958842302194,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.05639925167822163,0.0,0.0,1.0,0.5204868269574562,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.37795752173434577,1.0,1.0,0.0,0.11967916113567616,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.6783867062837019,0.0,1.0,0.0,0.4107052500869476,0,0,0,0,1,0,0,1,0,1,0,0
+0.24999999999999997,0.1391548365797293,1.0,0.0,1.0,0.4901999941597845,0,0,1,0,0,1,0,0,0,1,0,0
+0.0,0.06894464619786508,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.07142857142857142,0.05430835259161439,0.0,0.0,0.0,0.4840535155941271,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20160669087707717,1.0,1.0,0.0,0.6970354972308322,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15472653240893586,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,1,0,0,0,0,0,1
+0.5714285714285714,0.09370529327610873,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.19709475074281943,1.0,0.0,1.0,0.4133255669397803,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.0708704743039507,0.0,1.0,1.0,0.9121910841333913,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10223396060305931,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.00979421151094971,0.0,1.0,1.0,0.4603229620470974,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.2313744910311434,1.0,1.0,0.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.1138439528997469,1.0,1.0,0.0,0.6399187311966064,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05023660173874765,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.037746230879278087,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.12231759656652359,1.0,0.0,0.0,0.3250877505238663,1,0,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07257620776934082,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.1522504677011115,0.0,1.0,1.0,0.6155200677110325,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11450423682183337,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10157367668097282,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04330362055683944,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05425332893144051,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.21464729833828547,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.18482447452404535,0.0,1.0,1.0,0.5962182584447656,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.06393749312204247,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.125,0.20298228238142402,1.0,1.0,1.0,0.4708753344299414,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.0702101903818642,1.0,1.0,1.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.3214285714285714,0.1334323759216463,1.0,1.0,1.0,0.8922523883382062,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.2184989545504567,0.0,1.0,1.0,0.383421164823506,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.14713326730494114,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08622207549246176,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.0536480686695279,1.0,1.0,1.0,0.6993390090705507,0,0,1,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08776273797733025,0.0,1.0,1.0,0.3691093043872932,0,1,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022559700671288655,1.0,0.0,1.0,0.9228647175807431,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.025255860019808517,1.0,1.0,1.0,0.7691184526170824,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.8587542643336634,1.0,1.0,1.0,0.2740485661497681,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.08732254869593925,0.0,1.0,0.0,0.46182459874957915,0,0,0,1,0,0,0,1,0,1,0,0
+0.1607142857142857,0.08512160228898426,1.0,1.0,1.0,0.4978064484847755,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010894684714427203,0.0,0.0,0.0,0.3386431143464047,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.16303510509519095,1.0,0.0,1.0,0.5380335258497126,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.12815010454495435,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05216242984483327,0.0,1.0,0.0,0.3803633753089816,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.26152745680642675,0.0,1.0,0.0,0.8206450815998698,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11461428414218112,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055683944095961266,1.0,1.0,1.0,0.9318352763505735,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.049796412457356665,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.10714285714285712,0.03543523715197534,0.0,1.0,1.0,0.9153811453083084,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.16952789699570817,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.04748541873005392,0.0,1.0,1.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02580609662154727,1.0,1.0,0.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.09981291955540883,0.0,1.0,0.0,0.49148753840738346,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.18741058655221746,1.0,1.0,1.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.3502255970067129,1.0,1.0,1.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.24999999999999997,0.16391548365797293,1.0,1.0,1.0,0.9071687984489271,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17018818091779467,1.0,1.0,1.0,0.868365131914781,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.1572576207769341,1.0,1.0,0.0,0.32719918796985015,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.011444921316165951,1.0,0.0,1.0,0.5836541255891218,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.06718388907230109,1.0,1.0,1.0,0.868365131914781,0,0,1,0,0,0,0,1,0,0,0,1
+0.125,0.05898536370639375,0.0,1.0,1.0,0.5795267454297757,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.057114559260482006,1.0,1.0,1.0,0.5396118319655506,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.025145812699460767,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.058215032463959496,1.0,1.0,1.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.15538681633102236,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.12440849565313085,0.0,0.0,1.0,0.259549729145002,1,0,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0819852536590734,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.04803565533179267,0.0,1.0,0.0,0.8696334095426655,0,0,1,0,0,1,0,0,0,0,0,1
+0.125,0.07301639705073182,0.0,0.0,1.0,0.5005400663825238,0,1,0,0,0,0,1,0,0,1,0,0
+0.7321428571428571,0.08776273797733025,1.0,0.0,0.0,0.08470200286988733,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.09783206778914934,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.4439308902828216,0.0,1.0,0.0,0.38066739707842107,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.05749972488169913,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.6431165401122483,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.08633212281280951,1.0,1.0,0.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0.4768786443351435,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06965995378012545,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.1037746230879278,1.0,0.0,1.0,0.36745330533116055,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.19599427753934193,0.0,0.0,1.0,0.8105188297717469,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.21541762958071972,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.20947507428194126,1.0,1.0,0.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.543468691537361,1.0,1.0,0.0,0.6234592356361524,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.012875536480686695,1.0,1.0,1.0,0.7979365203848136,0,0,0,1,0,0,1,0,0,1,0,0
+1.0,0.33223286012985587,1.0,1.0,0.0,0.16559346885680104,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.36436667767139874,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07054033234290745,0.0,1.0,1.0,0.8547131179732448,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10366457576758006,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04715527676901067,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,1,0,0,1,0,0,0
+0.2857142857142857,0.10795642126114229,1.0,1.0,1.0,0.47653149674170675,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.16154946627049632,0.0,0.0,1.0,0.8404307054775538,0,0,0,1,0,1,0,0,0,0,0,1
+0.7678571428571428,0.5762077693408165,0.0,1.0,1.0,0.13279434507188115,0,1,0,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.02641135688345989,1.0,1.0,1.0,0.9394762243447705,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.10030813249697369,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0176625949158138,1.0,1.0,1.0,0.4148665750517137,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.11609992296687575,1.0,1.0,1.0,0.6535565228181367,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06019588423021899,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.6785714285714286,0.3809838230439089,0.0,1.0,0.0,0.18862334019529733,0,0,0,1,0,0,1,0,1,0,0,0
+0.6785714285714286,0.2501375591504347,1.0,1.0,1.0,0.3466918762732373,0,0,0,0,1,0,0,1,0,0,1,0
+0.6785714285714286,0.20441289754594477,1.0,1.0,0.0,0.22731351394237334,0,1,0,0,0,0,1,0,1,0,0,0
+0.125,0.38395510069329813,1.0,1.0,1.0,0.915529351477635,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05315285572796302,1.0,1.0,0.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.13744910311433917,0.0,1.0,1.0,0.8532586635277519,0,0,0,0,1,0,0,1,0,0,0,1
+0.6785714285714286,0.49702872235061074,1.0,1.0,1.0,0.33770845427133345,1,0,0,0,0,0,0,1,0,1,0,0
+0.4107142857142857,0.23528117090348852,1.0,1.0,1.0,0.8704016314829519,0,1,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.2750082535490261,1.0,1.0,1.0,0.8769881149919502,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16127434796962695,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.02569604930119952,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,0.0,1.0,0.8618695563470995,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08990866072411137,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09210960713106636,0.0,1.0,0.0,0.5545926067069147,0,1,0,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.39760096841641906,1.0,1.0,1.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.14285714285714285,0.03791130185979971,1.0,1.0,0.0,0.7613436169716516,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03251898316275999,0.0,0.0,1.0,0.5419816737959281,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.21255639925167824,0.0,1.0,0.0,0.491978924064202,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.2697259821723341,1.0,0.0,1.0,0.5367070201327678,0,0,0,1,0,0,1,0,0,0,1,0
+0.5178571428571429,0.22180037416088919,0.0,0.0,0.0,0.12004837083863333,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07246616044899307,0.0,0.0,1.0,0.44241615651549826,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.004897105755474855,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.05639925167822163,0.0,0.0,0.0,0.2247881972110395,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.021404203807637284,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1127434796962694,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.086552217453505,1.0,1.0,0.0,0.3842577810457014,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.41339275888632115,1.0,1.0,0.0,0.14138446961748627,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.07400682293386156,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.007428194123473095,0.0,1.0,1.0,0.44959115198776406,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.028942445251458126,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15511169803015298,1.0,1.0,1.0,0.520597760213557,0,0,0,0,1,0,0,1,0,1,0,0
+0.08928571428571427,0.06129635743369649,1.0,1.0,1.0,0.590693107733845,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0534279740288324,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.17321448222735777,1.0,1.0,1.0,0.6007809114857595,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.21475734565863322,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.08407615274568064,0.0,1.0,0.0,0.5853827741356347,0,1,0,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.05425332893144051,0.0,0.0,1.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.08341586882359414,1.0,0.0,1.0,0.8891646790727171,0,0,1,0,0,1,0,0,0,0,0,1
+0.4642857142857143,0.43039506988004844,1.0,0.0,1.0,0.22597418414948542,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.268460437988335,0.0,1.0,0.0,0.5664245567595892,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.4620886981402003,1.0,1.0,0.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.21563772422141522,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.09271486739297898,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11813579839330911,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06652360515021459,0.0,1.0,1.0,0.8547131179732448,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023935292175635527,1.0,0.0,0.0,0.6483538884899454,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.2396280400572246,1.0,0.0,0.0,0.08995843040536927,0,0,0,0,1,0,0,1,1,0,0,0
+0.125,0.10410476504897105,1.0,1.0,1.0,0.9292175567150299,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.03851656212171234,1.0,1.0,0.0,0.4768786443351435,0,0,0,0,1,1,0,0,1,0,0,0
+0.21428571428571427,0.050896885660834154,1.0,1.0,1.0,0.602425169638959,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.06866952789699571,0.0,0.0,1.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09541102674149884,1.0,1.0,1.0,0.8533241696172105,0,0,1,0,0,0,0,1,0,0,0,1
+0.5714285714285714,0.19555408825795093,0.0,1.0,0.0,0.3259888088958532,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.09557609772202047,1.0,0.0,0.0,0.23658638755348427,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.21398701441619897,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.8928571428571428,0.5054473423572136,1.0,1.0,1.0,0.6415019923339554,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.16176956091119182,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,0,1,0,1,0,0,0
+0.1607142857142857,0.034774953229888855,1.0,0.0,1.0,0.5894933892681636,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13810938703642567,1.0,1.0,1.0,0.520597760213557,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.10845163420270716,0.0,0.0,1.0,0.7793274409005319,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.020688896225376913,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,1.0,0.0,1.0,0.0,0.22634832395437568,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0532078793881369,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.12798503356443272,1.0,1.0,1.0,0.7174552542410282,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.13189171343677783,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.4107142857142857,0.12765489160338947,0.0,0.0,0.0,0.6811557639743221,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1594585671838891,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022889842632331906,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.051447122262572906,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.3138549576317817,1.0,1.0,1.0,0.8794147982207655,0,0,0,1,0,0,1,0,0,0,0,1
+0.7321428571428571,0.24738637614174097,1.0,0.0,0.0,0.2731769653496512,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.0529877847474414,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.09067899196654561,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.6239683063717398,0.0,0.0,0.0,0.46989347577510143,0,0,1,0,0,1,0,0,0,0,0,1
+0.0,0.01931330472103004,0.0,0.0,1.0,0.8714857078947565,0,1,0,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.05474854187300539,1.0,1.0,0.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.30218994167492025,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.17857142857142858,0.06382744580169472,1.0,1.0,1.0,0.6964156448050416,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.035765379113018594,1.0,0.0,0.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0061626499394739735,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.09513590844062947,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.17051832287883792,0.0,1.0,0.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.030097942115109497,0.0,1.0,0.0,0.8898642220971337,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.014746340926598437,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1129635743369649,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.1354682513480797,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06734896005282272,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.08638714647298339,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09227467811158797,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06509298998569385,0.0,0.0,1.0,0.2247881972110395,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.03895675140310334,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.19725982172334106,0.0,1.0,1.0,0.7472556622006505,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.028227137669197756,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.15010454495433037,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.05727963024100363,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.0233300319137229,1.0,0.0,0.0,0.43311856998291953,0,0,0,1,0,0,1,0,1,0,0,0
+1.0,0.3453835149114119,1.0,1.0,1.0,0.5003985149708152,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.03994717728623308,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.20353251898316277,1.0,1.0,1.0,0.7380360665955924,0,1,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.043908880818752064,1.0,0.0,1.0,0.5734408207763663,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13315725762077696,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.23841751953339935,1.0,1.0,1.0,0.2864857137967765,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.11692527786948388,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.056784417299438755,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.19296797622977882,0.0,1.0,1.0,0.7786348012503493,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.2195994277539342,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3482447452404534,1.0,1.0,1.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.10294926818531969,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,0.0,1.0,1.0,0.8898642220971337,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.18399911962143722,0.0,1.0,1.0,0.8769316531253295,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.24408495653130846,1.0,1.0,1.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.2424892703862661,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.17857142857142858,0.48024650599757895,1.0,1.0,0.0,0.4873369967986317,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.04429404643996919,0.0,1.0,1.0,0.8627147751385414,0,1,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.15412127214702323,1.0,1.0,0.0,0.15889026804756212,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.17558049961483438,0.0,0.0,0.0,0.27183024502308256,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.26961593485198637,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.09425552987784747,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.18377902498074172,1.0,1.0,1.0,0.6594326055741871,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3076923076923077,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06707384175195334,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.12947067238912735,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.18713546825134808,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.15951359084406297,1.0,0.0,1.0,0.2126849403417187,0,1,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.21239132827115662,1.0,0.0,0.0,0.2729209608367981,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.06531308462638935,1.0,1.0,0.0,0.7769155728885463,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.062011665015956854,0.0,1.0,1.0,0.419918901878422,0,0,0,0,1,0,0,1,0,0,1,0
+0.08928571428571427,0.06448772972378122,1.0,1.0,0.0,0.45755650252851443,0,1,0,0,0,0,0,1,1,0,0,0
+0.4642857142857143,0.18553978210630573,0.0,1.0,1.0,0.23787440014014885,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.17029822823814242,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.020633872565203038,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.040332342907450205,0.0,0.0,1.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.15081985253659075,0.0,0.0,1.0,0.31136986465762495,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.049466270496313414,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.001430615164520744,1.0,0.0,1.0,0.5836541255891218,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06168152305491362,0.0,0.0,0.0,0.16589905516842904,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.029987894794761747,0.0,1.0,0.0,0.39882243364250436,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.056784417299438755,0.0,1.0,0.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0064927919005172245,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.055463849455265765,1.0,0.0,1.0,0.8491471130930849,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.10652580609662154,0.0,1.0,1.0,0.8976829391418685,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.14223616154946628,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.18609001870804448,0.0,0.0,1.0,0.16589905516842904,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.05838010344448112,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11543963904478925,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010069329811819083,0.0,0.0,0.0,0.5527042865014291,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.20408275558490152,0.0,1.0,0.0,0.3257054908520209,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.13101133487399583,0.0,0.0,1.0,0.41167927397905774,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.12077693408165512,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,1,0,0,0,0,0,1
+0.10714285714285712,0.04335864421701331,1.0,1.0,1.0,0.5805283757278361,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.473643666776714,1.0,1.0,1.0,0.7117589057670359,0,0,0,1,0,0,0,1,0,0,0,1
+0.3571428571428571,0.05639925167822163,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3476394849785408,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06019588423021899,1.0,1.0,0.0,0.6586754162607802,0,0,0,1,0,0,1,0,0,0,1,0
+0.4107142857142857,0.3144602178936943,1.0,1.0,1.0,0.5709518098205263,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.020248706943985912,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.026246285902938263,1.0,1.0,1.0,0.9078496135630285,0,0,0,0,1,0,0,1,0,0,0,1
+0.19642857142857142,0.06569825024760646,1.0,0.0,1.0,0.5690789555483984,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.18542973478595798,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.10955210740618465,1.0,1.0,1.0,0.5291896437274354,0,1,0,0,0,1,0,0,1,0,0,0
+1.0,0.799603829646748,0.0,1.0,0.0,0.13365287548774402,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.09750192582810609,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.019753494002421042,0.0,1.0,0.0,0.6244595378782976,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15247056234180698,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.2272477165181028,1.0,1.0,1.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.05827005612413337,1.0,1.0,0.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14185099592824915,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08501155496863651,0.0,1.0,1.0,0.5545926067069147,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.07059535600308132,1.0,1.0,0.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.26702982282381427,1.0,1.0,1.0,0.8456935742394528,0,1,0,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.5707604269836029,1.0,1.0,1.0,0.2592550937927965,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.06250687795752173,0.0,1.0,0.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23885770881479035,1.0,1.0,0.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17326950588753165,1.0,1.0,1.0,0.7215189623993398,0,0,0,0,1,0,1,0,0,0,1,0
+0.03571428571428571,0.04385385715857819,0.0,1.0,1.0,0.6383519380348411,0,0,1,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.4091009133927589,0.0,1.0,0.0,0.15312328184273788,0,0,0,1,0,1,0,0,1,0,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-val.csv
new file mode 100644
index 0000000..4856718
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR-val.csv
@@ -0,0 +1,301 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.24999999999999997,0.07169582920655881,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.06107626279300099,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.02806206668867613,0.0,1.0,1.0,0.6997655328413775,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14575767580059426,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.7321428571428571,0.15302079894354573,1.0,0.0,0.0,0.23681941072883386,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.24551557169582922,1.0,1.0,1.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.090073731704633,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11131286453174864,1.0,1.0,1.0,0.5291642625196091,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.035105095190932106,1.0,1.0,0.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.07538241443820842,0.0,1.0,1.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.1352481567073842,1.0,1.0,1.0,0.9173729937673434,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03389457466710685,1.0,0.0,1.0,0.4901999941597845,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.06404754044239022,1.0,1.0,1.0,0.7088870911194225,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05849015076482887,1.0,0.0,1.0,0.8955009029182879,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12809508088478047,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.1825685044569165,0.0,1.0,1.0,0.8769316531253295,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.06289204357873886,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10267414988445031,1.0,1.0,1.0,0.7333540005028889,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09695168922636734,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09298998569384835,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.05904038736656762,1.0,1.0,1.0,0.7691184526170824,0,0,0,0,1,0,1,0,0,0,1,0
+0.5714285714285714,0.09502586112028172,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.047210300429184546,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.05777484318256851,1.0,1.0,1.0,0.8887675866521803,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.0505667436997909,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.16501595686145043,1.0,1.0,1.0,0.2510304635302003,0,0,0,1,0,0,0,1,1,0,0,0
+0.625,0.1275998679432156,1.0,0.0,1.0,0.6681686332576849,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.07147573456586331,1.0,1.0,1.0,0.7387979215486421,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.024760647078243648,1.0,1.0,1.0,0.9209417015586914,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.32067789149334214,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05260261912622427,0.0,0.0,1.0,0.5214711461502933,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.060801144492131615,0.0,0.0,0.0,0.3676314741249905,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15010454495433037,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07015516672169032,0.0,1.0,1.0,0.8575078279322482,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.037526136238582586,0.0,0.0,1.0,0.8373645845436015,0,0,1,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.12534389787608674,0.0,1.0,1.0,0.46064338284484674,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.15758776273797734,0.0,1.0,0.0,0.3403565857713484,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.09122922856828436,1.0,1.0,0.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.05447342357213601,0.0,1.0,1.0,0.8898642220971337,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02569604930119952,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.0695499064597777,1.0,0.0,1.0,0.5216017218323117,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.08688235941454825,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02145922746781116,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39517992736876856,0.0,1.0,1.0,0.3259888088958532,0,0,1,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.0710905689446462,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5160669087707714,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.00968416419060196,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.24061846594035435,1.0,1.0,0.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.45234951028942444,0.0,0.0,1.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.1904368878617806,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.6877407285132606,1.0,1.0,1.0,0.7743267172453298,0,0,0,1,0,0,1,0,0,0,0,1
+0.6785714285714286,0.4199405744470122,1.0,1.0,1.0,0.20677830560523824,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.5444040937603168,0.0,0.0,1.0,0.4443440748703878,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05480356553317926,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1840541432816111,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.18526466380543635,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.034995047870584356,0.0,0.0,1.0,0.8001795737339381,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07131066358534169,1.0,0.0,0.0,0.3160895023001086,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.13002090899086607,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04682513480796743,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1513700891383295,0.0,1.0,1.0,0.6052277358503451,0,0,0,1,0,1,0,0,0,1,0,0
+0.14285714285714285,0.1116430064927919,1.0,1.0,1.0,0.8820768558276499,0,0,1,0,0,0,0,1,0,0,0,1
+0.10714285714285712,0.06993507208099482,1.0,1.0,1.0,0.6913008425849491,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.04357873885770881,0.0,1.0,0.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3538571585781886,1.0,1.0,1.0,0.24702882446640287,0,0,1,0,0,0,0,1,0,1,0,0
+0.5714285714285714,0.10426983602949268,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2512930560140861,1.0,1.0,0.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.7797402883239792,1.0,0.0,0.0,0.21486216341529946,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02426543413667877,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.05441839991196214,1.0,1.0,1.0,0.5354741988477034,0,0,1,0,0,0,0,1,0,1,0,0
+0.3035714285714286,0.1384945526576428,1.0,1.0,0.0,0.6311298802642114,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.33069219764498736,1.0,1.0,1.0,0.70490883351029,0,0,0,1,0,1,0,0,0,1,0,0
+0.24999999999999997,0.027511830086937385,0.0,1.0,0.0,0.33991531191906393,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06894464619786508,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.125,0.04974138879718279,0.0,1.0,1.0,0.8937163943553702,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.16011885110597557,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.10311433916584131,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.02839220864971938,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3685484758446132,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.0482007263123143,1.0,1.0,1.0,0.7460536409951634,0,0,0,0,1,0,1,0,0,0,1,0
+0.24999999999999997,0.05315285572796302,0.0,0.0,0.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10806646858149004,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.20864971937933313,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.04418399911962144,1.0,1.0,1.0,0.4768786443351435,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14559260482007264,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.327335754374381,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.41311764058545175,0.0,1.0,1.0,0.491978924064202,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.06173654671508748,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.33564432706063607,1.0,1.0,0.0,0.32719918796985015,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.2304941124683614,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.8485748872014967,1.0,0.0,0.0,0.21486216341529946,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.030483107736326624,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3035714285714286,0.07439198855507868,1.0,1.0,1.0,0.4148665750517137,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.3421371189611533,1.0,1.0,0.0,0.7134293371692706,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06729393639264884,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.2851876306811929,0.0,1.0,1.0,0.8298727730544322,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.05436337625178826,1.0,1.0,1.0,0.9417068119041873,1,0,0,0,0,0,1,0,0,0,0,1
+0.5178571428571429,0.13832948167712117,0.0,1.0,1.0,0.7345947491674498,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09199955981071861,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1522504677011115,1.0,1.0,1.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.052657642786398146,0.0,0.0,0.0,0.1941383022647156,0,1,0,0,0,1,0,0,1,0,0,0
+0.21428571428571427,0.13068119291295258,1.0,1.0,0.0,0.4664405275790275,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.023880268515461652,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.08928571428571427,0.07813359744690217,0.0,0.0,0.0,0.5214711461502933,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.02321998459337515,0.0,1.0,1.0,0.9120334763614909,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.12820512820512822,1.0,1.0,0.0,0.8962138068279807,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.3336634752943766,1.0,1.0,1.0,0.8837277100655102,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1641906019588423,0.0,0.0,0.0,0.1941383022647156,0,1,0,0,0,1,0,0,1,0,0,0
+0.10714285714285712,0.3883569935072081,1.0,1.0,1.0,0.6608113136649546,1,0,0,0,0,0,0,1,0,1,0,0
+0.1607142857142857,0.10184879498184218,0.0,0.0,1.0,0.4796070935730555,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.32898646417959726,0.0,0.0,0.0,0.1271934988531199,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.15219544404093763,1.0,0.0,1.0,0.5621939327570208,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.014966435567293938,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0926598437328051,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.1376141740948608,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06943985913942995,1.0,1.0,0.0,0.6449158361571704,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.12396830637173985,1.0,1.0,1.0,0.539932226665866,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.05705953560030813,0.0,1.0,1.0,0.8301501720360537,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.6679322108506658,1.0,1.0,0.0,0.3516352721472709,0,0,1,0,0,0,0,1,0,1,0,0
+0.08928571428571427,0.05981071860900186,1.0,1.0,0.0,0.7460536409951634,0,1,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.12061186310113349,1.0,1.0,1.0,0.3554393160412404,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.06129635743369649,0.0,1.0,1.0,0.8788408139676869,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11742049081104876,1.0,1.0,1.0,0.38472306870888867,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.1402553097832068,1.0,1.0,1.0,0.4917873581105606,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06575327390778034,1.0,1.0,1.0,0.712739903075119,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.08215032463959503,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.06036095521074061,1.0,1.0,1.0,0.7762312226178342,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.25420931000330144,1.0,1.0,0.0,0.27088325935678026,0,0,1,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.0817651590183779,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.027731924727632886,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.18075272367117862,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.35721360184879497,1.0,1.0,1.0,0.8794147982207655,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.37839771101573677,0.0,0.0,0.0,0.6396221742784345,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.030593155056674374,0.0,0.0,1.0,0.2247881972110395,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14537251017937713,1.0,1.0,1.0,0.530059222287824,0,0,0,0,1,0,0,1,0,0,1,0
+0.5714285714285714,0.30538131396500495,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15709254979641246,1.0,0.0,1.0,0.8205202256524424,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06250687795752173,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.11852096401452623,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19962583911081766,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15637724221415208,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,1,0,0,0,1,0,0
+0.4285714285714286,0.1371739848134698,1.0,1.0,1.0,0.8656027328560056,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.07285132607021019,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0872125013755915,0.0,0.0,0.0,0.1840254387203246,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13579839330912294,0.0,0.0,0.0,0.3243121077691882,0,0,1,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.2527786948387807,1.0,1.0,0.0,0.7355348054855501,1,0,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.052217453505007144,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.17857142857142858,0.20485308682733577,1.0,0.0,1.0,0.2907185455724763,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.1336524705623418,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.10360955210740619,0.0,1.0,1.0,0.8532586635277519,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.05397821063057115,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.14427203697589966,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.11978650819852536,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+1.0,0.3877517332452955,1.0,1.0,0.0,0.121581563557335,0,0,0,0,1,1,0,0,1,0,0,0
+0.4642857142857143,0.40420380763728403,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.625,0.7664245625619016,1.0,1.0,1.0,0.2911457072868785,0,0,0,1,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06883459887751733,0.0,1.0,0.0,0.6447959977442328,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.19775503466490593,1.0,1.0,0.0,0.3285309193360387,0,0,0,1,0,0,0,1,0,1,0,0
+0.6785714285714286,0.20864971937933313,1.0,1.0,1.0,0.7473544772820154,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,0.0,1.0,1.0,0.45983457538120076,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.41619896555518876,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.03626059205458347,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.34158688235941453,1.0,1.0,0.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11962143721800374,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13965004952129417,0.0,0.0,1.0,0.8404307054775538,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024595576097722022,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.12248266754704522,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0706503796632552,1.0,0.0,0.0,0.5995907180616916,0,1,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.04500935402222955,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.18284362275778587,0.0,0.0,1.0,0.44241615651549826,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.05782986684274238,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3805436337625179,1.0,1.0,1.0,0.8152117172112082,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.09843732805106195,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.12286783316826234,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18526466380543635,1.0,0.0,1.0,0.4847650698535884,0,0,0,0,1,1,0,0,0,0,1,0
+0.3571428571428571,0.09948277759436558,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.08281060856168151,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.07202597116760207,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16017387476614944,1.0,1.0,0.0,0.34047407781017125,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.5486959392538792,1.0,1.0,1.0,0.7743267172453298,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.062231759656652355,0.0,1.0,0.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.10955210740618465,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12589413447782546,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.4300099042588313,0.0,0.0,0.0,0.1073942894895514,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.33927588863211183,0.0,1.0,0.0,0.26164945103668286,0,0,0,0,1,1,0,0,0,1,0,0
+0.3571428571428571,0.24683613954000222,0.0,0.0,0.0,0.3676314741249905,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.037416088918234836,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.1463629360625069,1.0,1.0,1.0,0.3565290863994199,0,0,0,1,0,0,0,1,1,0,0,0
+0.24999999999999997,0.11158798283261802,0.0,1.0,0.0,0.5545926067069147,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2318146803125344,1.0,1.0,0.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11797072741278751,0.0,0.0,1.0,0.594377341023918,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1103224386486189,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.625,0.4587872785297678,1.0,1.0,1.0,0.8024775365103721,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3022449653350941,0.0,0.0,1.0,0.2949724434247011,0,1,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.05227247716518102,0.0,0.0,1.0,0.24744012641724394,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08897325850115549,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.0335644327060636,0.0,1.0,1.0,0.4603229620470974,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.04258831297457907,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.36238582590513924,1.0,1.0,0.0,0.121581563557335,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.07868383404864092,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06399251678221635,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.03615054473423572,1.0,1.0,1.0,0.7468001783284491,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05188731154396391,0.0,1.0,0.0,0.28597713146577547,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1280400572246066,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2047980631671619,0.0,0.0,1.0,0.7832044401604136,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1930229998899527,0.0,0.0,1.0,0.3671745216659698,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.11389897655992078,1.0,1.0,1.0,0.9071687984489271,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.12176735996478487,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06014086057004511,1.0,1.0,1.0,0.7691184526170824,0,0,0,0,1,0,1,0,0,0,1,0
+0.4285714285714286,0.4167492021569275,1.0,1.0,1.0,0.8656027328560056,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.16644657202597118,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.16474083856058105,0.0,1.0,1.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.4107142857142857,0.46016287003411466,1.0,1.0,1.0,0.8472095845527893,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.08759766699680863,0.0,1.0,0.0,0.19533656176753816,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.12991086167051832,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.08928571428571427,0.09794211510949709,1.0,0.0,1.0,0.676443925470061,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.49669858038956755,1.0,1.0,1.0,0.8189994124469444,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.045449543303620554,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07131066358534169,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09040387366567623,0.0,1.0,1.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10404974138879718,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.425332893144052,1.0,1.0,0.0,0.8216427748498599,0,1,0,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.055793991416309016,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,1,0,0,0,0,0,1
+0.24999999999999997,0.10999229668757565,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.324254429404644,0.0,1.0,0.0,0.0955803074919803,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.1842742379223066,0.0,1.0,1.0,0.35908907651024774,0,0,0,1,0,1,0,0,1,0,0,0
+0.19642857142857142,0.07081545064377683,1.0,1.0,1.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.590073731704633,0.0,1.0,0.0,0.7468841586935758,1,0,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.2226807527236712,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.05172224056344228,0.0,1.0,0.0,0.27039187876105036,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.056344228018047754,0.0,1.0,0.0,0.39882243364250436,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.11863101133487398,1.0,1.0,1.0,0.9344131207554249,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33856058104985143,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.05705953560030813,1.0,1.0,1.0,0.9319236160108605,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.3138549576317817,0.0,1.0,1.0,0.3593859563426322,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.09392538791680423,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07164080554638494,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.03780125453945196,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09805216242984482,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.5846263893474194,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.2659843732805106,0.0,1.0,1.0,0.491978924064202,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.2712116209970287,1.0,1.0,0.0,0.31774368328346525,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.044018928139099814,1.0,1.0,1.0,0.7619553600617908,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2015516672169033,1.0,0.0,1.0,0.31967729971945585,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.15401122482667548,0.0,1.0,1.0,0.5639451140754361,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.44486629250577747,1.0,1.0,0.0,0.21396727904669746,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.08847804555959062,0.0,0.0,1.0,0.2848197012851972,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.25272367117860683,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,1,0,0,1,0,0,0
+0.24999999999999997,0.19423352041377795,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09783206778914934,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11191812479366127,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1363486299108617,1.0,1.0,1.0,0.7303600113933694,0,0,0,1,0,0,1,0,0,1,0,0
+0.4107142857142857,0.27748431825685044,1.0,1.0,0.0,0.3549890108481963,1,0,0,0,0,0,1,0,1,0,0,0
+1.0,0.4900957411687025,1.0,1.0,1.0,0.16559346885680104,0,0,1,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.09315505667436998,0.0,0.0,0.0,0.7904631079537381,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.148949048090679,1.0,1.0,1.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.11769560911191813,0.0,1.0,1.0,0.3403565857713484,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08979861340376362,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08060966215472654,0.0,1.0,1.0,0.383421164823506,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.33459887751733247,1.0,1.0,0.0,0.14138446961748627,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.0586552217453505,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.8058765269065697,1.0,1.0,0.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.4642857142857143,0.15555188731154398,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.0,0.07120061626499395,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.041652910751623196,1.0,0.0,1.0,0.5528336477984951,0,0,1,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.10008803785627819,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.7002310993727302,0.0,1.0,0.0,0.4721076735930411,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.1570375261362386,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.5714285714285714,0.20578848905029165,0.0,1.0,1.0,0.3694096935983138,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.17877187190491914,0.0,1.0,0.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.048750962914053037,1.0,1.0,0.0,0.6705704969262621,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.007813359744690218,0.0,0.0,1.0,0.4681806327003752,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.048475844613183675,0.0,1.0,0.0,0.33991531191906393,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.6785714285714286,0.17299438758666227,1.0,1.0,1.0,0.22696837267036857,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.3713546825134808,1.0,1.0,0.0,0.21980151898449196,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.6464179597226807,1.0,1.0,0.0,0.32719918796985015,0,1,0,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.14201606690877078,0.0,1.0,1.0,0.7583632325020305,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09436557719819522,0.0,1.0,1.0,0.4443019547874373,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07411687025420931,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.09975789589523495,1.0,1.0,1.0,0.7623117828116139,0,1,0,0,0,1,0,0,0,1,0,0
+0.3928571428571428,0.42456256190161773,1.0,1.0,1.0,0.5811811839164577,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.4034334763948498,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.24369979091009136,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.30543633762517886,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.1990756025090789,0.0,1.0,1.0,0.8301501720360537,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.11775063277209201,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.3631561571475735,1.0,1.0,0.0,0.6870648053224668,1,0,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.0675690546935182,1.0,0.0,0.0,0.9134302134303278,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10069329811819082,1.0,0.0,1.0,0.8891646790727171,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.03747111257840871,0.0,1.0,0.0,0.6730269139352237,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05469351821283151,0.0,1.0,1.0,0.9015902882244003,0,0,0,0,1,1,0,0,0,0,0,1
+0.10714285714285712,0.10355452844723231,0.0,1.0,1.0,0.41906888316914087,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0935402222955871,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.06597336854847584,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.07329151535160118,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-test.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-test.csv
new file mode 100644
index 0000000..5e83fb1
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-test.csv
@@ -0,0 +1,201 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.30802244965335096,1.0,0.0,1.0,0.31967729971945585,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.11290855067679102,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.04693518212831517,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0704302850225597,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.08093980411576977,1.0,1.0,1.0,0.4768786443351435,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1389347419390338,1.0,1.0,0.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08974358974358974,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,1,0,0,0,0,1,0
+0.19642857142857142,0.15291075162319798,1.0,1.0,1.0,0.9318352763505735,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.07741828986464179,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.11571475734565863,1.0,1.0,1.0,0.5855097720668199,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.13255199735886433,1.0,1.0,1.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07384175195333993,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09216463079124024,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.594475624518543,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09447562451854297,0.0,0.0,1.0,0.3676314741249905,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.13321228128095083,0.0,1.0,0.0,0.2581995981513079,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.17497523935292178,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.23236491691427316,1.0,1.0,1.0,0.48667382453412633,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.24347969626939586,0.0,0.0,1.0,0.8290700279000807,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.041212721470232194,1.0,0.0,1.0,0.8205202256524424,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.5893584241223726,1.0,1.0,0.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.05876526906569825,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.019753494002421042,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.1509298998569385,1.0,1.0,1.0,0.8769881149919502,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.847529437658193,1.0,0.0,1.0,0.5547639995361563,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17156377242214152,1.0,0.0,1.0,0.8695372347753674,0,0,0,0,1,1,0,0,0,0,0,1
+0.7857142857142857,0.24969736986904373,1.0,1.0,1.0,0.21980151898449196,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10982722570705403,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08237041928029053,1.0,0.0,1.0,0.2729209608367981,0,1,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.2666446572025971,0.0,1.0,0.0,0.22634832395437568,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.034885000550236606,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.32023770221195114,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06228678331682623,1.0,1.0,1.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.02767690106745901,0.0,1.0,1.0,0.6295469835412176,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.13464289644547156,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.01870804445911742,1.0,1.0,1.0,0.7623117828116139,0,1,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.09183448883019699,1.0,1.0,0.0,0.539932226665866,0,1,0,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07686805326290305,1.0,1.0,0.0,0.3692311661992841,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3090128755364807,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3132496973698691,0.0,1.0,0.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.026246285902938263,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14097061736546715,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,1,0,0,1,0,0,0
+0.7857142857142857,0.22328601298558381,0.0,0.0,0.0,0.0810126268631418,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.3266754704522945,1.0,0.0,1.0,0.26441754960248154,0,0,0,1,0,1,0,0,1,0,0,0
+0.24999999999999997,0.05749972488169913,0.0,1.0,1.0,0.5545926067069147,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3004842082095301,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.32243864861890614,1.0,1.0,1.0,0.6234592356361524,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.44767249917464513,1.0,1.0,0.0,0.591004897644727,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.09414548255749973,0.0,0.0,1.0,0.43705682731209555,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.04996148343787829,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.06382744580169472,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5178571428571429,0.385330692197645,1.0,1.0,1.0,0.8654407578963966,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.023825244855287777,0.0,1.0,1.0,0.681321561513423,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06399251678221635,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.2817761637504127,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.2217453505007153,0.0,1.0,0.0,0.383421164823506,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06096621547265324,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.0292175635523275,1.0,1.0,1.0,0.6696986103919891,0,0,0,0,1,0,0,1,0,0,1,0
+0.0,0.17222405634422802,0.0,1.0,1.0,0.9329250944686631,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.045779685264663805,1.0,1.0,1.0,0.7387979215486421,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.11483437878287663,1.0,1.0,1.0,0.47720039923434754,0,0,0,0,1,0,1,0,0,1,0,0
+0.053571428571428575,0.03279410146362936,1.0,1.0,1.0,0.9178375942401232,0,0,0,0,1,0,0,1,0,0,0,1
+0.7857142857142857,0.45322988885220644,0.0,0.0,1.0,0.20756498401823725,0,0,0,1,0,0,1,0,0,1,0,0
+0.19642857142857142,0.09106415758776273,1.0,1.0,1.0,0.6955811084767616,0,0,0,0,1,1,0,0,0,0,1,0
+0.14285714285714285,0.08550676791020138,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.022119511389897654,0.0,0.0,1.0,0.2848197012851972,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.059480576647958625,1.0,0.0,0.0,0.5995907180616916,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.15566193463189174,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1340926598437328,1.0,1.0,1.0,0.7377669250251128,0,0,0,1,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05942555298778475,1.0,1.0,1.0,0.7215189623993398,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05051172003961703,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.17585561791570375,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.052217453505007144,0.0,1.0,1.0,0.6908920489455583,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02701661714537251,0.0,0.0,0.0,0.2844193852648953,1,0,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.39198855507868385,1.0,1.0,1.0,0.6594326055741871,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10559040387366568,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.05986574226917574,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33944095961263343,0.0,1.0,1.0,0.22408792958769466,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.039892153626059204,1.0,0.0,1.0,0.311205543841052,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13392758886321118,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23528117090348852,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.038571585781886214,0.0,1.0,0.0,0.6155200677110325,0,1,0,0,0,1,0,0,0,1,0,0
+0.8928571428571428,0.8635963464289644,1.0,1.0,0.0,0.3004289224837474,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.18190822053483,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.0,0.06630351050951909,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09271486739297898,1.0,1.0,0.0,0.6014238410536501,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.1581379993397161,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.07114559260482008,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14933421371189612,1.0,1.0,1.0,0.3103438400976657,0,0,0,0,1,0,0,1,1,0,0,0
+0.24999999999999997,0.06558820292725871,1.0,1.0,0.0,0.41602094847665755,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.7431495543083525,1.0,1.0,1.0,0.5481571705989746,0,0,0,0,1,0,0,1,0,0,0,1
+0.2857142857142857,0.3245295477055134,1.0,1.0,1.0,0.6408270671508309,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.3336084516342027,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.41900517222405637,1.0,1.0,0.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.05760977220204688,1.0,0.0,1.0,0.5621939327570208,0,0,1,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.121712336304611,1.0,0.0,0.0,0.2849263027670955,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.3582590513920986,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,1,0,0,0,0,0,1
+0.5714285714285714,0.5129305601408606,1.0,1.0,0.0,0.8216427748498599,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39391438318476946,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.36519203257400684,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18548475844613185,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.06063607351160999,0.0,0.0,1.0,0.3382029230298863,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06767910201386595,1.0,1.0,1.0,0.650463907744803,1,0,0,0,0,0,0,1,0,0,1,0
+0.24999999999999997,0.042808407615274574,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.035875426433366345,1.0,0.0,0.0,0.43311856998291953,0,0,0,1,0,1,0,0,1,0,0,0
+0.10714285714285712,0.1136238582590514,1.0,1.0,1.0,0.5291896437274354,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.09519093210080334,0.0,0.0,0.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.07169582920655881,0.0,0.0,1.0,0.41718500620577564,0,0,0,1,0,1,0,0,0,1,0,0
+0.10714285714285712,0.05447342357213601,0.0,1.0,0.0,0.5654556515140055,0,0,0,0,1,0,0,1,0,0,1,0
+0.14285714285714285,0.05353802134918015,1.0,1.0,0.0,0.7138311294496358,0,0,0,0,1,1,0,0,0,1,0,0
+0.3035714285714286,0.12963574336964895,0.0,1.0,1.0,0.31273337858185585,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06377242214152085,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.07252118410916694,0.0,0.0,1.0,0.7569602202607998,0,0,1,0,0,1,0,0,0,0,0,1
+0.01785714285714285,0.1759656652360515,1.0,1.0,1.0,0.9540917374927895,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1992956971497744,0.0,1.0,1.0,0.5541068942854244,1,0,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.116430064927919,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4087157477715418,0.0,1.0,1.0,0.7844990820387668,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5033564432706064,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.5714285714285714,0.5126004181798173,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.0928799383735006,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10448993067018818,1.0,1.0,0.0,0.38048671184038707,0,0,1,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.350170573346539,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.14355672939363928,1.0,1.0,1.0,0.9262782795387031,1,0,0,0,0,0,1,0,0,0,0,1
+0.2857142857142857,0.13090128755364808,1.0,1.0,1.0,0.5956378735010521,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.32761087267525035,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.18851105975569496,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.19642857142857142,0.114229118520964,1.0,1.0,1.0,0.6449158361571704,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.17244415098492352,1.0,1.0,0.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.1546164850885881,1.0,1.0,0.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05771981952239463,1.0,1.0,1.0,0.7333540005028889,0,0,1,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.06426763508308572,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.04104765048971058,0.0,0.0,1.0,0.8565716032957927,0,0,0,1,0,0,1,0,0,0,0,1
+0.2857142857142857,0.1780015406624849,1.0,1.0,1.0,0.9000458966532298,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17211400902388027,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.05359304500935402,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.02723671178606801,0.0,1.0,0.0,0.6088670652522917,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.08858809287993837,1.0,1.0,1.0,0.7134293371692706,1,0,0,0,0,0,1,0,0,1,0,0
+0.2857142857142857,0.2213051612193243,0.0,0.0,1.0,0.22169998970541557,0,0,0,0,1,0,1,0,1,0,0,0
+0.053571428571428575,0.11912622427643886,1.0,1.0,1.0,0.7174552542410282,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.19406844943325632,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.10625068779575217,1.0,1.0,1.0,0.539932226665866,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.239022779795312,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.028007043028502255,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.23550126554418402,0.0,1.0,1.0,0.8239601442576517,0,0,0,1,0,1,0,0,0,0,0,1
+0.10714285714285712,0.09210960713106636,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17591064157587763,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+1.0,0.7580609662154726,1.0,1.0,0.0,0.29141195522842805,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14383184769450866,1.0,1.0,0.0,0.624582242671011,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.05529877847474414,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.22669747991636405,1.0,1.0,0.0,0.26590645487362585,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.03719599427753935,0.0,1.0,1.0,0.8896713869108571,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.030373060415978873,1.0,1.0,0.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.04632992186640256,0.0,1.0,1.0,0.5692869095347953,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.07378672829316606,1.0,1.0,1.0,0.9140325439012884,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.23131946737096953,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023770221195113902,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.3362495873225487,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.055463849455265765,0.0,0.0,1.0,0.8290700279000807,0,1,0,0,0,1,0,0,0,0,0,1
+0.4107142857142857,0.12490370859469573,1.0,0.0,0.0,0.39742586223642357,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.00484208209530098,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.2857142857142857,0.37454605480356556,1.0,1.0,1.0,0.6871883067965017,0,0,0,1,0,1,0,0,0,1,0,0
+0.2857142857142857,0.10922196544514141,1.0,1.0,0.0,0.3790392890575462,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,0.0,1.0,1.0,0.6244595378782976,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.6095521074061846,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05496863651370089,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10278419720479806,1.0,1.0,1.0,0.7001676841419655,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05694948827996038,0.0,1.0,0.0,0.3307077506353231,0,0,0,1,0,1,0,0,1,0,0,0
+0.7857142857142857,0.18311874105865522,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.14295146913172666,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.17932210850665786,0.0,0.0,1.0,0.276086333823588,0,0,0,1,0,0,1,0,1,0,0,0
+0.07142857142857142,0.05029162539892153,1.0,1.0,1.0,0.4679701801537487,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.2050181578078574,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.18498954550456698,0.0,1.0,1.0,0.5545926067069147,0,0,0,0,1,0,1,0,0,1,0,0
+1.2142857142857142,0.2941014636293606,1.0,0.0,0.0,0.09102457054520846,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025255860019808517,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.43903378452734676,1.0,1.0,0.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.17409486079013978,1.0,1.0,0.0,0.4148665750517137,0,1,0,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.6802024870694399,1.0,1.0,0.0,0.3516352721472709,0,0,1,0,0,0,0,1,0,1,0,0
+0.03571428571428571,0.08253549026081215,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.15555188731154398,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.2791900517222406,0.0,1.0,1.0,0.8368159662459218,0,0,0,0,1,0,0,1,0,0,0,1
+0.0,0.06844943325630021,1.0,1.0,1.0,0.7836588733899705,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.12809508088478047,0.0,1.0,1.0,0.32271911415921684,1,0,0,0,0,0,0,1,1,0,0,0
+0.053571428571428575,0.11439418950148562,0.0,1.0,1.0,0.6637429098283977,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.022229558710245404,1.0,1.0,0.0,0.4917873581105606,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.4337515131506548,0.0,1.0,1.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19880048420820953,1.0,1.0,1.0,0.8189994124469444,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.39738087377572356,0.0,0.0,1.0,0.48957791029126796,1,0,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.07180587652690656,1.0,1.0,0.0,0.9005973681731101,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.08928571428571427,0.02971277649389237,0.0,1.0,1.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.27423792230659183,1.0,1.0,0.0,0.5394438034973912,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.11863101133487398,0.0,0.0,0.0,0.18745666815478274,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1496093320127655,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.3035714285714286,0.6839440959612633,1.0,1.0,0.0,0.8613899174477632,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.045779685264663805,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.20166171453725104,1.0,1.0,0.0,0.5235081475203799,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.19709475074281943,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09232970177176185,1.0,1.0,0.0,0.6598740118142182,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.5286123032904149,1.0,1.0,1.0,0.5291642625196091,0,0,0,1,0,0,1,0,0,1,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-train.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-train.csv
new file mode 100644
index 0000000..1627cd1
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-train.csv
@@ -0,0 +1,501 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.5714285714285714,0.20408275558490152,1.0,1.0,1.0,0.2740485661497681,1,0,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.18306371739848135,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.12462859029382635,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09601628700341147,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+1.0,0.5566193463189171,1.0,1.0,1.0,0.6276600176708876,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.39303400462198745,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.008748761967646089,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,1,0,0,0,0,1,0
+0.08928571428571427,0.12418840101243535,1.0,1.0,1.0,0.9112945990675128,0,0,0,0,1,0,0,1,0,0,0,1
+0.4642857142857143,0.25096291405304283,0.0,0.0,1.0,0.737517899393192,0,0,0,1,0,1,0,0,0,0,0,1
+0.7857142857142857,0.3471992956971498,1.0,0.0,0.0,0.28983662188342285,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.48332783096731596,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.1816881258941345,1.0,1.0,0.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.2857142857142857,0.3421371189611533,1.0,1.0,1.0,0.6403744137297571,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1250137559150435,1.0,1.0,1.0,0.32719918796985015,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.14141080664685815,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.029327610872675252,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.19329811819082207,1.0,0.0,1.0,0.34386083170097576,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.39809618135798397,0.0,1.0,0.0,0.7914015454381981,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06344228018047761,0.0,1.0,1.0,0.3257054908520209,0,0,1,0,0,1,0,0,1,0,0,0
+0.125,0.24876196764608782,1.0,1.0,1.0,0.7620595412443067,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.20743919885550788,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.052327500825354895,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.28232640035215145,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.28947947617475517,1.0,0.0,1.0,0.7905497651969635,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.04666006382744581,0.0,1.0,1.0,0.862481750443303,1,0,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.32871134587872786,1.0,1.0,0.0,0.28436974428843736,0,0,0,0,1,0,0,1,0,1,0,0
+0.125,0.06289204357873886,0.0,1.0,1.0,0.8939030026966095,0,1,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.09183448883019699,1.0,1.0,1.0,0.7769155728885463,0,0,0,1,0,1,0,0,0,1,0,0
+0.19642857142857142,0.2638384505337295,0.0,1.0,1.0,0.8769316531253295,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.3136898866512601,0.0,0.0,0.0,0.1493953955906259,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.026466380543633764,1.0,1.0,1.0,0.5499994958063575,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.02085396720589854,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08572686255089688,0.0,0.0,0.0,0.8228490355418768,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.05161219324309453,0.0,1.0,0.0,0.523377616709452,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.16424562561901618,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.6223175965665235,0.0,0.0,0.0,0.3243121077691882,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.04974138879718279,1.0,1.0,1.0,0.7387979215486421,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.19808517662594918,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11604489930670188,1.0,1.0,0.0,0.3842577810457014,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07059535600308132,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.21789369428854408,1.0,1.0,0.0,0.4390666360957615,0,0,1,0,0,0,1,0,0,0,1,0
+0.4642857142857143,0.19847034224716628,1.0,1.0,1.0,0.2864857137967765,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06289204357873886,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.08528667326950588,1.0,0.0,1.0,0.8291441697520612,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.1101023440079234,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03224386486189061,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06025090789039286,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.2500825354902608,0.0,1.0,1.0,0.3694096935983138,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.025200836359634642,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.05716958292065588,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08880818752063387,0.0,1.0,0.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.5342797402883239,0.0,1.0,0.0,0.26164945103668286,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.27797953119841534,1.0,1.0,1.0,0.36651873073537256,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.08924837680202487,1.0,1.0,1.0,0.48838112309751835,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.12044679212061186,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06261692527786948,0.0,1.0,1.0,0.8832307883744815,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.6558270056124132,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.13271706833938596,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.1607142857142857,0.06377242214152085,0.0,1.0,1.0,0.8854921566472022,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1866952789699571,0.0,1.0,1.0,0.3691093043872932,0,0,0,0,1,0,1,0,1,0,0,0
+0.7857142857142857,0.358093980411577,0.0,1.0,0.0,0.10812898709422081,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.0880928799383735,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.17420490811048753,1.0,1.0,1.0,0.3128458574189485,0,0,1,0,0,0,1,0,1,0,0,0
+0.4107142857142857,0.2678001540662485,1.0,1.0,1.0,0.8921185284956621,0,0,0,1,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07411687025420931,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.06228678331682623,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.05397821063057115,1.0,1.0,1.0,0.9319236160108605,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13101133487399583,0.0,1.0,0.0,0.5382443052921533,0,0,1,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.10454495433036205,0.0,0.0,0.0,0.27183024502308256,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.1474634092659844,0.0,1.0,1.0,0.6634322478338307,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.3582590513920986,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,0,1,0,0,0,0,1
+0.3035714285714286,0.13728403213381754,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11417409486079015,1.0,1.0,1.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05249257180587652,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.04396390447892594,0.0,0.0,1.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.11235831407505227,1.0,1.0,0.0,0.9262782795387031,1,0,0,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.06597336854847584,0.0,1.0,1.0,0.9185698239945864,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.39952679652250467,1.0,1.0,0.0,0.36651873073537256,0,0,0,0,1,0,0,1,1,0,0,0
+0.08928571428571427,0.03675580499614835,0.0,1.0,0.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.1153295917244415,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07444701221525256,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.125,0.2011114779355123,1.0,1.0,1.0,0.4708753344299414,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0482007263123143,0.0,0.0,1.0,0.7904631079537381,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.14707824364476726,0.0,1.0,1.0,0.4849768887446002,0,0,1,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.1620446792120612,0.0,1.0,1.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.24573566633652472,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.7871134587872785,1.0,0.0,0.0,0.6577144367789873,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2884890502916254,1.0,1.0,1.0,0.21396727904669746,0,0,0,0,1,0,0,1,1,0,0,0
+0.3571428571428571,0.0734565863321228,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.18377902498074172,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.1380543633762518,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10300429184549356,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.09640145262462858,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.09084406294706723,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.06839440959612633,0.0,1.0,1.0,0.8898642220971337,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.052767690106745896,1.0,1.0,0.0,0.7619553600617908,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20518322878837902,1.0,1.0,1.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.4110817651590184,0.0,1.0,1.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.1870804445911742,0.0,0.0,1.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.08968856608341587,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09304500935402223,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.034334763948497854,0.0,0.0,1.0,0.8105188297717469,0,1,0,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.2718168812589414,1.0,1.0,1.0,0.8704016314829519,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.5627269725982172,1.0,1.0,1.0,0.7682673905629588,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.22565203037306042,0.0,1.0,1.0,0.4443019547874373,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.06195664135578298,0.0,1.0,1.0,0.8575078279322482,0,0,0,1,0,0,1,0,0,0,0,1
+0.4642857142857143,0.1786068009243975,1.0,1.0,1.0,0.49210970933040776,0,0,1,0,0,0,1,0,0,1,0,0
+0.6428571428571428,0.3162759986794322,1.0,1.0,0.0,0.20892794417015212,0,0,1,0,0,0,1,0,1,0,0,0
+0.4285714285714286,0.20666886761307363,1.0,1.0,0.0,0.30391227130738524,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.1701331572576208,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.13051612193243095,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11742049081104876,1.0,1.0,0.0,0.47720039923434754,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07752833718498954,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1781666116430065,0.0,0.0,1.0,0.7832044401604136,0,0,0,1,0,0,1,0,0,0,0,1
+1.0,0.39440959612633436,1.0,1.0,1.0,0.21615402523359528,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.10938703642566304,0.0,0.0,1.0,0.7569602202607998,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.059590623968306375,1.0,1.0,0.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.3214285714285714,0.05683944095961263,0.0,0.0,1.0,0.7970917912232778,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.17981732144822274,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.018432926158248045,1.0,0.0,1.0,0.5528336477984951,0,0,1,0,0,1,0,0,0,1,0,0
+0.625,0.25767580059425554,1.0,0.0,0.0,0.3730439023791205,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.056564322658743255,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.08352591614394189,1.0,0.0,1.0,0.8618695563470995,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.17051832287883792,1.0,1.0,0.0,0.3103438400976657,0,1,0,0,0,0,0,1,1,0,0,0
+0.08928571428571427,0.0627819962583911,1.0,1.0,1.0,0.7001676841419655,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.17453505007153078,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10432485968966655,0.0,0.0,1.0,0.8406172417730042,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.06988004842082095,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3627159678661825,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09909761197314845,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15109497083746012,1.0,0.0,0.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18763068119291296,0.0,1.0,1.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06558820292725871,0.0,0.0,0.0,0.2287663808190244,0,0,0,1,0,1,0,0,1,0,0,0
+0.08928571428571427,0.04886101023440079,1.0,0.0,1.0,0.36745330533116055,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.3037856278199626,0.0,0.0,1.0,0.737517899393192,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.12930560140860572,1.0,1.0,0.0,0.3232634558617386,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.19252778694838782,0.0,1.0,1.0,0.501464322292528,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.15456146142841423,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022669747991636405,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09298998569384835,1.0,1.0,1.0,0.36651873073537256,0,1,0,0,0,0,0,1,1,0,0,0
+0.5714285714285714,0.1140640475404424,1.0,0.0,1.0,0.4028715067363472,0,0,0,1,0,1,0,0,0,1,0,0
+0.7857142857142857,0.6577528337184989,1.0,1.0,1.0,0.3131231838824953,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.28925938153405967,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.24999999999999997,0.010069329811819083,0.0,0.0,0.0,0.7901372777915644,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.23995818201826785,0.0,1.0,0.0,0.9121910841333913,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.13821943435677342,0.0,0.0,0.0,0.49006922670215497,0,0,0,0,1,0,1,0,0,1,0,0
+0.4642857142857143,0.2213051612193243,1.0,1.0,1.0,0.8302239413120919,0,0,1,0,0,0,1,0,0,0,0,1
+0.625,0.6333223286012986,1.0,1.0,1.0,0.49778098104998286,0,0,0,1,0,1,0,0,0,1,0,0
+0.5714285714285714,0.10867172884340266,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.06734896005282272,0.0,1.0,1.0,0.6244595378782976,0,1,0,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06338725652030373,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.09150434686915374,1.0,1.0,1.0,0.6598740118142182,0,1,0,0,0,0,1,0,0,1,0,0
+0.17857142857142858,0.030373060415978873,1.0,1.0,1.0,0.9052897264392974,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.05617915703752613,1.0,1.0,0.0,0.5758465012701344,0,0,1,0,0,0,0,1,0,0,1,0
+0.3571428571428571,0.055463849455265765,0.0,1.0,1.0,0.45368904432201734,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.18240343347639487,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17255419830527127,1.0,1.0,0.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.1098822493672279,0.0,1.0,1.0,0.5785653349165971,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.30560140860570045,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.03906679872345109,0.0,1.0,1.0,0.8976829391418685,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.044349070100143065,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3322878837900297,1.0,1.0,1.0,0.42442143235694413,0,1,0,0,0,0,1,0,0,0,1,0
+0.08928571428571427,0.13772422141520854,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.10784637394079453,1.0,0.0,0.0,0.8715443022869169,0,0,1,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.5528227137669197,1.0,1.0,0.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.5714285714285714,0.7740728513260702,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.2200396170353252,1.0,1.0,0.0,0.5394438034973912,1,0,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.03736106525806096,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.08928571428571427,0.039011775063277215,0.0,1.0,0.0,0.599794302613941,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.13931990756025092,0.0,1.0,1.0,0.8721840349758069,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.44613183668977663,0.0,1.0,1.0,0.26164945103668286,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.11285352701661715,0.0,0.0,1.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.7857142857142857,0.18245845713656875,1.0,1.0,1.0,0.40471109118578136,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.20446792120611865,0.0,1.0,0.0,0.3403565857713484,0,0,0,0,1,1,0,0,1,0,0,0
+0.03571428571428571,0.05430835259161439,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03994717728623308,0.0,0.0,0.0,0.2364918904353026,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.03659073401562672,0.0,1.0,0.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.03835149114119071,1.0,1.0,0.0,0.5813781424290714,0,0,0,1,0,0,0,1,0,0,1,0
+0.3571428571428571,0.14168592494772753,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.3035714285714286,0.11213821943435677,0.0,0.0,1.0,0.7688898730617176,0,1,0,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.145867723120942,0.0,1.0,1.0,0.8978106855080499,0,1,0,0,0,1,0,0,0,0,0,1
+0.7321428571428571,0.23803235391218225,1.0,1.0,1.0,0.38456298371763664,1,0,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3845053372950369,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.13733905579399142,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.08215032463959503,1.0,1.0,1.0,0.6014238410536501,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.06140640475404424,0.0,0.0,0.0,0.31109338775758766,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06905469351821283,1.0,1.0,1.0,0.8533241696172105,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.005117200396170352,0.0,1.0,1.0,0.4603229620470974,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.10856168152305491,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.06850445691647408,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.11428414218113787,0.0,0.0,0.0,0.4681806327003752,0,1,0,0,0,0,1,0,0,0,1,0
+0.7321428571428571,0.6364036535710355,1.0,1.0,0.0,0.20599972812749537,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.06789919665456146,1.0,0.0,0.0,0.5528336477984951,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4844833278309673,1.0,1.0,1.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.07730824254429404,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.06619346318917134,0.0,1.0,1.0,0.8627147751385414,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.053813139650049524,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.037526136238582586,0.0,1.0,1.0,0.7170669202424681,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.4311654011224827,1.0,1.0,0.0,0.47720039923434754,0,0,0,0,1,0,1,0,0,1,0,0
+0.03571428571428571,0.06030593155056674,1.0,1.0,1.0,0.9209417015586914,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.1437218003741609,1.0,1.0,1.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.15808297567954221,0.0,1.0,0.0,0.28637876253981037,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.3938593595245956,0.0,0.0,0.0,0.1140139540448384,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055793991416309016,1.0,0.0,0.0,0.5216017218323117,0,0,1,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.07175085286673269,0.0,0.0,1.0,0.8847736592891744,0,0,0,1,0,1,0,0,0,0,0,1
+0.3035714285714286,0.18719049191152196,1.0,1.0,1.0,0.678114089607729,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20254209310003302,1.0,1.0,0.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.03571428571428571,0.061131286453174866,1.0,1.0,1.0,0.5706358388636316,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.052437548145702645,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.18031253438978762,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.4184549356223176,0.0,0.0,0.0,0.594377341023918,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.17095851216022892,0.0,1.0,1.0,0.31273337858185585,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.10179377132166832,1.0,1.0,0.0,0.624582242671011,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.0505667436997909,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.47854077253218885,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.24999999999999997,0.09276989105315285,1.0,0.0,1.0,0.3727423047912524,0,0,0,1,0,1,0,0,1,0,0,0
+0.3571428571428571,0.1794871794871795,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3695939253879168,1.0,1.0,1.0,0.6500940471086876,0,0,0,1,0,1,0,0,0,1,0,0
+0.4642857142857143,0.11461428414218112,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09843732805106195,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.49636843842852424,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.3035714285714286,0.07274127874986244,0.0,1.0,1.0,0.8471418368374785,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.011224826675470454,1.0,0.0,1.0,0.6581569155182169,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.33311323869263787,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.10713106635853417,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.7857142857142857,0.40601958842302194,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.05639925167822163,0.0,0.0,1.0,0.5204868269574562,0,1,0,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.37795752173434577,1.0,1.0,0.0,0.11967916113567616,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.6783867062837019,0.0,1.0,0.0,0.4107052500869476,0,0,0,0,1,0,0,1,0,1,0,0
+0.24999999999999997,0.1391548365797293,1.0,0.0,1.0,0.4901999941597845,0,0,1,0,0,1,0,0,0,1,0,0
+0.0,0.06894464619786508,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.07142857142857142,0.05430835259161439,0.0,0.0,0.0,0.4840535155941271,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.20160669087707717,1.0,1.0,0.0,0.6970354972308322,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15472653240893586,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,1,0,0,0,0,0,1
+0.5714285714285714,0.09370529327610873,1.0,1.0,0.0,0.39646283586920644,0,0,0,0,1,0,0,1,0,1,0,0
+0.4642857142857143,0.19709475074281943,1.0,0.0,1.0,0.4133255669397803,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.0708704743039507,0.0,1.0,1.0,0.9121910841333913,0,1,0,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10223396060305931,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.00979421151094971,0.0,1.0,1.0,0.4603229620470974,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.2313744910311434,1.0,1.0,0.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.3035714285714286,0.1138439528997469,1.0,1.0,0.0,0.6399187311966064,0,1,0,0,0,1,0,0,0,0,1,0
+0.14285714285714285,0.05023660173874765,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.037746230879278087,1.0,1.0,1.0,0.9345335095933344,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.12231759656652359,1.0,0.0,0.0,0.3250877505238663,1,0,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07257620776934082,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.1522504677011115,0.0,1.0,1.0,0.6155200677110325,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.11450423682183337,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.10157367668097282,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04330362055683944,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.05425332893144051,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.21464729833828547,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.18482447452404535,0.0,1.0,1.0,0.5962182584447656,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.06393749312204247,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.125,0.20298228238142402,1.0,1.0,1.0,0.4708753344299414,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.0702101903818642,1.0,1.0,1.0,0.42657246837101115,0,0,0,0,1,0,0,1,1,0,0,0
+0.3214285714285714,0.1334323759216463,1.0,1.0,1.0,0.8922523883382062,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.2184989545504567,0.0,1.0,1.0,0.383421164823506,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.14713326730494114,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08622207549246176,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.0536480686695279,1.0,1.0,1.0,0.6993390090705507,0,0,1,0,0,0,1,0,0,0,1,0
+0.19642857142857142,0.08776273797733025,0.0,1.0,1.0,0.3691093043872932,0,1,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.022559700671288655,1.0,0.0,1.0,0.9228647175807431,0,0,0,1,0,1,0,0,0,0,0,1
+0.03571428571428571,0.025255860019808517,1.0,1.0,1.0,0.7691184526170824,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.8587542643336634,1.0,1.0,1.0,0.2740485661497681,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.08732254869593925,0.0,1.0,0.0,0.46182459874957915,0,0,0,1,0,0,0,1,0,1,0,0
+0.1607142857142857,0.08512160228898426,1.0,1.0,1.0,0.4978064484847755,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010894684714427203,0.0,0.0,0.0,0.3386431143464047,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.16303510509519095,1.0,0.0,1.0,0.5380335258497126,0,1,0,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.12815010454495435,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.05216242984483327,0.0,1.0,0.0,0.3803633753089816,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.26152745680642675,0.0,1.0,0.0,0.8206450815998698,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11461428414218112,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.055683944095961266,1.0,1.0,1.0,0.9318352763505735,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.049796412457356665,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.10714285714285712,0.03543523715197534,0.0,1.0,1.0,0.9153811453083084,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.16952789699570817,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.04748541873005392,0.0,1.0,1.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02580609662154727,1.0,1.0,0.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.09981291955540883,0.0,1.0,0.0,0.49148753840738346,1,0,0,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.18741058655221746,1.0,1.0,1.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.3502255970067129,1.0,1.0,1.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.24999999999999997,0.16391548365797293,1.0,1.0,1.0,0.9071687984489271,1,0,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.17018818091779467,1.0,1.0,1.0,0.868365131914781,0,0,1,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.1572576207769341,1.0,1.0,0.0,0.32719918796985015,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.011444921316165951,1.0,0.0,1.0,0.5836541255891218,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.06718388907230109,1.0,1.0,1.0,0.868365131914781,0,0,1,0,0,0,0,1,0,0,0,1
+0.125,0.05898536370639375,0.0,1.0,1.0,0.5795267454297757,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.057114559260482006,1.0,1.0,1.0,0.5396118319655506,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.025145812699460767,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.058215032463959496,1.0,1.0,1.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.08928571428571427,0.15538681633102236,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.12440849565313085,0.0,0.0,1.0,0.259549729145002,1,0,0,0,0,1,0,0,1,0,0,0
+0.03571428571428571,0.0819852536590734,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.04803565533179267,0.0,1.0,0.0,0.8696334095426655,0,0,1,0,0,1,0,0,0,0,0,1
+0.125,0.07301639705073182,0.0,0.0,1.0,0.5005400663825238,0,1,0,0,0,0,1,0,0,1,0,0
+0.7321428571428571,0.08776273797733025,1.0,0.0,0.0,0.08470200286988733,0,0,1,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.09783206778914934,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.4107142857142857,0.4439308902828216,0.0,1.0,0.0,0.38066739707842107,0,0,0,0,1,0,0,1,0,1,0,0
+0.14285714285714285,0.05749972488169913,0.0,0.0,0.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.6431165401122483,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.08633212281280951,1.0,1.0,0.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0.4768786443351435,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06965995378012545,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.1037746230879278,1.0,0.0,1.0,0.36745330533116055,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.19599427753934193,0.0,0.0,1.0,0.8105188297717469,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.21541762958071972,1.0,1.0,0.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.20947507428194126,1.0,1.0,0.0,0.6500940471086876,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.543468691537361,1.0,1.0,0.0,0.6234592356361524,0,0,1,0,0,0,0,1,0,0,0,1
+0.03571428571428571,0.012875536480686695,1.0,1.0,1.0,0.7979365203848136,0,0,0,1,0,0,1,0,0,1,0,0
+1.0,0.33223286012985587,1.0,1.0,0.0,0.16559346885680104,0,0,1,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.36436667767139874,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.07054033234290745,0.0,1.0,1.0,0.8547131179732448,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.10366457576758006,1.0,1.0,1.0,0.921786795969523,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04715527676901067,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,1,0,0,1,0,0,0
+0.2857142857142857,0.10795642126114229,1.0,1.0,1.0,0.47653149674170675,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.16154946627049632,0.0,0.0,1.0,0.8404307054775538,0,0,0,1,0,1,0,0,0,0,0,1
+0.7678571428571428,0.5762077693408165,0.0,1.0,1.0,0.13279434507188115,0,1,0,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.02641135688345989,1.0,1.0,1.0,0.9394762243447705,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.10030813249697369,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0176625949158138,1.0,1.0,1.0,0.4148665750517137,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.11609992296687575,1.0,1.0,1.0,0.6535565228181367,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.06019588423021899,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.6785714285714286,0.3809838230439089,0.0,1.0,0.0,0.18862334019529733,0,0,0,1,0,0,1,0,1,0,0,0
+0.6785714285714286,0.2501375591504347,1.0,1.0,1.0,0.3466918762732373,0,0,0,0,1,0,0,1,0,0,1,0
+0.6785714285714286,0.20441289754594477,1.0,1.0,0.0,0.22731351394237334,0,1,0,0,0,0,1,0,1,0,0,0
+0.125,0.38395510069329813,1.0,1.0,1.0,0.915529351477635,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05315285572796302,1.0,1.0,0.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.13744910311433917,0.0,1.0,1.0,0.8532586635277519,0,0,0,0,1,0,0,1,0,0,0,1
+0.6785714285714286,0.49702872235061074,1.0,1.0,1.0,0.33770845427133345,1,0,0,0,0,0,0,1,0,1,0,0
+0.4107142857142857,0.23528117090348852,1.0,1.0,1.0,0.8704016314829519,0,1,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.2750082535490261,1.0,1.0,1.0,0.8769881149919502,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16127434796962695,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.02569604930119952,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.055683944095961266,1.0,0.0,1.0,0.8618695563470995,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.08990866072411137,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.09210960713106636,0.0,1.0,0.0,0.5545926067069147,0,1,0,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.39760096841641906,1.0,1.0,1.0,0.16857126089732433,0,0,0,1,0,0,0,1,1,0,0,0
+0.14285714285714285,0.03791130185979971,1.0,1.0,0.0,0.7613436169716516,0,0,0,1,0,0,1,0,0,0,1,0
+0.14285714285714285,0.03251898316275999,0.0,0.0,1.0,0.5419816737959281,0,0,0,1,0,1,0,0,0,1,0,0
+0.3571428571428571,0.21255639925167824,0.0,1.0,0.0,0.491978924064202,0,1,0,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.2697259821723341,1.0,0.0,1.0,0.5367070201327678,0,0,0,1,0,0,1,0,0,0,1,0
+0.5178571428571429,0.22180037416088919,0.0,0.0,0.0,0.12004837083863333,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.07246616044899307,0.0,0.0,1.0,0.44241615651549826,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.004897105755474855,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.19642857142857142,0.05639925167822163,0.0,0.0,0.0,0.2247881972110395,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.021404203807637284,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1127434796962694,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.086552217453505,1.0,1.0,0.0,0.3842577810457014,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.41339275888632115,1.0,1.0,0.0,0.14138446961748627,0,0,0,0,1,0,0,1,1,0,0,0
+0.03571428571428571,0.07400682293386156,1.0,1.0,1.0,0.9522216912358249,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.007428194123473095,0.0,1.0,1.0,0.44959115198776406,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.028942445251458126,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.15511169803015298,1.0,1.0,1.0,0.520597760213557,0,0,0,0,1,0,0,1,0,1,0,0
+0.08928571428571427,0.06129635743369649,1.0,1.0,1.0,0.590693107733845,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0534279740288324,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.17321448222735777,1.0,1.0,1.0,0.6007809114857595,0,0,0,1,0,0,1,0,1,0,0,0
+0.24999999999999997,0.21475734565863322,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.08407615274568064,0.0,1.0,0.0,0.5853827741356347,0,1,0,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.05425332893144051,0.0,0.0,1.0,0.8461497236114719,0,1,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.08341586882359414,1.0,0.0,1.0,0.8891646790727171,0,0,1,0,0,1,0,0,0,0,0,1
+0.4642857142857143,0.43039506988004844,1.0,0.0,1.0,0.22597418414948542,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.268460437988335,0.0,1.0,0.0,0.5664245567595892,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.4620886981402003,1.0,1.0,0.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.21563772422141522,1.0,1.0,1.0,0.34047407781017125,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.09271486739297898,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11813579839330911,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06652360515021459,0.0,1.0,1.0,0.8547131179732448,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.023935292175635527,1.0,0.0,0.0,0.6483538884899454,0,0,0,1,0,0,1,0,0,1,0,0
+0.7857142857142857,0.2396280400572246,1.0,0.0,0.0,0.08995843040536927,0,0,0,0,1,0,0,1,1,0,0,0
+0.125,0.10410476504897105,1.0,1.0,1.0,0.9292175567150299,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.03851656212171234,1.0,1.0,0.0,0.4768786443351435,0,0,0,0,1,1,0,0,1,0,0,0
+0.21428571428571427,0.050896885660834154,1.0,1.0,1.0,0.602425169638959,1,0,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.06866952789699571,0.0,0.0,1.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.09541102674149884,1.0,1.0,1.0,0.8533241696172105,0,0,1,0,0,0,0,1,0,0,0,1
+0.5714285714285714,0.19555408825795093,0.0,1.0,0.0,0.3259888088958532,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.09557609772202047,1.0,0.0,0.0,0.23658638755348427,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.21398701441619897,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.8928571428571428,0.5054473423572136,1.0,1.0,1.0,0.6415019923339554,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.16176956091119182,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,0,1,0,1,0,0,0
+0.1607142857142857,0.034774953229888855,1.0,0.0,1.0,0.5894933892681636,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13810938703642567,1.0,1.0,1.0,0.520597760213557,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.10845163420270716,0.0,0.0,1.0,0.7793274409005319,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.020688896225376913,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,1.0,0.0,1.0,0.0,0.22634832395437568,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.0532078793881369,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.053571428571428575,0.12798503356443272,1.0,1.0,1.0,0.7174552542410282,0,0,1,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.13189171343677783,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,1,0,0,1,0,0,0
+0.4107142857142857,0.12765489160338947,0.0,0.0,0.0,0.6811557639743221,0,0,1,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1594585671838891,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.03571428571428571,0.022889842632331906,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.14285714285714285,0.051447122262572906,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.3138549576317817,1.0,1.0,1.0,0.8794147982207655,0,0,0,1,0,0,1,0,0,0,0,1
+0.7321428571428571,0.24738637614174097,1.0,0.0,0.0,0.2731769653496512,0,1,0,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.0529877847474414,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.09067899196654561,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.6239683063717398,0.0,0.0,0.0,0.46989347577510143,0,0,1,0,0,1,0,0,0,0,0,1
+0.0,0.01931330472103004,0.0,0.0,1.0,0.8714857078947565,0,1,0,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.05474854187300539,1.0,1.0,0.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.30218994167492025,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.17857142857142858,0.06382744580169472,1.0,1.0,1.0,0.6964156448050416,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.035765379113018594,1.0,0.0,0.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0061626499394739735,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.09513590844062947,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.17051832287883792,0.0,1.0,0.0,0.709384236480772,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.030097942115109497,0.0,1.0,0.0,0.8898642220971337,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.014746340926598437,0.0,1.0,1.0,0.895580236194126,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1129635743369649,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.1354682513480797,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06734896005282272,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.08638714647298339,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09227467811158797,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06509298998569385,0.0,0.0,1.0,0.2247881972110395,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.03895675140310334,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.5714285714285714,0.19725982172334106,0.0,1.0,1.0,0.7472556622006505,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.028227137669197756,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.15010454495433037,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.05727963024100363,0.0,1.0,1.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.0233300319137229,1.0,0.0,0.0,0.43311856998291953,0,0,0,1,0,0,1,0,1,0,0,0
+1.0,0.3453835149114119,1.0,1.0,1.0,0.5003985149708152,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.03994717728623308,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.20353251898316277,1.0,1.0,1.0,0.7380360665955924,0,1,0,0,0,0,1,0,0,0,1,0
+0.10714285714285712,0.043908880818752064,1.0,0.0,1.0,0.5734408207763663,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.13315725762077696,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.4642857142857143,0.23841751953339935,1.0,1.0,1.0,0.2864857137967765,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.11692527786948388,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.056784417299438755,0.0,0.0,0.0,0.24744012641724394,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.19296797622977882,0.0,1.0,1.0,0.7786348012503493,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.2195994277539342,1.0,1.0,0.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.3482447452404534,1.0,1.0,1.0,0.30992319586753697,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.10294926818531969,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,0.0,1.0,1.0,0.8898642220971337,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.18399911962143722,0.0,1.0,1.0,0.8769316531253295,0,1,0,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.24408495653130846,1.0,1.0,1.0,0.7324169714111652,0,0,1,0,0,0,0,1,0,0,0,1
+0.19642857142857142,0.2424892703862661,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.17857142857142858,0.48024650599757895,1.0,1.0,0.0,0.4873369967986317,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.04429404643996919,0.0,1.0,1.0,0.8627147751385414,0,1,0,0,0,0,1,0,0,0,0,1
+0.7857142857142857,0.15412127214702323,1.0,1.0,0.0,0.15889026804756212,0,0,1,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.17558049961483438,0.0,0.0,0.0,0.27183024502308256,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.26961593485198637,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.09425552987784747,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,1,0,0,0,0,0,1
+0.24999999999999997,0.18377902498074172,1.0,1.0,1.0,0.6594326055741871,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3076923076923077,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.06707384175195334,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.12947067238912735,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.18713546825134808,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.15951359084406297,1.0,0.0,1.0,0.2126849403417187,0,1,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.21239132827115662,1.0,0.0,0.0,0.2729209608367981,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.06531308462638935,1.0,1.0,0.0,0.7769155728885463,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.062011665015956854,0.0,1.0,1.0,0.419918901878422,0,0,0,0,1,0,0,1,0,0,1,0
+0.08928571428571427,0.06448772972378122,1.0,1.0,0.0,0.45755650252851443,0,1,0,0,0,0,0,1,1,0,0,0
+0.4642857142857143,0.18553978210630573,0.0,1.0,1.0,0.23787440014014885,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.17029822823814242,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.020633872565203038,1.0,1.0,1.0,0.7138311294496358,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.040332342907450205,0.0,0.0,1.0,0.49006922670215497,0,1,0,0,0,1,0,0,0,1,0,0
+0.4642857142857143,0.15081985253659075,0.0,0.0,1.0,0.31136986465762495,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.049466270496313414,1.0,1.0,1.0,0.8899154955245911,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.001430615164520744,1.0,0.0,1.0,0.5836541255891218,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.06168152305491362,0.0,0.0,0.0,0.16589905516842904,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.029987894794761747,0.0,1.0,0.0,0.39882243364250436,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.056784417299438755,0.0,1.0,0.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.0064927919005172245,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.055463849455265765,1.0,0.0,1.0,0.8491471130930849,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.10652580609662154,0.0,1.0,1.0,0.8976829391418685,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.14223616154946628,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.18609001870804448,0.0,0.0,1.0,0.16589905516842904,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.05838010344448112,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11543963904478925,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.010069329811819083,0.0,0.0,0.0,0.5527042865014291,0,1,0,0,0,1,0,0,0,1,0,0
+0.19642857142857142,0.20408275558490152,0.0,1.0,0.0,0.3257054908520209,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.13101133487399583,0.0,0.0,1.0,0.41167927397905774,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.12077693408165512,1.0,1.0,1.0,0.9264124244522206,0,1,0,0,0,1,0,0,0,0,0,1
+0.10714285714285712,0.04335864421701331,1.0,1.0,1.0,0.5805283757278361,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.473643666776714,1.0,1.0,1.0,0.7117589057670359,0,0,0,1,0,0,0,1,0,0,0,1
+0.3571428571428571,0.05639925167822163,1.0,1.0,1.0,0.5640737884995887,0,0,1,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.3476394849785408,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.06019588423021899,1.0,1.0,0.0,0.6586754162607802,0,0,0,1,0,0,1,0,0,0,1,0
+0.4107142857142857,0.3144602178936943,1.0,1.0,1.0,0.5709518098205263,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.020248706943985912,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.10714285714285712,0.026246285902938263,1.0,1.0,1.0,0.9078496135630285,0,0,0,0,1,0,0,1,0,0,0,1
+0.19642857142857142,0.06569825024760646,1.0,0.0,1.0,0.5690789555483984,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.18542973478595798,1.0,1.0,1.0,0.23796927750170668,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.10955210740618465,1.0,1.0,1.0,0.5291896437274354,0,1,0,0,0,1,0,0,1,0,0,0
+1.0,0.799603829646748,0.0,1.0,0.0,0.13365287548774402,0,0,0,0,1,0,0,1,0,1,0,0
+0.3571428571428571,0.09750192582810609,0.0,1.0,1.0,0.8013985500332685,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.019753494002421042,0.0,1.0,0.0,0.6244595378782976,0,1,0,0,0,0,1,0,0,0,1,0
+0.3571428571428571,0.15247056234180698,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.2272477165181028,1.0,1.0,1.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.10714285714285712,0.05827005612413337,1.0,1.0,0.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.14185099592824915,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08501155496863651,0.0,1.0,1.0,0.5545926067069147,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.07059535600308132,1.0,1.0,0.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.26702982282381427,1.0,1.0,1.0,0.8456935742394528,0,1,0,0,0,0,0,1,0,0,0,1
+0.4642857142857143,0.5707604269836029,1.0,1.0,1.0,0.2592550937927965,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.06250687795752173,0.0,1.0,0.0,0.8696334095426655,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.23885770881479035,1.0,1.0,0.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.17326950588753165,1.0,1.0,1.0,0.7215189623993398,0,0,0,0,1,0,1,0,0,0,1,0
+0.03571428571428571,0.04385385715857819,0.0,1.0,1.0,0.6383519380348411,0,0,1,0,0,0,1,0,0,0,1,0
+0.7857142857142857,0.4091009133927589,0.0,1.0,0.0,0.15312328184273788,0,0,0,1,0,1,0,0,1,0,0,0
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-val.csv b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-val.csv
new file mode 100644
index 0000000..4856718
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/RS_NS_NI_MS_NB_HE_ME_RW_OLR_NP-val.csv
@@ -0,0 +1,301 @@
+month,credit_amount,sex,age,credit,pred_credit,employment=A71,employment=A72,employment=A73,employment=A74,employment=A75,housing=A151,housing=A152,housing=A153,status=A11,status=A12,status=A13,status=A14
+0.24999999999999997,0.07169582920655881,1.0,1.0,0.0,0.3989479019699039,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.06107626279300099,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.02806206668867613,0.0,1.0,1.0,0.6997655328413775,0,0,0,1,0,0,1,0,0,1,0,0
+0.24999999999999997,0.14575767580059426,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.7321428571428571,0.15302079894354573,1.0,0.0,0.0,0.23681941072883386,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.24551557169582922,1.0,1.0,1.0,0.5547218618582729,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.090073731704633,1.0,1.0,1.0,0.6156438955835984,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.11131286453174864,1.0,1.0,1.0,0.5291642625196091,0,0,0,1,0,0,1,0,0,1,0,0
+0.14285714285714285,0.035105095190932106,1.0,1.0,0.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.07538241443820842,0.0,1.0,1.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.19642857142857142,0.1352481567073842,1.0,1.0,1.0,0.9173729937673434,0,1,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.03389457466710685,1.0,0.0,1.0,0.4901999941597845,0,0,1,0,0,0,1,0,0,1,0,0
+0.07142857142857142,0.06404754044239022,1.0,1.0,1.0,0.7088870911194225,0,0,1,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05849015076482887,1.0,0.0,1.0,0.8955009029182879,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12809508088478047,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.1825685044569165,0.0,1.0,1.0,0.8769316531253295,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.06289204357873886,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10267414988445031,1.0,1.0,1.0,0.7333540005028889,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.09695168922636734,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09298998569384835,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.05904038736656762,1.0,1.0,1.0,0.7691184526170824,0,0,0,0,1,0,1,0,0,0,1,0
+0.5714285714285714,0.09502586112028172,1.0,1.0,0.0,0.27443988356010096,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.047210300429184546,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.19642857142857142,0.05777484318256851,1.0,1.0,1.0,0.8887675866521803,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.0505667436997909,1.0,1.0,1.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.5714285714285714,0.16501595686145043,1.0,1.0,1.0,0.2510304635302003,0,0,0,1,0,0,0,1,1,0,0,0
+0.625,0.1275998679432156,1.0,0.0,1.0,0.6681686332576849,0,0,1,0,0,0,1,0,0,0,0,1
+0.08928571428571427,0.07147573456586331,1.0,1.0,1.0,0.7387979215486421,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.024760647078243648,1.0,1.0,1.0,0.9209417015586914,0,0,0,0,1,0,0,1,0,0,0,1
+0.14285714285714285,0.32067789149334214,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05260261912622427,0.0,0.0,1.0,0.5214711461502933,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.060801144492131615,0.0,0.0,0.0,0.3676314741249905,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15010454495433037,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07015516672169032,0.0,1.0,1.0,0.8575078279322482,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.037526136238582586,0.0,0.0,1.0,0.8373645845436015,0,0,1,0,0,0,1,0,0,0,0,1
+0.4107142857142857,0.12534389787608674,0.0,1.0,1.0,0.46064338284484674,0,1,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.15758776273797734,0.0,1.0,0.0,0.3403565857713484,0,1,0,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.09122922856828436,1.0,1.0,0.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.14285714285714285,0.05447342357213601,0.0,1.0,1.0,0.8898642220971337,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02569604930119952,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.0695499064597777,1.0,0.0,1.0,0.5216017218323117,0,0,1,0,0,0,1,0,0,1,0,0
+0.19642857142857142,0.08688235941454825,1.0,1.0,1.0,0.9173729937673434,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.02145922746781116,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.39517992736876856,0.0,1.0,1.0,0.3259888088958532,0,0,1,0,0,1,0,0,0,1,0,0
+0.14285714285714285,0.0710905689446462,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.5160669087707714,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.00968416419060196,1.0,1.0,1.0,0.9418146463783325,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.24061846594035435,1.0,1.0,0.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.45234951028942444,0.0,0.0,1.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.5714285714285714,0.1904368878617806,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.6877407285132606,1.0,1.0,1.0,0.7743267172453298,0,0,0,1,0,0,1,0,0,0,0,1
+0.6785714285714286,0.4199405744470122,1.0,1.0,1.0,0.20677830560523824,0,0,0,1,0,0,0,1,1,0,0,0
+1.0,0.5444040937603168,0.0,0.0,1.0,0.4443440748703878,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.05480356553317926,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.1840541432816111,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.18526466380543635,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.034995047870584356,0.0,0.0,1.0,0.8001795737339381,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.07131066358534169,1.0,0.0,0.0,0.3160895023001086,0,0,0,1,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10570045119401342,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.5714285714285714,0.13002090899086607,1.0,1.0,1.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.04682513480796743,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.1513700891383295,0.0,1.0,1.0,0.6052277358503451,0,0,0,1,0,1,0,0,0,1,0,0
+0.14285714285714285,0.1116430064927919,1.0,1.0,1.0,0.8820768558276499,0,0,1,0,0,0,0,1,0,0,0,1
+0.10714285714285712,0.06993507208099482,1.0,1.0,1.0,0.6913008425849491,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.04357873885770881,0.0,1.0,0.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.3538571585781886,1.0,1.0,1.0,0.24702882446640287,0,0,1,0,0,0,0,1,0,1,0,0
+0.5714285714285714,0.10426983602949268,1.0,0.0,0.0,0.2184939341622545,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2512930560140861,1.0,1.0,0.0,0.43499565515022004,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.7797402883239792,1.0,0.0,0.0,0.21486216341529946,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.02426543413667877,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.05441839991196214,1.0,1.0,1.0,0.5354741988477034,0,0,1,0,0,0,0,1,0,1,0,0
+0.3035714285714286,0.1384945526576428,1.0,1.0,0.0,0.6311298802642114,0,0,0,0,1,1,0,0,0,1,0,0
+0.24999999999999997,0.33069219764498736,1.0,1.0,1.0,0.70490883351029,0,0,0,1,0,1,0,0,0,1,0,0
+0.24999999999999997,0.027511830086937385,0.0,1.0,0.0,0.33991531191906393,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06894464619786508,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.024815670738417523,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.125,0.04974138879718279,0.0,1.0,1.0,0.8937163943553702,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.16011885110597557,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.10311433916584131,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.02839220864971938,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.3685484758446132,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,1,0,0,0,1,0,0
+0.08928571428571427,0.0482007263123143,1.0,1.0,1.0,0.7460536409951634,0,0,0,0,1,0,1,0,0,0,1,0
+0.24999999999999997,0.05315285572796302,0.0,0.0,0.0,0.2364918904353026,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.10806646858149004,1.0,0.0,1.0,0.3387603186528358,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.20864971937933313,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.04418399911962144,1.0,1.0,1.0,0.4768786443351435,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14559260482007264,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.327335754374381,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.3571428571428571,0.41311764058545175,0.0,1.0,1.0,0.491978924064202,0,0,0,0,1,1,0,0,0,1,0,0
+0.14285714285714285,0.06173654671508748,1.0,1.0,0.0,0.5600008807692614,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.33564432706063607,1.0,1.0,0.0,0.32719918796985015,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.2304941124683614,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.8485748872014967,1.0,0.0,0.0,0.21486216341529946,0,0,1,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.030483107736326624,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3035714285714286,0.07439198855507868,1.0,1.0,1.0,0.4148665750517137,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.3421371189611533,1.0,1.0,0.0,0.7134293371692706,1,0,0,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.06729393639264884,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.2851876306811929,0.0,1.0,1.0,0.8298727730544322,1,0,0,0,0,1,0,0,0,0,0,1
+0.03571428571428571,0.05436337625178826,1.0,1.0,1.0,0.9417068119041873,1,0,0,0,0,0,1,0,0,0,0,1
+0.5178571428571429,0.13832948167712117,0.0,1.0,1.0,0.7345947491674498,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09199955981071861,1.0,1.0,0.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.1522504677011115,1.0,1.0,1.0,0.5493424244597425,0,0,0,0,1,0,1,0,0,0,1,0
+0.3571428571428571,0.052657642786398146,0.0,0.0,0.0,0.1941383022647156,0,1,0,0,0,1,0,0,1,0,0,0
+0.21428571428571427,0.13068119291295258,1.0,1.0,0.0,0.4664405275790275,0,0,0,0,1,1,0,0,1,0,0,0
+0.14285714285714285,0.023880268515461652,1.0,1.0,0.0,0.46045295935712555,0,0,1,0,0,1,0,0,1,0,0,0
+0.08928571428571427,0.07813359744690217,0.0,0.0,0.0,0.5214711461502933,0,1,0,0,0,0,1,0,0,1,0,0
+0.03571428571428571,0.02321998459337515,0.0,1.0,1.0,0.9120334763614909,1,0,0,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.12820512820512822,1.0,1.0,0.0,0.8962138068279807,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.3336634752943766,1.0,1.0,1.0,0.8837277100655102,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1641906019588423,0.0,0.0,0.0,0.1941383022647156,0,1,0,0,0,1,0,0,1,0,0,0
+0.10714285714285712,0.3883569935072081,1.0,1.0,1.0,0.6608113136649546,1,0,0,0,0,0,0,1,0,1,0,0
+0.1607142857142857,0.10184879498184218,0.0,0.0,1.0,0.4796070935730555,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.32898646417959726,0.0,0.0,0.0,0.1271934988531199,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.15219544404093763,1.0,0.0,1.0,0.5621939327570208,0,0,1,0,0,0,1,0,0,0,1,0
+0.14285714285714285,0.014966435567293938,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0926598437328051,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.7857142857142857,0.1376141740948608,1.0,1.0,1.0,0.7359170768632337,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.06943985913942995,1.0,1.0,0.0,0.6449158361571704,0,0,1,0,0,0,1,0,0,1,0,0
+0.4642857142857143,0.12396830637173985,1.0,1.0,1.0,0.539932226665866,0,0,0,0,1,0,1,0,0,1,0,0
+0.3571428571428571,0.05705953560030813,0.0,1.0,1.0,0.8301501720360537,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.6679322108506658,1.0,1.0,0.0,0.3516352721472709,0,0,1,0,0,0,0,1,0,1,0,0
+0.08928571428571427,0.05981071860900186,1.0,1.0,0.0,0.7460536409951634,0,1,0,0,0,0,1,0,0,0,1,0
+0.4107142857142857,0.12061186310113349,1.0,1.0,1.0,0.3554393160412404,0,0,0,0,1,0,1,0,1,0,0,0
+0.10714285714285712,0.06129635743369649,0.0,1.0,1.0,0.8788408139676869,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.11742049081104876,1.0,1.0,1.0,0.38472306870888867,0,0,0,0,1,1,0,0,1,0,0,0
+0.08928571428571427,0.1402553097832068,1.0,1.0,1.0,0.4917873581105606,0,0,1,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.06575327390778034,1.0,1.0,1.0,0.712739903075119,0,0,0,1,0,0,1,0,0,0,1,0
+0.03571428571428571,0.08215032463959503,1.0,1.0,1.0,0.9303800002041727,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.06036095521074061,1.0,1.0,1.0,0.7762312226178342,0,0,0,1,0,0,1,0,0,0,1,0
+0.3571428571428571,0.25420931000330144,1.0,1.0,0.0,0.27088325935678026,0,0,1,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.0817651590183779,0.0,1.0,1.0,0.9086619979671742,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.027731924727632886,1.0,1.0,1.0,0.7138311294496358,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.18075272367117862,1.0,1.0,1.0,0.8501233211422209,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.0675690546935182,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.4642857142857143,0.35721360184879497,1.0,1.0,1.0,0.8794147982207655,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.37839771101573677,0.0,0.0,0.0,0.6396221742784345,0,1,0,0,0,1,0,0,0,0,0,1
+0.19642857142857142,0.030593155056674374,0.0,0.0,1.0,0.2247881972110395,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.14537251017937713,1.0,1.0,1.0,0.530059222287824,0,0,0,0,1,0,0,1,0,0,1,0
+0.5714285714285714,0.30538131396500495,1.0,1.0,1.0,0.4297441010936714,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15709254979641246,1.0,0.0,1.0,0.8205202256524424,0,1,0,0,0,0,1,0,0,0,0,1
+0.19642857142857142,0.06250687795752173,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,1,0,0,0,0,0,1
+0.24999999999999997,0.11852096401452623,0.0,1.0,1.0,0.8384025299994181,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.19962583911081766,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.15637724221415208,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,1,0,0,0,1,0,0
+0.4285714285714286,0.1371739848134698,1.0,1.0,1.0,0.8656027328560056,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.07285132607021019,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.3035714285714286,0.0872125013755915,0.0,0.0,0.0,0.1840254387203246,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.13579839330912294,0.0,0.0,0.0,0.3243121077691882,0,0,1,0,0,1,0,0,0,1,0,0
+0.7857142857142857,0.2527786948387807,1.0,1.0,0.0,0.7355348054855501,1,0,0,0,0,1,0,0,0,0,0,1
+0.3571428571428571,0.052217453505007144,1.0,1.0,0.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.17857142857142858,0.20485308682733577,1.0,0.0,1.0,0.2907185455724763,1,0,0,0,0,0,0,1,1,0,0,0
+0.3571428571428571,0.1336524705623418,0.0,1.0,1.0,0.8301501720360537,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.10360955210740619,0.0,1.0,1.0,0.8532586635277519,0,0,0,0,1,0,0,1,0,0,0,1
+0.3571428571428571,0.05397821063057115,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,1,0,0,1,0,0,0
+0.3571428571428571,0.14427203697589966,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.11978650819852536,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+1.0,0.3877517332452955,1.0,1.0,0.0,0.121581563557335,0,0,0,0,1,1,0,0,1,0,0,0
+0.4642857142857143,0.40420380763728403,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.625,0.7664245625619016,1.0,1.0,1.0,0.2911457072868785,0,0,0,1,0,0,1,0,1,0,0,0
+0.08928571428571427,0.06883459887751733,0.0,1.0,0.0,0.6447959977442328,0,0,0,0,1,0,1,0,0,1,0,0
+0.7857142857142857,0.19775503466490593,1.0,1.0,0.0,0.3285309193360387,0,0,0,1,0,0,0,1,0,1,0,0
+0.6785714285714286,0.20864971937933313,1.0,1.0,1.0,0.7473544772820154,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,0.0,1.0,1.0,0.45983457538120076,1,0,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.41619896555518876,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.03626059205458347,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.34158688235941453,1.0,1.0,0.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.11962143721800374,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.13965004952129417,0.0,0.0,1.0,0.8404307054775538,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.024595576097722022,1.0,1.0,0.0,0.5082778240326536,0,1,0,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.12248266754704522,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.14285714285714285,0.0706503796632552,1.0,0.0,0.0,0.5995907180616916,0,1,0,0,0,1,0,0,0,1,0,0
+0.03571428571428571,0.04500935402222955,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.14285714285714285,0.18284362275778587,0.0,0.0,1.0,0.44241615651549826,0,0,1,0,0,0,1,0,0,1,0,0
+0.24999999999999997,0.05782986684274238,1.0,1.0,1.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.6785714285714286,0.3805436337625179,1.0,1.0,1.0,0.8152117172112082,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.09843732805106195,1.0,0.0,0.0,0.4752943973393717,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.12286783316826234,1.0,0.0,1.0,0.23658638755348427,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.18526466380543635,1.0,0.0,1.0,0.4847650698535884,0,0,0,0,1,1,0,0,0,0,1,0
+0.3571428571428571,0.09948277759436558,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.08281060856168151,0.0,1.0,1.0,0.28637876253981037,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.07202597116760207,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.16017387476614944,1.0,1.0,0.0,0.34047407781017125,0,0,1,0,0,1,0,0,1,0,0,0
+0.7857142857142857,0.5486959392538792,1.0,1.0,1.0,0.7743267172453298,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.062231759656652355,0.0,1.0,0.0,0.2488657768966351,0,0,1,0,0,0,1,0,1,0,0,0
+0.3035714285714286,0.10955210740618465,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.12589413447782546,1.0,1.0,0.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.5714285714285714,0.4300099042588313,0.0,0.0,0.0,0.1073942894895514,0,0,1,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.33927588863211183,0.0,1.0,0.0,0.26164945103668286,0,0,0,0,1,1,0,0,0,1,0,0
+0.3571428571428571,0.24683613954000222,0.0,0.0,0.0,0.3676314741249905,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.037416088918234836,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.1463629360625069,1.0,1.0,1.0,0.3565290863994199,0,0,0,1,0,0,0,1,1,0,0,0
+0.24999999999999997,0.11158798283261802,0.0,1.0,0.0,0.5545926067069147,0,1,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2318146803125344,1.0,1.0,0.0,0.7918122026495104,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.11797072741278751,0.0,0.0,1.0,0.594377341023918,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.1103224386486189,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.625,0.4587872785297678,1.0,1.0,1.0,0.8024775365103721,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.3022449653350941,0.0,0.0,1.0,0.2949724434247011,0,1,0,0,0,0,0,1,0,1,0,0
+0.14285714285714285,0.05227247716518102,0.0,0.0,1.0,0.24744012641724394,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08897325850115549,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.03571428571428571,0.0335644327060636,0.0,1.0,1.0,0.4603229620470974,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.04258831297457907,1.0,1.0,0.0,0.38472306870888867,0,1,0,0,0,0,1,0,1,0,0,0
+1.0,0.36238582590513924,1.0,1.0,0.0,0.121581563557335,0,0,0,0,1,0,1,0,1,0,0,0
+0.14285714285714285,0.07868383404864092,1.0,1.0,1.0,0.5082778240326536,0,0,0,0,1,0,1,0,1,0,0,0
+0.3571428571428571,0.06399251678221635,1.0,1.0,1.0,0.8627767388444526,0,0,1,0,0,0,1,0,0,0,0,1
+0.07142857142857142,0.03615054473423572,1.0,1.0,1.0,0.7468001783284491,0,1,0,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.05188731154396391,0.0,1.0,0.0,0.28597713146577547,1,0,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.1280400572246066,1.0,1.0,1.0,0.46045295935712555,0,0,1,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.2047980631671619,0.0,0.0,1.0,0.7832044401604136,0,0,0,1,0,1,0,0,0,0,0,1
+0.3571428571428571,0.1930229998899527,0.0,0.0,1.0,0.3671745216659698,1,0,0,0,0,1,0,0,0,1,0,0
+0.24999999999999997,0.11389897655992078,1.0,1.0,1.0,0.9071687984489271,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.12176735996478487,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06014086057004511,1.0,1.0,1.0,0.7691184526170824,0,0,0,0,1,0,1,0,0,0,1,0
+0.4285714285714286,0.4167492021569275,1.0,1.0,1.0,0.8656027328560056,0,1,0,0,0,1,0,0,0,0,0,1
+0.3035714285714286,0.16644657202597118,1.0,1.0,1.0,0.8962138068279807,0,0,0,0,1,0,1,0,0,0,0,1
+0.24999999999999997,0.16474083856058105,0.0,1.0,1.0,0.5069022744695886,0,0,1,0,0,0,1,0,0,1,0,0
+0.4107142857142857,0.46016287003411466,1.0,1.0,1.0,0.8472095845527893,0,0,1,0,0,0,1,0,0,0,0,1
+0.5714285714285714,0.08759766699680863,0.0,1.0,0.0,0.19533656176753816,0,1,0,0,0,0,1,0,1,0,0,0
+0.3571428571428571,0.12991086167051832,1.0,1.0,1.0,0.8839295784323556,0,0,0,0,1,0,1,0,0,0,0,1
+0.08928571428571427,0.09794211510949709,1.0,0.0,1.0,0.676443925470061,0,0,0,1,0,0,1,0,0,1,0,0
+0.3571428571428571,0.49669858038956755,1.0,1.0,1.0,0.8189994124469444,0,0,1,0,0,0,0,1,0,0,0,1
+0.14285714285714285,0.045449543303620554,1.0,1.0,1.0,0.9122329885255572,0,0,1,0,0,0,1,0,0,0,0,1
+0.10714285714285712,0.07131066358534169,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09040387366567623,0.0,1.0,1.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.10404974138879718,1.0,1.0,1.0,0.9393970480266846,0,0,0,1,0,0,1,0,0,0,0,1
+0.5714285714285714,0.425332893144052,1.0,1.0,0.0,0.8216427748498599,0,1,0,0,0,1,0,0,0,0,0,1
+0.14285714285714285,0.055793991416309016,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,1,0,0,0,0,0,1
+0.24999999999999997,0.10999229668757565,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.7857142857142857,0.324254429404644,0.0,1.0,0.0,0.0955803074919803,0,0,0,0,1,0,0,1,1,0,0,0
+0.3035714285714286,0.1842742379223066,0.0,1.0,1.0,0.35908907651024774,0,0,0,1,0,1,0,0,1,0,0,0
+0.19642857142857142,0.07081545064377683,1.0,1.0,1.0,0.6874873703107746,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.590073731704633,0.0,1.0,0.0,0.7468841586935758,1,0,0,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.2226807527236712,1.0,1.0,0.0,0.6598740118142182,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.05172224056344228,0.0,1.0,0.0,0.27039187876105036,1,0,0,0,0,0,0,1,1,0,0,0
+0.14285714285714285,0.056344228018047754,0.0,1.0,0.0,0.39882243364250436,0,1,0,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.11863101133487398,1.0,1.0,1.0,0.9344131207554249,1,0,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.33856058104985143,1.0,1.0,1.0,0.6014238410536501,0,1,0,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.05705953560030813,1.0,1.0,1.0,0.9319236160108605,0,0,0,0,1,0,1,0,0,0,0,1
+0.6785714285714286,0.3138549576317817,0.0,1.0,1.0,0.3593859563426322,0,0,0,1,0,0,1,0,0,1,0,0
+0.03571428571428571,0.09392538791680423,0.0,1.0,1.0,0.5122497385912862,0,0,0,1,0,0,1,0,1,0,0,0
+0.3571428571428571,0.07164080554638494,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.3571428571428571,0.03780125453945196,1.0,1.0,1.0,0.8839295784323556,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09805216242984482,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.5714285714285714,0.5846263893474194,1.0,1.0,1.0,0.8216427748498599,0,0,0,0,1,0,1,0,0,0,0,1
+0.3571428571428571,0.2659843732805106,0.0,1.0,1.0,0.491978924064202,0,0,0,0,1,0,1,0,0,1,0,0
+0.5714285714285714,0.2712116209970287,1.0,1.0,0.0,0.31774368328346525,0,0,0,1,0,0,1,0,1,0,0,0
+0.03571428571428571,0.044018928139099814,1.0,1.0,1.0,0.7619553600617908,1,0,0,0,0,0,1,0,0,1,0,0
+0.5714285714285714,0.2015516672169033,1.0,0.0,1.0,0.31967729971945585,0,0,1,0,0,0,1,0,0,0,1,0
+0.24999999999999997,0.15401122482667548,0.0,1.0,1.0,0.5639451140754361,0,1,0,0,0,0,1,0,0,0,1,0
+0.5714285714285714,0.44486629250577747,1.0,1.0,0.0,0.21396727904669746,0,0,0,0,1,0,0,1,1,0,0,0
+0.14285714285714285,0.08847804555959062,0.0,0.0,1.0,0.2848197012851972,0,1,0,0,0,1,0,0,1,0,0,0
+0.14285714285714285,0.25272367117860683,1.0,1.0,0.0,0.5082778240326536,0,0,0,0,1,1,0,0,1,0,0,0
+0.24999999999999997,0.19423352041377795,1.0,1.0,1.0,0.9073342365921735,0,1,0,0,0,0,1,0,0,0,0,1
+0.14285714285714285,0.09783206778914934,1.0,1.0,1.0,0.6731420559668704,0,0,1,0,0,0,1,0,0,1,0,0
+0.3571428571428571,0.11191812479366127,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.19642857142857142,0.1363486299108617,1.0,1.0,1.0,0.7303600113933694,0,0,0,1,0,0,1,0,0,1,0,0
+0.4107142857142857,0.27748431825685044,1.0,1.0,0.0,0.3549890108481963,1,0,0,0,0,0,1,0,1,0,0,0
+1.0,0.4900957411687025,1.0,1.0,1.0,0.16559346885680104,0,0,1,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.09315505667436998,0.0,0.0,0.0,0.7904631079537381,0,1,0,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.148949048090679,1.0,1.0,1.0,0.38472306870888867,0,0,0,0,1,0,1,0,1,0,0,0
+0.24999999999999997,0.11769560911191813,0.0,1.0,1.0,0.3403565857713484,0,1,0,0,0,0,1,0,1,0,0,0
+0.24999999999999997,0.08979861340376362,0.0,0.0,0.0,0.20364878828289232,0,0,1,0,0,1,0,0,1,0,0,0
+0.4642857142857143,0.08060966215472654,0.0,1.0,1.0,0.383421164823506,0,0,1,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.33459887751733247,1.0,1.0,0.0,0.14138446961748627,0,0,0,0,1,0,0,1,1,0,0,0
+0.19642857142857142,0.0586552217453505,1.0,1.0,1.0,0.9016367060401769,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.8058765269065697,1.0,1.0,0.0,0.5706358388636316,0,0,0,0,1,0,1,0,1,0,0,0
+0.4642857142857143,0.15555188731154398,1.0,1.0,1.0,0.85555465817234,0,0,0,0,1,0,1,0,0,0,0,1
+0.0,0.07120061626499395,1.0,1.0,1.0,0.9558919804312949,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.041652910751623196,1.0,0.0,1.0,0.5528336477984951,0,0,1,0,0,0,1,0,0,1,0,0
+0.10714285714285712,0.10008803785627819,1.0,1.0,1.0,0.9187129412749173,0,0,1,0,0,0,1,0,0,0,0,1
+0.24999999999999997,0.7002310993727302,0.0,1.0,0.0,0.4721076735930411,1,0,0,0,0,0,0,1,0,1,0,0
+0.24999999999999997,0.1570375261362386,1.0,1.0,1.0,0.49745855352837404,0,0,0,1,0,0,1,0,1,0,0,0
+0.5714285714285714,0.20578848905029165,0.0,1.0,1.0,0.3694096935983138,0,1,0,0,0,0,1,0,0,1,0,0
+0.14285714285714285,0.17877187190491914,0.0,1.0,0.0,0.3538845096386129,0,0,1,0,0,0,1,0,1,0,0,0
+0.08928571428571427,0.048750962914053037,1.0,1.0,0.0,0.6705704969262621,0,0,0,0,1,0,0,1,0,1,0,0
+0.19642857142857142,0.007813359744690218,0.0,0.0,1.0,0.4681806327003752,0,1,0,0,0,1,0,0,0,0,1,0
+0.24999999999999997,0.048475844613183675,0.0,1.0,0.0,0.33991531191906393,1,0,0,0,0,0,1,0,1,0,0,0
+0.14285714285714285,0.06459777704412897,1.0,1.0,1.0,0.7543809345848546,0,0,0,1,0,0,1,0,0,1,0,0
+0.6785714285714286,0.17299438758666227,1.0,1.0,1.0,0.22696837267036857,1,0,0,0,0,0,1,0,1,0,0,0
+0.7857142857142857,0.3713546825134808,1.0,1.0,0.0,0.21980151898449196,0,0,0,1,0,0,1,0,1,0,0,0
+0.4642857142857143,0.6464179597226807,1.0,1.0,0.0,0.32719918796985015,0,1,0,0,0,0,1,0,1,0,0,0
+0.4642857142857143,0.14201606690877078,0.0,1.0,1.0,0.7583632325020305,0,0,1,0,0,0,1,0,0,0,0,1
+0.3571428571428571,0.09436557719819522,0.0,1.0,1.0,0.4443019547874373,0,0,1,0,0,1,0,0,0,1,0,0
+0.3571428571428571,0.07411687025420931,1.0,1.0,1.0,0.8456935742394528,0,0,0,0,1,0,0,1,0,0,0,1
+0.03571428571428571,0.09975789589523495,1.0,1.0,1.0,0.7623117828116139,0,1,0,0,0,1,0,0,0,1,0,0
+0.3928571428571428,0.42456256190161773,1.0,1.0,1.0,0.5811811839164577,0,1,0,0,0,0,1,0,0,1,0,0
+0.7857142857142857,0.4034334763948498,1.0,1.0,1.0,0.2839698095777251,1,0,0,0,0,0,0,1,0,1,0,0
+0.3571428571428571,0.24369979091009136,1.0,1.0,1.0,0.903630573225354,0,0,0,1,0,0,1,0,0,0,0,1
+0.14285714285714285,0.30543633762517886,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,1,0,0,0,0,0,1
+0.3571428571428571,0.1990756025090789,0.0,1.0,1.0,0.8301501720360537,0,0,0,0,1,1,0,0,0,0,0,1
+0.14285714285714285,0.11775063277209201,1.0,1.0,1.0,0.9264124244522206,0,0,0,0,1,0,1,0,0,0,0,1
+0.19642857142857142,0.3631561571475735,1.0,1.0,0.0,0.6870648053224668,1,0,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.0675690546935182,1.0,0.0,0.0,0.9134302134303278,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.10069329811819082,1.0,0.0,1.0,0.8891646790727171,0,0,1,0,0,0,1,0,0,0,0,1
+0.03571428571428571,0.03747111257840871,0.0,1.0,0.0,0.6730269139352237,0,1,0,0,0,0,1,0,0,1,0,0
+0.08928571428571427,0.05469351821283151,0.0,1.0,1.0,0.9015902882244003,0,0,0,0,1,1,0,0,0,0,0,1
+0.10714285714285712,0.10355452844723231,0.0,1.0,1.0,0.41906888316914087,0,1,0,0,0,1,0,0,1,0,0,0
+0.24999999999999997,0.0935402222955871,1.0,1.0,1.0,0.9234069094768464,0,0,0,1,0,0,1,0,0,0,0,1
+0.03571428571428571,0.06184659403543523,1.0,1.0,1.0,0.5231864157229879,0,0,1,0,0,0,1,0,1,0,0,0
+0.03571428571428571,0.06597336854847584,1.0,1.0,1.0,0.7623117828116139,0,0,0,0,1,0,1,0,0,1,0,0
+0.24999999999999997,0.07329151535160118,1.0,1.0,1.0,0.9073342365921735,0,0,0,0,1,0,1,0,0,0,0,1
diff --git a/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/german_AIF_test.json b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/german_AIF_test.json
new file mode 100644
index 0000000..1e56c7a
--- /dev/null
+++ b/pipeline/logs/2020-07-30_18-09-09-739__german_AIF_test/german_AIF_test.json
@@ -0,0 +1 @@
+{"size": 1000, "features": 8, "categorical features": ["credit", "age", "employment", "housing", "sex", "status"], "numerical features": ["credit_amount", "month"], "domain": {"status": [4, ["A11", "A12", "A14", "A13"]], "month": [4, 72], "credit_amount": [250, 18424], "employment": [5, ["A75", "A73", "A74", "A71", "A72"]], "housing": [3, ["A152", "A153", "A151"]], "sex": [2, ["male", "female"]], "age": [2, ["old", "young"]], "credit": [2, ["good", "bad"]]}}
\ No newline at end of file
diff --git a/pipeline/model/classifiers.py b/pipeline/model/classifiers.py
index 248311f..27e514b 100644
--- a/pipeline/model/classifiers.py
+++ b/pipeline/model/classifiers.py
@@ -3,85 +3,252 @@
 """
 
 import pandas as pd
+import numpy as np
+from aif360.datasets import BinaryLabelDataset
 from sklearn.linear_model import SGDClassifier
 from sklearn.tree import DecisionTreeClassifier
-from pipeline.model.inprocessor import Model
 
+from sklearn.model_selection import GridSearchCV
+from sklearn.pipeline import Pipeline
+from pipeline.step import Step
 
-class SK_LogisticRegression(Model):
-    def __init__(self, df, target_col, loss_func="log", instance_weights=[], seed=0):
+
+class SK_LogisticRegression(Step):
+    def __init__(self, target_col, seed, loss_func="log", instance_weights=[], target_positive=1):
         """
-        :param df: pandas dataframe, stores the data to fit the classifier.
         :param target_col: str, the name of the target variable in above data.
+        :param seed: integer, the seed for random state.
         :param loss_func: str, the name of the loss function used in linear model. Same as the loss parameter in sklearn.linear_model.SGDClassifier.
                          The possible options are ‘hinge’, ‘log’, ‘modified_huber’, ‘squared_hinge’, ‘perceptron’, or a regression loss: ‘squared_loss’, ‘huber’, ‘epsilon_insensitive’, or ‘squared_epsilon_insensitive’.
         :param instance_weights: list of float, each number represents the weight of the sample in above data.
-        :param seed: integer, the seed for random state.
+        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
+
         """
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.loss_func = loss_func
+        self.instance_weights = instance_weights
+        self.target_positive = target_positive
+        self.fitted_step = None
+
+    def fit(self, df):
+        if len(self.instance_weights) == 0:
+            self.instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
+
+        self.fitted_step = SGDClassifier(loss=self.loss_func, random_state=self.seed).fit(np.array(df.drop(columns=[self.target_col])), np.array(df[self.target_col]), sample_weight=self.instance_weights)
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
+        after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+
+        favorable_class_idx = list(self.fitted_step.classes_).index(self.target_positive)
+        after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.fitted_step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
+
+        return after_df
+
+    def name(self):
+        return "SK_LogisticRegression"
+
+    def abbr_name(self):
+        return "LR"
+
+    def step_name(self):
+        return "Model"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
 
-        cur_step = SGDClassifier(loss=loss_func, random_state=seed)
-        super().__init__("@".join(["SK_LogisticRegression", target_col]), cur_step, df, target_col, instance_weights=instance_weights)
+    def fit_only_on_train(self):
+        return True
 
 
-class SK_DecisionTree(Model):
-    def __init__(self, df, target_col, instance_weights=[], seed=0):
+class SK_DecisionTree(Step):
+    def __init__(self, target_col, seed, instance_weights=[], target_positive=1):
         """
-        :param df: pandas dataframe, stores the data to fit the classifier.
         :param target_col: str, the name of the target variable in above data.
-        :param instance_weights: list of float, each number represents the weight of the sample in above data.
         :param seed: integer, the seed for random state.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
+
         """
-        cur_step = DecisionTreeClassifier(random_state=seed)
-        super().__init__("@".join(["SK_DecisionTree", target_col]), cur_step, df, target_col, instance_weights=instance_weights)
 
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.instance_weights = instance_weights
+        self.target_positive = target_positive
+        self.fitted_step = None
+
+    def fit(self, df):
+        if len(self.instance_weights) == 0:
+            self.instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
+
+        self.fitted_step = DecisionTreeClassifier(random_state=self.seed).fit(np.array(df.drop(columns=[self.target_col])), np.array(df[self.target_col]), sample_weight=self.instance_weights)
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
+        after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+
+        favorable_class_idx = list(self.fitted_step.classes_).index(self.target_positive)
+        after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.fitted_step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
+
+        return after_df
+
+    def name(self):
+        return "SK_DecisionTree"
+
+    def abbr_name(self):
+        return "DT"
+
+    def step_name(self):
+        return "Model"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
-class OPT_LogisticRegression(Model):
-    def __init__(self, df, target_col, loss_func="log", max_iter=1000, instance_weights=[], seed=0):
+class OPT_LogisticRegression(Step):
+    def __init__(self, target_col, seed, loss_func="log", max_iter=1000, instance_weights=[], target_positive=1):
         """
-        :param df: pandas dataframe, stores the data to fit the classifier.
         :param target_col: str, the name of the target variable in above data.
+        :param seed: integer, random seed.
         :param loss_func: str, the name of the loss function used in linear model. Same as the loss parameter in sklearn.linear_model.SGDClassifier.
                          The possible options are ‘hinge’, ‘log’, ‘modified_huber’, ‘squared_hinge’, ‘perceptron’, or a regression loss: ‘squared_loss’, ‘huber’, ‘epsilon_insensitive’, or ‘squared_epsilon_insensitive’.
         :param max_iter: integer, max number of iterations of the model.
         :param instance_weights: list of float, each number represents the weight of the sample in above data.
-        :param seed: integer, random seed.
+        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
         """
-        # Update below parameters according to the loss function used
-        param_grid = {
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.loss_func = loss_func
+        self.max_iter = max_iter
+        self.instance_weights = instance_weights
+        self.target_positive = target_positive
+        self.fitted_step = None
+
+        self.param_grid = {
             'learner__loss': [loss_func],
             'learner__penalty': ['l2', 'l1', 'elasticnet'],
             'learner__alpha': [0.00005, 0.0001, 0.005, 0.001]
         }
-        cur_step = SGDClassifier(max_iter=max_iter, random_state=seed)
-        super().__init__("@".join(["OPT_LogisticRegression", target_col]), cur_step, df, target_col, instance_weights=instance_weights, hyper_tune=True, param_grid=param_grid)
 
-class OPT_DecisionTree(Model):
-    def __init__(self, df, target_col, instance_weights=[], seed=0):
+    def fit(self, df):
+        if len(self.instance_weights) == 0:
+            self.instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
+
+        search = GridSearchCV(Pipeline([('learner', SGDClassifier(max_iter=self.max_iter, random_state=self.seed))]), self.param_grid, scoring='roc_auc', cv=5, verbose=1, n_jobs=-1)
+        self.fitted_step = search.fit(np.array(df.drop(columns=[self.target_col])), np.array(df[self.target_col]), None,
+                                 **{'learner__sample_weight': self.instance_weights})
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
+        after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+
+        favorable_class_idx = list(self.fitted_step.classes_).index(self.target_positive)
+        after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.fitted_step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
+
+        return after_df
+
+    def name(self):
+        return "OPT_LogisticRegression"
+
+    def abbr_name(self):
+        return "OLR"
+
+    def step_name(self):
+        return "Model"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class OPT_DecisionTree(Step):
+    def __init__(self, target_col, seed, instance_weights=[], target_positive=1):
         """
-        :param df: pandas dataframe, stores the data to fit the classifier.
         :param target_col: str, the name of the target variable in above data.
-        :param instance_weights: list of float, each number represents the weight of the sample in above data.
         :param seed: integer, random seed.
+        :param instance_weights: list of float, each number represents the weight of the sample in above data.
+        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
         """
-        param_grid = {
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.instance_weights = instance_weights
+        self.target_positive = target_positive
+        self.fitted_step = None
+
+        self.param_grid = {
             'learner__min_samples_split': range(20, 500, 10),
             'learner__max_depth': range(15, 30, 2),
             'learner__min_samples_leaf': [3, 4, 5, 10],
             "learner__criterion": ["gini", "entropy"]
         }
 
-        cur_step = DecisionTreeClassifier(random_state=seed)
-        super().__init__("@".join(["OPT_DecisionTree", target_col]), cur_step, df, target_col, instance_weights=instance_weights, hyper_tune=True, param_grid=param_grid)
+    def fit(self, df):
+        if len(self.instance_weights) == 0:
+            self.instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
+
+        search = GridSearchCV(Pipeline([('learner', DecisionTreeClassifier(random_state=self.seed))]), self.param_grid, scoring='roc_auc', cv=5, verbose=1, n_jobs=-1)
+        self.fitted_step = search.fit(np.array(df.drop(columns=[self.target_col])), np.array(df[self.target_col]), None,
+                                 **{'learner__sample_weight': self.instance_weights})
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
+        after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+
+        favorable_class_idx = list(self.fitted_step.classes_).index(self.target_positive)
+        after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.fitted_step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
 
+        return after_df
+
+    def name(self):
+        return "OPT_DecisionTree"
+
+    def abbr_name(self):
+        return "ODT"
+
+    def step_name(self):
+        return "Model"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
-    cur_o = SK_LogisticRegression(data, "income-per-year")
-    # cur_o = SK_DecisionTree(data, "income-per-year")
-    # cur_o = OPT_LogisticRegression(data, "income-per-year")
-    # cur_o = OPT_DecisionTree(data, "income-per-year")
+    data = pd.read_csv("../../data/german_pre_encoded.csv")
+    # cur_o = SK_LogisticRegression("credit", 0)
+    cur_o = SK_DecisionTree("credit", 0)
+    # cur_o = OPT_LogisticRegression("credit", 0)
+    # cur_o = OPT_DecisionTree("credit", 0)
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
-
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/german_after_" + cur_o.name() + ".csv", index=False)
diff --git a/pipeline/model/fair_classifiers.py b/pipeline/model/fair_classifiers.py
index 085e968..f92c5ad 100644
--- a/pipeline/model/fair_classifiers.py
+++ b/pipeline/model/fair_classifiers.py
@@ -3,35 +3,72 @@
 """
 
 import pandas as pd
+from aif360.datasets import BinaryLabelDataset
 from aif360.algorithms.inprocessing import AdversarialDebiasing
 from aif360.algorithms.inprocessing import MetaFairClassifier
 from aif360.algorithms.inprocessing import PrejudiceRemover
-from pipeline.model.inprocessor import Model
+from pipeline.step import Step
+
 import warnings
 warnings.filterwarnings("ignore")
 
-class AIF_AdversarialDebiasing(Model):
+class AIF_AdversarialDebiasing(Step):
 
-    def __init__(self, df, target_col, sensitive_att, seed=0):
+    def __init__(self, target_col, sensitive_att, seed):
         """
-        :param df: pandas dataframe, stores the data to fit the fair classifier.
-        :param target_col: str, the name of the target variable in above data.
+        :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
-        :param seed: integer, random seed.
-
+        :param seed: integer, the seed for random process.
         """
 
         import tensorflow as tf
-        sess = tf.Session()
-        cur_step = AdversarialDebiasing(unprivileged_groups=[{sensitive_att: 0}], privileged_groups=[{sensitive_att: 1}], scope_name='debiased_classifier', debias=True, sess=sess, seed=seed)
-        super().__init__("@".join(["AIF_AdversarialDebiasing", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+        self.sess = tf.Session()
+
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.fitted_step = None
+
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        self.fitted_step = AdversarialDebiasing(unprivileged_groups=[{self.sensitive_att: 0}], privileged_groups=[{self.sensitive_att: 1}], scope_name='debiased_classifier', debias=True, sess=self.sess, seed=self.seed).fit(aif_df)
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
+
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
+
+    def name(self):
+        return "AdversarialDebiasing"
 
+    def abbr_name(self):
+        return "AD"
 
-class AIF_MetaFairClassifier(Model):
+    def step_name(self):
+        return "FairInprocessor"
 
-    def __init__(self, df, target_col, sensitive_att, fairness_penalty=0.8, fair_metric="sr"):
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class AIF_MetaFairClassifier(Step):
+
+    def __init__(self, target_col, sensitive_att, fairness_penalty=0.8, fair_metric="sr"):
         """
-        :param df: pandas dataframe, stores the data to fit the fair classifier.
         :param target_col: str, the name of the target variable in above data.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
         :param fairness_penalty: float in [0,1], fairness penalty parameter. default is 0.8. The same parameter in aif360.algorithms.inprocessing.MetaFairClassifier.
@@ -39,31 +76,112 @@ def __init__(self, df, target_col, sensitive_att, fairness_penalty=0.8, fair_met
                             The same parameter in aif360.algorithms.inprocessing.MetaFairClassifier.
         """
 
-        cur_step = MetaFairClassifier(tau=fairness_penalty, sensitive_attr=sensitive_att, type=fair_metric)
-        super().__init__("@".join(["AIF_MetaFairClassifier", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.fairness_penalty = fairness_penalty
+        self.fair_metric = fair_metric
+
+        self.fitted_step = None
+
+
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
+                                    protected_attribute_names=[self.sensitive_att])
+        self.fitted_step = MetaFairClassifier(tau=self.fairness_penalty, sensitive_attr=self.sensitive_att, type=self.fair_metric).fit(aif_df)
+
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
 
-class AIF_PrejudiceRemover(Model):
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
 
-    def __init__(self, df, target_col, sensitive_att, fairness_penalty=1.0):
+        return after_df
+
+    def name(self):
+        return "MetaFairClassifier"
+
+    def abbr_name(self):
+        return "MFC"
+
+    def step_name(self):
+        return "FairInprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class AIF_PrejudiceRemover(Step):
+
+    def __init__(self, target_col, sensitive_att, fairness_penalty=1.0):
         """
-        :param df: pandas dataframe, stores the data to fit the fair classifier.
         :param target_col: str, the name of the target variable in above data.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
         :param fairness_penalty: float in [0,1], fairness penalty parameter. default is 1. The same parameter in aif360.algorithms.inprocessing.PrejudiceRemover.
 
         """
         # TODO: fix the bug that cannot import lib of 'getoutput'
-        cur_step = PrejudiceRemover(eta=fairness_penalty, sensitive_attr=sensitive_att, class_attr=target_col)
-        super().__init__("@".join(["AIF_PrejudiceRemover", sensitive_att]), cur_step, df, target_col, sensitive_att=sensitive_att, fair_aware=True)
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.fairness_penalty = fairness_penalty
+
+        self.fitted_step = None
+
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
+                                    protected_attribute_names=[self.sensitive_att])
+        self.fitted_step = PrejudiceRemover(eta=self.fairness_penalty, sensitive_attr=self.sensitive_att, class_attr=self.target_col).fit(aif_df)
+
+        return self
 
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
+
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
+
+    def name(self):
+        return "PrejudiceRemover"
+
+    def abbr_name(self):
+        return "PR"
+
+    def step_name(self):
+        return "FairInprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
-    cur_o = AIF_AdversarialDebiasing(data, "income-per-year", "sex")
-    # cur_o = AIF_MetaFairClassifier(data, "income-per-year", "sex")
-    # cur_o = AIF_PrejudiceRemover(data, "income-per-year", "sex")
+    data = pd.read_csv("../../data/german_pre_encoded.csv")
+    # cur_o = AIF_AdversarialDebiasing("credit", "sex")
+    cur_o = AIF_MetaFairClassifier("credit", "sex")
+    # cur_o = AIF_PrejudiceRemover("credit", "sex") # not working TODO: fix
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
-
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/german_after_" + cur_o.name() + ".csv", index=False)
diff --git a/pipeline/model/inprocessor.py b/pipeline/model/inprocessor.py
deleted file mode 100644
index 3df3909..0000000
--- a/pipeline/model/inprocessor.py
+++ /dev/null
@@ -1,74 +0,0 @@
-"""
-    Super class for all the supported classifier classes including fair-classifiers.
-"""
-import numpy as np
-from aif360.datasets import BinaryLabelDataset
-from sklearn.model_selection import GridSearchCV
-from sklearn.pipeline import Pipeline
-from pipeline.step import Step
-
-class Model(Step):
-    def __init__(self, step_name, step, df, target_col, instance_weights=[], hyper_tune=False, param_grid={}, sensitive_att=None, fair_aware=False, target_positive=1):
-        """
-        :param step_name: str, name of the current input step.
-        :param step: object of the initialized class.
-        :param df: pandas dataframe, stores the data.
-        :param target_col: str, the name of the target attribute.
-        :param instance_weights: list of float in [0,1], each float represents the weight of the sample in above data.
-        :param hyper_tune: boolean, whether to tune the hyper-parameter. Default is False.
-        :param param_grid: dict, stores the search range of the hyper-parameter. When hyper_tune is True, this must be provided.
-        :param sensitive_att: str, the name of a sensitive attribute.
-        :param fair_aware: boolean, whether the model is fair-aware. Default is False.
-        :param target_positive: integer, 0 or 1, represents the positive value of the target attribute. Default is 1.
-        """
-
-        super().__init__(step_name=step_name, df=df, sensitive_att=sensitive_att, target_col=target_col)
-        # assume the data set has been encoded to numerical values
-        if fair_aware: # fair classifiers
-            # intitialize a binary label dataset from AIF 360
-            aif_df = BinaryLabelDataset(df=df, label_names=[target_col], protected_attribute_names=[sensitive_att])
-            fitted_step = step.fit(aif_df)
-            input_score = False
-        else: # regular classifiers
-            if len(instance_weights) == 0:
-                instance_weights = [1 for _ in range(1, df.shape[0] + 1)]
-            if hyper_tune: # grid search for best hyper parameters
-                if not param_grid:
-                    print("Need to specify the search range of the hyper parameters - 'param_grid' is empty!")
-                    raise ValueError
-
-                search = GridSearchCV(Pipeline([('learner', step)]), param_grid, scoring='roc_auc', cv=5, verbose=1, n_jobs=-1)
-                fitted_step = search.fit(np.array(df.drop(columns=[target_col])), np.array(df[target_col]), None, **{'learner__sample_weight': instance_weights})
-            else:
-                fitted_step = step.fit(np.array(df.drop(columns=[target_col])), np.array(df[target_col]), sample_weight=instance_weights)
-            input_score = True
-
-        self.input_score = input_score
-        self.step = fitted_step
-        self.target_positive = target_positive
-
-
-    def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the learned discretizer.
-        :return: pandas dataframe, stores the data after discretize.
-        """
-
-        # initialize AIF360 BinaryLabelDataset
-
-        if self.input_score:  # for regular model, generate score prediction
-            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[])
-            after_df, _ = aif_pred_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
-
-            favorable_class_idx = list(self.step.classes_).index(self.target_positive)
-            after_df[self.pred_target_col] = [x[favorable_class_idx] for x in self.step.predict_proba(np.array(df.drop(columns=[self.target_col])))]
-
-        else:  # for fair model, generate label prediction
-            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col],
-                                             protected_attribute_names=[self.sensitive_att])
-
-            after_aif_df = self.step.predict(aif_pred_df)
-            after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
-            after_df[self.pred_target_col] = after_aif_df.labels
-
-        return after_df
diff --git a/pipeline/postprocess/fair_postprocessors.py b/pipeline/postprocess/fair_postprocessors.py
index efa5b1c..412fdc6 100644
--- a/pipeline/postprocess/fair_postprocessors.py
+++ b/pipeline/postprocess/fair_postprocessors.py
@@ -3,64 +3,217 @@
 """
 
 import pandas as pd
+import numpy as np
+from aif360.datasets import BinaryLabelDataset
 from aif360.algorithms.postprocessing import CalibratedEqOddsPostprocessing
 from aif360.algorithms.postprocessing import EqOddsPostprocessing
 from aif360.algorithms.postprocessing import RejectOptionClassification
-from pipeline.postprocess.postprocessor import Postprocessor
+from pipeline.step import Step
 
-class AIF_EqOddsPostprocessing(Postprocessor):
+class NoFairPostprocessor(Step):
+    def __init__(self):
+        self.fitted_step = None
 
-    def __init__(self, df, target_col, sensitive_att, threshold=0.5, seed=0):
+    def fit(self, df):
+        pass
+
+    def apply(self, df):
+        return df
+
+    def name(self):
+        return "NoPostprocessor"
+
+    def abbr_name(self):
+        return "NP"
+
+    def step_name(self):
+        return "FairPostprocessor"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class AIF_EqOddsPostprocessing(Step):
+
+    def __init__(self, target_col, sensitive_att, seed, threshold=0.5):
         """
-        :param df: pandas dataframe, stores the data to fit the postprocessor.
         :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param seed: integer, the seed for random process.
         :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
-        :param seed: integer, the seed for random state.
         """
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.clf_threshold = threshold
+        self.fitted_step = None
+
+
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_true_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]), label_names=[self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+
+        aif_pred_df = aif_true_df.copy()
+        aif_pred_df.labels = np.array([int(x >= self.clf_threshold) for x in df[self.pred_target_col]])
+
+        self.fitted_step = EqOddsPostprocessing([{self.sensitive_att: 0}], [{self.sensitive_att: 1}], self.seed).fit(aif_true_df, aif_pred_df)
+
+        return self
+
+    def apply(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        df["pred_label_" + self.target_col] = [int(x >= self.clf_threshold) for x in df[self.pred_target_col]]
+        aif_pred_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]),
+                                             label_names=["pred_label_" + self.target_col],
+                                             protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
+
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
+
+    def name(self):
+        return "Eq_odds"
+
+    def abbr_name(self):
+        return "EQ"
+
+    def step_name(self):
+        return "FairPostprocessor"
+
+    def input_encoded_data(self):
+        return True
 
-        cur_step = EqOddsPostprocessing([{sensitive_att: 0}], [{sensitive_att: 1}], seed)
-        super().__init__("@".join(["AIF_EqOddsPostprocessing", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=False, clf_threshold=threshold)
+    def output_encoded_data(self):
+        return False
 
+    def fit_only_on_train(self):
+        return True
 
-class AIF_CalibratedEqOddsPostprocessing(Postprocessor):
+class AIF_CalibratedEqOddsPostprocessing(Step):
 
-    def __init__(self, df, target_col, sensitive_att, threshold=0.5, seed=0, cost_constraint='weighted'):
+    def __init__(self, target_col, sensitive_att, seed, cost_constraint='weighted'):
         """
         :param df: pandas dataframe, stores the data to fit the postprocessor.
         :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
-        :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
-        :param seed: integer, the seed for random state.
+        :param seed: integer, the seed for random process.
         :param cost_constraint: str, the fairness constraints format, value from [fpr, fnr, weighted].
                                 The same parameter as in aif360.algorithms.postprocessing.CalibratedEqOddsPostprocessing.
         """
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+        self.seed = seed
+        self.cost_constraint = cost_constraint
+        self.fitted_step = None
+
+    def fit(self, df):
+
+        aif_true_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]), label_names=[self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+        aif_pred_df = aif_true_df.copy()
+        aif_pred_df.scores = df[self.pred_target_col]
+        self.fitted_step = CalibratedEqOddsPostprocessing([{self.sensitive_att: 0}], [{self.sensitive_att: 1}], cost_constraint=self.cost_constraint, seed=self.seed).fit(aif_true_df, aif_pred_df)
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], scores_names=[self.pred_target_col],
+                                             protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
 
-        cur_step = CalibratedEqOddsPostprocessing([{sensitive_att: 0}], [{sensitive_att: 1}], cost_constraint=cost_constraint, seed=seed)
-        super().__init__("@".join(["AIF_CalibratedEqOddsPostprocessing", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=True, clf_threshold=threshold)
+    def name(self):
+        return "Calibrated_eq_odds"
 
-class AIF_RejectOptionPostprocessing(Postprocessor):
+    def abbr_name(self):
+        return "CEQ"
 
-    def __init__(self, df, target_col, sensitive_att, threshold=0.5):
+    def step_name(self):
+        return "FairPostprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class AIF_RejectOptionPostprocessing(Step):
+
+    def __init__(self, target_col, sensitive_att):
         """
         :param df: pandas dataframe, stores the data to fit the postprocessor.
         :param target_col: str, the name of the target variable in above data. Assume 1 represents the favorable class.
         :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
-        :param threshold: float in [0, 1], the classification threshold to generate the predicted class label.
         """
-        # TODO: fix the bug that reject option doesn't return results
-        cur_step = RejectOptionClassification([{sensitive_att: 0}], [{sensitive_att: 1}])
-        super().__init__("@".join(["AIF_RejectOptionClassification", sensitive_att]), cur_step, df, sensitive_att, target_col, input_score=True, clf_threshold=threshold)
+        # TODO: fix the bug that reject option outputs error
 
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
 
+        self.fitted_step = None
+
+    def fit(self, df):
+        aif_true_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]), label_names=[self.target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+        aif_pred_df = aif_true_df.copy()
+        aif_pred_df.scores = df[self.pred_target_col]
+        self.fitted_step = RejectOptionClassification([{self.sensitive_att: 0}], [{self.sensitive_att: 1}]).fit(aif_true_df, aif_pred_df)
+        return self
+
+    def apply(self, df):
+        aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], scores_names=[self.pred_target_col],
+                                         protected_attribute_names=[self.sensitive_att])
+
+        after_aif_df = self.fitted_step.predict(aif_pred_df)
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        after_df[self.pred_target_col] = after_aif_df.labels
+
+        return after_df
+
+    def name(self):
+        return "Reject_option"
+
+    def abbr_name(self):
+        return "RO"
+
+    def step_name(self):
+        return "FairPostprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_post.csv")
-    # cur_o = AIF_RejectOptionPostprocessing(data, "income-per-year", "sex")
-    cur_o = AIF_EqOddsPostprocessing(data, "income-per-year", "sex")
-    # cur_o = AIF_CalibratedEqOddsPostprocessing(data, "income-per-year", "sex")
+    data = pd.read_csv("../../data/german_pre_encoded_after_model.csv")
+    # cur_o = AIF_RejectOptionPostprocessing("credit", "sex")
+    # cur_o = AIF_EqOddsPostprocessing("credit", "sex", 0)
+    cur_o = AIF_CalibratedEqOddsPostprocessing("credit", "sex", 0)
+
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
 
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/german_after_" + cur_o.name() + ".csv", index=False)
diff --git a/pipeline/postprocess/postprocessor.py b/pipeline/postprocess/postprocessor.py
deleted file mode 100644
index f10494c..0000000
--- a/pipeline/postprocess/postprocessor.py
+++ /dev/null
@@ -1,59 +0,0 @@
-"""
-    Super class for all the supported postprocessor classes.
-"""
-import numpy as np
-from aif360.datasets import BinaryLabelDataset
-from pipeline.step import Step
-
-class Postprocessor(Step):
-    def __init__(self, step_name, step, df, sensitive_att, target_col, input_score=True, clf_threshold=0.5):
-        """
-        :param step_name: str, name of the current input step.
-        :param step: object of the initialized class.
-        :param df: pandas dataframe, stores the data.
-        :param sensitive_att: str, the name of a sensitive attribute.
-        :param target_col: str, the name of the target attribute.
-        :param input_score: boolean, represent whether the post-processor takes predicted score as input. Default is True.
-        :param clf_threshold: float in [0, 1], represents the threshold to categorize class labels from predicted scores.
-        """
-        if "pred_"+target_col not in df.columns:
-            print("Require the predictions for ",target_col, " existing in the data!")
-            raise ValueError
-        super().__init__(step_name=step_name, df=df, sensitive_att=sensitive_att, target_col=target_col)
-        # assume the data set has been encoded to numerical values,
-        # intitialize a BinaryLabelDataset from AIF 360
-        aif_true_df = BinaryLabelDataset(df=df.drop(columns=["pred_"+target_col]), label_names=[target_col], protected_attribute_names=[sensitive_att])
-
-        aif_pred_df = aif_true_df.copy()
-
-        if input_score:
-            aif_pred_df.scores = df["pred_"+target_col]
-        else:
-            aif_pred_df.labels = np.array([int(x >= clf_threshold) for x in df["pred_"+target_col]])
-        self.input_score = input_score
-        self.step = step.fit(aif_true_df, aif_pred_df)
-        self.clf_threshold = clf_threshold
-
-
-
-    def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the learned discretizer.
-        :return: pandas dataframe, stores the data after discretize.
-        """
-
-        # initialize AIF360 BinaryLabelDataset
-
-        if self.input_score: # use score prediction to fit model, e.g. RejectOptionClassification, CalibratedEqOddsPostprocessing
-            aif_pred_df = BinaryLabelDataset(df=df, label_names=[self.target_col], scores_names=[self.pred_target_col],
-                                             protected_attribute_names=[self.sensitive_att])
-        else: # use label prediction to fit model, e.g. EqOddsPostprocessing
-            df["pred_label_"+self.target_col] = [int(x >= self.clf_threshold) for x in df[self.pred_target_col]]
-            aif_pred_df = BinaryLabelDataset(df=df.drop(columns=[self.pred_target_col]), label_names=["pred_label_"+self.target_col],
-                                         protected_attribute_names=[self.sensitive_att])
-
-        after_aif_df = self.step.predict(aif_pred_df)
-        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
-        after_df[self.pred_target_col] = after_aif_df.labels
-
-        return after_df
\ No newline at end of file
diff --git a/pipeline/preprocess/categorizers.py b/pipeline/preprocess/categorizers.py
index 83d8d90..f0afb99 100644
--- a/pipeline/preprocess/categorizers.py
+++ b/pipeline/preprocess/categorizers.py
@@ -2,50 +2,138 @@
     Classes to discretize numerical attributes into categorical attributes.
 """
 import pandas as pd
+import numpy as np
 from sklearn.preprocessing import KBinsDiscretizer
 from sklearn.preprocessing import Binarizer
-from pipeline.preprocess.preprocessor import Preprocessor
+from pipeline.step import Step
 
+class NoBinarizer(Step):
+    def __init__(self):
+        self.fitted_step = None
 
-class SK_Discretizer(Preprocessor):
-    def __init__(self, df, num_atts, bin_size, encode='ordinal', strategy='kmeans'):
+    def fit(self, df):
+        pass
+
+    def apply(self, df):
+        return df
+
+    def name(self):
+        return "NoBinarizer"
+
+    def abbr_name(self):
+        return "NB"
+
+    def step_name(self):
+        return "Categorizer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class SK_Discretizer(Step):
+    def __init__(self, num_atts, bin_size, encode='ordinal', strategy='kmeans'):
         """
-        :param df: pandas dataframe, stores the data to fit the discretizer.
         :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
         :param bin_size: list of integer, each integer represents the number of bins to categorize the corresponding numerical attribute.
         :param encode: same parameter with sklearn KBinsDiscretizer
         :param strategy: same parameter with sklearn KBinsDiscretizer
         """
-        cur_step = {}
-        for idx, ai in enumerate(num_atts):
-            cur_step[ai] = KBinsDiscretizer(n_bins=bin_size[idx], encode=encode, strategy=strategy)
+        self.focus_atts = num_atts
         self.bin_size = bin_size
-        super().__init__("@".join([strategy+"Categorizer"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+        self.encode = encode
+        self.strategy = strategy
+
+        self.fitted_step = None
+
+    def fit(self, df):
+        fitted_step = {}
+        for idx, ai in enumerate(self.focus_atts):
+            fitted_step[ai] = KBinsDiscretizer(n_bins=self.bin_size[idx], encode=self.encode, strategy=self.strategy).fit(np.array(df[ai]).reshape(-1, 1))
+        self.fitted_step = fitted_step
+        return self
+
+    def apply(self, df):
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
+
+    def name(self):
+        return "SK_Discretizer"
+
+    def abbr_name(self):
+        return "DS"
+
+    def step_name(self):
+        return "Categorizer"
 
+    def input_encoded_data(self):
+        return False
 
-class SK_Binarizer(Preprocessor):
-    def __init__(self, df, num_atts, bin_thresholds, copy=True):
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class SK_Binarizer(Step):
+    def __init__(self, num_atts, bin_thresholds, copy=True):
         """
-        :param df: pandas dataframe, stores the data to fit the binarizer.
         :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
         :param bin_thresholds: list of float, each float represents the value to binarize the corresponding numerical attributes.
                                Values below or equal to this threshold are replaced by 0, above it by 1.
         :param copy: same parameter with sklearn Binarizer
         """
-        cur_step = {}
-        for idx, ai in enumerate(num_atts):
-            cur_step[ai] = Binarizer(threshold=bin_thresholds[idx], copy=copy)
-
+        self.focus_atts = num_atts
         self.bin_thresholds = bin_thresholds
-        super().__init__("@".join(["BinaryCategorizer"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+        self.copy = copy
+
+        self.fitted_step = None
+
+    def fit(self, df):
+        fitted_step = {}
+        for idx, ai in enumerate(self.focus_atts):
+            fitted_step[ai] = Binarizer(threshold=self.bin_thresholds[idx], copy=self.copy).fit(np.array(df[ai]).reshape(-1, 1))
+        self.fitted_step = fitted_step
+        return self
+
+    def apply(self, df):
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
 
+    def name(self):
+        return "SK_Binarizer"
+
+    def abbr_name(self):
+        return "BI"
+
+    def step_name(self):
+        return "Categorizer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult.csv")
-    cur_o = SK_Discretizer(data, ["fnlwgt", "age"], [2, 3])
-    # cur_o = SK_Binarizer(data, ["fnlwgt", "age"], [100000, 30])
 
-    after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+    data = pd.read_csv("../../data/german.csv")
+    # cur_o = SK_Discretizer(["month", "age"], [2, 3])
+    cur_o = SK_Binarizer(["month", "age"], [24, 30])
 
-    print(cur_o.get_name())
\ No newline at end of file
+    cur_o.fit(data)
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/german_"+cur_o.name()+".csv", index=False)
diff --git a/pipeline/preprocess/encoders.py b/pipeline/preprocess/encoders.py
index b2a6712..e81ab60 100644
--- a/pipeline/preprocess/encoders.py
+++ b/pipeline/preprocess/encoders.py
@@ -1,98 +1,144 @@
 """
     Classes to encode the string values for categorical attributes.
 """
-"""
-    Classes to discretize numerical attributes into categorical attributes.
-"""
+
 import numpy as np
 import pandas as pd
 from sklearn.preprocessing import OrdinalEncoder
-from sklearn.preprocessing import OneHotEncoder
-from pipeline.preprocess.preprocessor import Preprocessor
+from pipeline.step import Step
 
-class SK_OrdinalEncoder(Preprocessor):
-    def __init__(self, df, cate_atts, sort_label, sort_positive_value):
+class SK_OrdinalEncoder(Step):
+    def __init__(self, cate_atts, sort_label, sort_positive_value):
         """
-        :param df: pandas dataframe, stores the data to fit the encoder.
         :param cate_atts: list of str, each str represents the name of a categorical attribute in above data.
-        :param encode_order_dict: dict, key (str) represents the name of categorical attribute, value is a list of str, representing the ordered categories of each each categorical attribute.
         :param sort_label: str, name of the target varible to determine the order of ordinal encodings.
+        :param sort_positive_value: str, value of the target varible to determine the order of ordinal encodings.
+
         """
-        cur_step = {}
-        for ci in cate_atts:
-            value_counts = {}
-            for vi in df[ci].unique():
-                value_counts[vi] = df[(df[ci] == vi) & (df[sort_label] == sort_positive_value)].shape[0]
-            value_orders = sorted(value_counts.keys(), key=lambda x: value_counts[x])
-            cur_step[ci] = OrdinalEncoder(categories=[value_orders])
 
+        self.focus_atts = cate_atts
         self.sort_label = sort_label
         self.sort_positive_value = sort_positive_value
-        super().__init__("@".join(["OrdinalEncoder"]+cate_atts), df, step=cur_step, focus_atts=cate_atts, fit_flag=True)
 
+        self.fitted_step = None
+
+    def fit(self, df):
+        fitted_step = {}
+        for idx, ai in enumerate(self.focus_atts):
+            value_counts = {}
+            for vi in df[ai].unique():
+                value_counts[vi] = df[(df[ai] == vi) & (df[self.sort_label] == self.sort_positive_value)].shape[0]
+            value_orders = sorted(value_counts.keys(), key=lambda x: value_counts[x])
+            fitted_step[ai] = OrdinalEncoder(categories=[value_orders]).fit(np.array(df[ai]).reshape(-1, 1))
+        self.fitted_step = fitted_step
+        return self
+
+    def apply(self, df):
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
+
+    def name(self):
+        return "SK_OrdinalEncoder"
+
+    def abbr_name(self):
+        return "OE"
+
+    def step_name(self):
+        return "Encoder"
 
-class SK_OneHotEncoder(Preprocessor):
-    def __init__(self, df, cate_atts):
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return True
+
+    def fit_only_on_train(self):
+        return True
+
+
+class OneHotEncoder(Step):
+    def __init__(self, cate_atts):
         """
-        :param df: pandas dataframe, stores the data to fit the encoder.
         :param cate_atts: list of str, each str represents the name of a categorical attribute in above data.
-        :param encode_order_dict: dict, key (str) represents the name of categorical attribute, value is a list of str, representing the ordered categories of each each categorical attribute.
-        :param sort_label: str, name of the target variable to determine the order of ordinal encodings.
         """
-        # TODO: fix the bug that sklearn one-hot encoder change the dimension
-        # cur_step = {}
-        # for ci in cate_atts:
-        #     cur_step[ci] = OneHotEncoder()
-        super().__init__("@".join(["OneHotEncoder"]+cate_atts), df, step=None, focus_atts=cate_atts, fit_flag=False)
+        self.focus_atts = cate_atts
+
+    def fit(self, df):
+        pass
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the learned encoder.
-        :return: pandas dataframe, stores the data after encode.
-        """
         after_df = pd.get_dummies(df, columns=self.focus_atts, prefix_sep='=')
-        # after_df = df[list(set(df.columns).difference(self.focus_atts))]
-        # for ci in self.focus_atts:
-        #     ci_encode_array = self.step[ci].transform(np.array(df[ci]).reshape(-1, 1)).toarray()
-        #     ci_encode_df = pd.DataFrame(ci_encode_array, columns=[ci+"="+x for x in self.step[ci].categories_[0]])
-        #     after_df = pd.concat([after_df, ci_encode_df], axis=1)
         return after_df
 
-class CustomCateAttsEncoder(Preprocessor):
-    def __init__(self, df, sensitive_atts, protected_values):
-        """ To encode sensitive attribute and target feature.
-        :param df: pandas dataframe, stores the data to fit the encoder.
-        :param sensitive_atts: list of str, each str represents the name of a sensitive attribute.
-        :param protected_values: dict, key is the str in sensitive_atts, value is a list of str, each str represent the protected values for the key sensitive attribute.
+    def name(self):
+        return "OneHotEncoder"
+
+    def abbr_name(self):
+        return "HE"
+
+    def step_name(self):
+        return "Encoder"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return True
+
+    def fit_only_on_train(self):
+        return False
+
+# That can only apply in SensitiveAttributeEncoder
+class MappingEncoder(Step):
+    def __init__(self, focus_atts, mapping_dict):
+        """ Encode sensitive attribute and target feature through mapping string to numberical values according to the input dictionary.
+        :param focus_atts: list of str, each str represents the name of an attribute.
+        :param mapping_dict: dict, key is the value (str) of the attribute in focus_atts, value is int that encode the value. E.g. {"female": 0, "male": 1} for the values of attribute gender.
         """
-        super().__init__("@".join(["SensitiveAttEncoder"]+sensitive_atts), df, step=None, focus_atts=sensitive_atts, fit_flag=False)
-        for x in sensitive_atts:
-            if sum([vi not in df[x].unique() for vi in protected_values[x]]) > 0:
-                print("Some input values of sensitive attribute ", x, " are not valid!")
-                raise ValueError
-        self.protected_values = protected_values
+
+        self.focus_atts = focus_atts
+        self.mapping_dict = mapping_dict
+
+    def fit(self, df):
+        pass
+
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the encoder.
-        :return: pandas dataframe, stores the data after encode.
-        """
         after_df = df.copy()
         for si in self.focus_atts:
-            after_df[si] = after_df[si].apply(lambda x: int(x not in self.protected_values[si]))
+            after_df[si] = after_df[si].apply(lambda x: self.mapping_dict[x])
         return after_df
 
+    def name(self):
+        return "MappingEncoder"
+
+    def abbr_name(self):
+        return "ME"
+
+    def step_name(self):
+        return "SpecialEncoder"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return True
+
+    def fit_only_on_train(self):
+        return False
 
 
 if __name__ == '__main__':
-    data = pd.read_csv("../data/train/adult__Imputer.csv")
-    # data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
-    # data = pd.read_csv("../../data/adult_pre_SensitiveAttsEncoder_sex_race_income-per-year.csv")
-    # cur_o = SK_OrdinalEncoder(data, ["sex", "race"], "income-per-year", ">50K")
-    cur_o = SK_OneHotEncoder(data, ["workclass", "education", "marital-status", "occupation", "relationship", "native-country"])
-    # cur_o = CustomCateAttsEncoder(data, ["sex", "race", "income-per-year"], {"sex": ["Female"], "race": ["Black"], "income-per-year": ["<=50K"]})
+    data = pd.read_csv("../../data/german_AIF.csv")
 
-    after_data = cur_o.apply(data)
-    after_data.to_csv("../data/adult_"+cur_o.get_name()+".csv", index=False)
+    cur_o = SK_OrdinalEncoder(["sex"], "credit", "good")
+    # cur_o = OneHotEncoder(["sex", "credit", "age"])
 
-    print(cur_o.get_name())
\ No newline at end of file
+    # cur_o = MappingEncoder(["sex", "credit"], {"female": 0, "male": 1, "good": 1, "bad": 0})
+
+    cur_o.fit(data)
+    after_data = cur_o.apply(data)
+    after_data.to_csv("../../data/german_AIF_"+cur_o.name()+".csv", index=False)
diff --git a/pipeline/preprocess/fair_preprocessors.py b/pipeline/preprocess/fair_preprocessors.py
index aeb519f..908ad19 100644
--- a/pipeline/preprocess/fair_preprocessors.py
+++ b/pipeline/preprocess/fair_preprocessors.py
@@ -2,39 +2,75 @@
     Class of fairness preprocessing interventions
 """
 
-import numpy as np
 import pandas as pd
 from aif360.datasets import BinaryLabelDataset
 from aif360.algorithms.preprocessing import Reweighing as Reweighing
 from aif360.algorithms.preprocessing import LFR as LFR
 from aif360.algorithms.preprocessing import DisparateImpactRemover
-from pipeline.preprocess.preprocessor import Preprocessor
+from pipeline.step import Step
 
-class AIF_Reweighing(Preprocessor):
+class AIF_Reweighing(Step):
 
-    def __init__(self, df, target_col, sensitive_att):
+    def __init__(self, target_col, sensitive_att):
         """
-        :param df: pandas dataframe, stores the data to fit the scaler.
         :param target_col: str, the name of the target variable in above data.
-        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
-        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param sensitive_att: str, the name of a sensitive attribute in above data.
 
         """
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
 
-        cur_step = Reweighing([{sensitive_att: 0}], [{sensitive_att: 1}])
-        super().__init__("@".join(["AIF_Reweighing", sensitive_att]), df, step=cur_step, fit_flag=True, weight_flag=True, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
+        self.fitted_step = None
 
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        self.fitted_step = Reweighing([{self.sensitive_att: 0}], [{self.sensitive_att: 1}]).fit(aif_df)
+        return self
 
-class AIF_DIRemover(Preprocessor):
+    def apply(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        after_aif_df = self.fitted_step.transform(aif_df)
+        # TODO: double check whether to return weights
+        preprocessed_weights = after_aif_df.instance_weights
 
-    def __init__(self, df, target_col, sensitive_att, repair_level):
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        
+        return after_df
+
+    def name(self):
+        return "Reweighing"
+
+    def abbr_name(self):
+        return "RW"
+
+    def step_name(self):
+        return "FairPreprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class AIF_DIRemover(Step):
+    def __init__(self, target_col, sensitive_att, repair_level):
         """
-        :param df: pandas dataframe, stores the data to fit the scaler.
         :param target_col: str, the name of the target variable in above data.
-        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
-        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param sensitive_att: str, the name of a sensitive attribute in above data.
 
         """
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
+
+        self.fitted_step = None
+
         if repair_level is None or not isinstance(repair_level, float):
             print("Input repair_level is not valid! Should be float within [0,1]!")
             raise ValueError
@@ -42,41 +78,109 @@ def __init__(self, df, target_col, sensitive_att, repair_level):
             if repair_level < 0 or repair_level > 1:
                 print("Input repair_level is not valid! Should be float within [0,1]!")
                 raise ValueError
+
         self.repair_level = repair_level
-        cur_step = DisparateImpactRemover(repair_level=repair_level, sensitive_attribute=sensitive_att)
 
-        super().__init__("@".join(["AIF_DIRemover", sensitive_att]), df, step=cur_step, fit_flag=False, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
 
+    def fit(self, df):
+        # TODO: Rewrite the AIF DIRemover to afford the separation of train, validation, and test.
+        # aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        # self.fitted_step = DisparateImpactRemover(repair_level=self.repair_level, sensitive_attribute=self.sensitive_att).fit(aif_df)
+        #
+        pass
+
+    def apply(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        after_aif_df = DisparateImpactRemover(repair_level=self.repair_level, sensitive_attribute=self.sensitive_att).fit_transform(aif_df)
+
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        return after_df
+
+    def name(self):
+        return "DIRemover"
+
+    def abbr_name(self):
+        return "RI"
 
-class AIF_LFR(Preprocessor):
+    def step_name(self):
+        return "FairPreprocessor"
 
-    def __init__(self, df, target_col, sensitive_att):
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class AIF_LFR(Step):
+
+    def __init__(self, target_col, sensitive_att):
         """ NOTE: very sensitive to input data, refer the example in AIF 360 for this preprocessor
-        :param df: pandas dataframe, stores the data to fit the scaler.
         :param target_col: str, the name of the target variable in above data.
-        :param target_positive_value: str, the value of above target variable that represents positive outcome. default is 1.
-        :param sensitive_att: str, the name of a sensitive attribute in above data. If none, call auto_detection to update. Value 0 represent protected.
+        :param sensitive_att: str, the name of a sensitive attribute in above data.
 
         """
         # TODO: fix the bug of LFR for not returning categorical atts
         # TODO: experiment with the same data used by AIF360 tutorial to compare whether the categorical atts are returned
-        cur_step = LFR([{sensitive_att: 0}], [{sensitive_att: 1}])
-        super().__init__("@".join(["AIF_LFR", sensitive_att]), df, step=cur_step, fit_flag=True, sensitive_att=sensitive_att, target_col=target_col, fair_aware=True)
+        self.sensitive_att = sensitive_att
+        self.target_col = target_col
+        self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
 
+        self.fitted_step = None
 
+    def fit(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        self.fitted_step = LFR([{self.sensitive_att: 0}], [{self.sensitive_att: 1}]).fit(aif_df)
+        return self
 
-if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
-    # cur_o = AIF_Reweighing(data, "income-per-year", "sex")
-    # cur_o = AIF_LFR(data, "income-per-year", "sex")
-    cur_o = AIF_DIRemover(data, "income-per-year", "sex", 0.8)
+    def apply(self, df):
+        # wrap the input dataframe with AIF 360 object "BinaryLabelDataset"
+        aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
+        after_aif_df = self.fitted_step.transform(aif_df)
+
+        after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
+        return after_df
+
+    def name(self):
+        return "LFR"
 
+    def abbr_name(self):
+        return "LR"
+
+    def step_name(self):
+        return "FairPreprocessor"
+
+    def input_encoded_data(self):
+        return True
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+if __name__ == '__main__':
+    data = pd.read_csv("../../data/german_pre_encoded.csv")
+    # name_mapping = {"female": 0, "male": 1, "young": 0, "old": 1, "bad": 0, "good": 1}
+    # for atti in ["credit", "sex", "age"]:
+    #     data[atti] = data[atti].apply(lambda x: name_mapping[x])
+    #
+    # data.to_csv("../../data/german_pre_encoded.csv")
+
+    # cur_o = AIF_Reweighing("credit", "sex")
+    # cur_o = AIF_LFR("credit", "sex") # TODO: bug not working for this dataset, test another dataset
+    cur_o = AIF_DIRemover("credit", "sex", 0.8)
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    # for Reweighing
+    # # for Reweighing
     # after_data, new_weights = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
+    after_data.to_csv("../../data/german_after_"+cur_o.name()+".csv", index=False)
 
-    print(cur_o.get_name())
+    # print(cur_o.get_name())
     # for Reweighing
     # print(len(new_weights))
     # print(new_weights)
\ No newline at end of file
diff --git a/pipeline/preprocess/filters.py b/pipeline/preprocess/filters.py
index b2a4426..264deb4 100644
--- a/pipeline/preprocess/filters.py
+++ b/pipeline/preprocess/filters.py
@@ -2,68 +2,93 @@
     Classes to filter slice of data.
 """
 import pandas as pd
-from pipeline.preprocess.preprocessor import Preprocessor
-
-# utility functions
-def wrap_filter(att, value):
-    if isinstance(value, str): # string or null
-        if value in ["?", ""]: # for null
-            return '{}!={}'.format(att, att)
-        else:
-            return '{}=="{}"'.format(att, value)
-    else: # numerical value
-        return '{}=={}'.format(att, value)
+from pipeline.step import Step
+
 
 # TODO: add multiple filter class
-class RowFilter(Preprocessor):
+class RowFilter(Step):
 
-    def __init__(self, df, column, value):
+    def __init__(self, column, value):
         """
-        :param column: str, name of the column name to be filtered.
-        :param value: str, integer, float, the value of the column to be filtered.
+        :param column: str, name of the column name to be focused.
+        :param value: str, integer, float, the value of the column to keep.
         """
-        if column is None or value is None:
-            print("Need to specify column and value to create filter!")
-            raise ValueError
-        if column not in df.columns:
-            print("Need to specify valid column!")
-            raise ValueError
-        if value not in df[column].unique():
-            print("Need to specify valid value!")
-            raise ValueError
+
         self.column = column
         self.value = value
-        super().__init__(step_name="@".join(["RowFilter", column, str(value)]), df=df, focus_atts=[column], fit_flag=False)
+
+    def fit(self, df):
+        pass
+
+    # utility functions to wrap the input filter to a df query
+    def wrap_filter(self):
+        if isinstance(self.value, str):  # string or null
+            if self.value in ["?", ""]:  # for null
+                return '{}!={}'.format(self.column, self.column)
+            else:
+                return '{}=="{}"'.format(self.column, self.value)
+        else:  # numerical value
+            return '{}=={}'.format(self.column, self.value)
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to be filtered.
-        :return: pandas dataframe, stores the data after filter.
-        """
-        return df.query(wrap_filter(self.column, self.value))
+        return df.query(self.wrap_filter())
+
+    def name(self):
+        return "RowFilter"
+
+    def abbr_name(self):
+        return "RF"
+
+    def step_name(self):
+        return "Filter"
+
+    def input_encoded_data(self):
+        return False
 
-class RemoveColumnFilter(Preprocessor):
-    def __init__(self, df, exclude_cols):
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class ColumnFilter(Step):
+    def __init__(self, remove_cols):
         """
-        :param exclude_cols: list of string, each string represents the name of the column to be excluded.
+        :param exclude_cols: list of string, each string represents the name of the column to be filtered out.
         """
+        self.remove_cols = remove_cols
 
-        super().__init__(step_name="RemoveColumnFilter", df=df, focus_atts=exclude_cols, fit_flag=False)
+    def fit(self, df):
+        pass
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to be filtered.
-        :return: pandas dataframe, stores the data after filter.
-        """
-        return df.drop(columns=self.focus_atts)
+        return df.drop(columns=self.remove_cols)
+
+    def name(self):
+        return "ColumnFilter"
+
+    def abbr_name(self):
+        return "CF"
+
+    def step_name(self):
+        return "Filter"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
 
 if __name__ == '__main__':
 
-    data = pd.read_csv("../../data/adult_pre_reweigh.csv")
-    # cur_o = RowFilter(data, "sex", 0)
-    cur_o = RemoveColumnFilter(data, ["sex","race"])
+    data = pd.read_csv("../../data/german_AIF.csv")
+    cur_o = RowFilter("sex", "female")
+    # cur_o = ColumnFilter(["sex", "age"])
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_" + cur_o.get_name() + ".csv", index=False)
+    after_data.to_csv("../../data/german_AIF_" + cur_o.name() + ".csv", index=False)
 
-    print(cur_o.get_name())
diff --git a/pipeline/preprocess/imputers.py b/pipeline/preprocess/imputers.py
index 36aee84..66cfc57 100644
--- a/pipeline/preprocess/imputers.py
+++ b/pipeline/preprocess/imputers.py
@@ -4,65 +4,143 @@
 import numpy as np
 import pandas as pd
 import datawig
-from pipeline.preprocess.preprocessor import Preprocessor
 from sklearn.impute import SimpleImputer
+from pipeline.step import Step
 
-class DropNAImputer(Preprocessor):
-    def __init__(self, df, na_mark=None):
+class NoImputer(Step):
+    def __init__(self):
+        self.fitted_step = None
+
+    def fit(self, df):
+        pass
+
+    def apply(self, df):
+        return df
+
+    def name(self):
+        return "NoImputer"
+
+    def abbr_name(self):
+        return "NI"
+
+    def step_name(self):
+        return "Imputer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class DropNAImputer(Step):
+    def __init__(self, na_mark=None):
         """
-        :param df: pandas dataframe, stores the data to fit the imputer.
         :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
         """
-        super().__init__("DropNAImputer", df=df, fit_flag=False, na_mark=na_mark)
+        self.na_mark = na_mark
+
+    def fit(self, df):
+        pass
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to impute.
-        :return: pandas dataframe, stores the data after impute.
-        """
         if self.na_mark:
             df = df.replace({self.na_mark:np.nan})
         return df.dropna()
 
-class ModeImputer(Preprocessor):
-    def __init__(self, df, num_atts, cate_atts, na_mark=None):
+    def name(self):
+        return "DropNAImputer"
+
+    def abbr_name(self):
+        return "DN"
+
+    def step_name(self):
+        return "Imputer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class ModeImputer(Step):
+    def __init__(self, num_atts, cate_atts, na_mark=None):
         """
-        :param df: pandas dataframe, stores the data to fit the imputer.
         :param num_atts: list of str, each str represents the name of numerical column to be imputed using the mean value.
         :param cate_atts: list of str, each str represents the name of categorical column to be imputed using the most frequent value.
         :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
         """
-        if len(set(num_atts).intersection(cate_atts)) > 0:
-            print("Some attributes are both in num_atts and cate_atts!")
-            raise ValueError
+        self.num_atts = num_atts
+        self.cate_atts = cate_atts
+        self.na_mark = na_mark
+
+        self.fitted_step = None
 
-        cur_step = {}
-        if len(cate_atts) > 0:
-            for ci in cate_atts:
-                cur_step[ci] = SimpleImputer(strategy='most_frequent')
-        if len(num_atts) > 0:
-            for ni in num_atts:
-                cur_step[ni] = SimpleImputer(strategy='mean')
+    def fit(self, df):
+        fitted_step = {}
+        if len(self.cate_atts) > 0:
+            for ci in self.cate_atts:
+                fitted_step[ci] = SimpleImputer(strategy='most_frequent').fit(np.array(df[ci]).reshape(-1, 1))
+        if len(self.num_atts) > 0:
+            for ni in self.num_atts:
+                fitted_step[ni] = SimpleImputer(strategy='mean').fit(np.array(df[ni]).reshape(-1, 1))
+        self.fitted_step = fitted_step
 
-        super().__init__("@".join(["ModeImputer"]+num_atts+cate_atts), df, step=cur_step, focus_atts=cate_atts+num_atts, fit_flag=True, na_mark=na_mark)
+        return self
 
+    def apply(self, df):
+        if self.na_mark:
+            df = df.replace({self.na_mark:np.nan})
+        after_df = df.copy()
+        for ai in self.num_atts + self.cate_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
+
+    def name(self):
+        return "SK_ModeImputer"
+
+    def abbr_name(self):
+        return "MI"
+
+    def step_name(self):
+        return "Imputer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
 
-class DatawigImputer(Preprocessor):
-    def __init__(self, df, impute_atts, na_mark=None, output_path="datawig/", num_epochs=50):
+    def fit_only_on_train(self):
+        return True
+
+class DatawigImputer(Step):
+    def __init__(self, impute_atts, na_mark=None, output_path="datawig/", num_epochs=50):
         """
-        :param df: pandas dataframe, stores the data to fit the imputer.
         :param impute_atts: list of str, each str represents the name of column to be imputed using datawig model. Column can be categorical or numerical.
         :param na_mark: str, represents the symbol of missing values. Default is None, i.e. NaN represents the missing values.
         :param output_path: str, the path to store the learned datawig model.
         :param num_epochs: integer, the maximum iteration of datawig model.
         """
-        super().__init__("@".join(["DatawigImputer"] + impute_atts), df, focus_atts=impute_atts, fit_flag=False, na_mark=na_mark)
+        self.focus_atts = impute_atts
+        self.na_mark = na_mark
+        self.output_path = output_path
+        self.num_epochs = num_epochs
+
+        self.fitted_step = None
 
+    def fit(self, df):
         learned_imputers = {}
-        for ai in impute_atts:
-            learned_imputers[ai] = datawig.SimpleImputer(input_columns=list(set(df.columns).difference(ai)),
-                                                          output_column=ai, output_path=output_path).fit(train_df=df, num_epochs=num_epochs)
-        self.step = learned_imputers
+        for ai in self.focus_atts:
+            learned_imputers[ai] = datawig.SimpleImputer(input_columns=list(set(df.columns).difference(ai)), output_column=ai, output_path=self.output_path).fit(train_df=df, num_epochs=self.num_epochs)
+        self.fitted_step = learned_imputers
+        return self
 
     def apply(self, df):
         """
@@ -73,16 +151,33 @@ def apply(self, df):
             df = df.replace({self.na_mark:np.nan})
         after_df = df.copy()
         for ai in self.focus_atts:
-            after_df[ai] = self.step[ai].predict(df)[ai + '_imputed']
+            after_df[ai] = self.fitted_step[ai].predict(df)[ai + '_imputed']
         return after_df
 
+    def name(self):
+        return "DatawigImputer"
+
+    def abbr_name(self):
+        return "DW"
+
+    def step_name(self):
+        return "Imputer"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
-    cur_o = DropNAImputer(data, na_mark="?")
-    # cur_o = ModeImputer(data, ["fnlwgt"], ["workclass"], na_mark="?")
-    # cur_o = DatawigImputer(data, ["workclass"], na_mark="?")
+    data = pd.read_csv("../../data/adult_AIF.csv")
+    # cur_o = DropNAImputer(na_mark="?")
+    cur_o = ModeImputer([],["workclass"], na_mark="?")
+    # cur_o = DatawigImputer(["workclass"], na_mark="?") # TODO: test after fix dependency issue
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
-
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/adult_AIF_"+cur_o.name()+".csv", index=False)
diff --git a/pipeline/preprocess/preprocessor.py b/pipeline/preprocess/preprocessor.py
deleted file mode 100644
index 6c5ffae..0000000
--- a/pipeline/preprocess/preprocessor.py
+++ /dev/null
@@ -1,71 +0,0 @@
-"""
-    Super class for all the supported preprocessor classes.
-"""
-import numpy as np
-from aif360.datasets import BinaryLabelDataset
-from pipeline.step import Step
-
-class Preprocessor(Step):
-    def __init__(self, step_name, df, step=None, focus_atts=[], fit_flag=True, weight_flag=False, sensitive_att=None, target_col=None, fair_aware=False, na_mark=None):
-        """
-        :param step_name: str, name of the current input step.
-        :param df: pandas dataframe, stores the data.
-        :param step: object of the initialized class. If none, initialize here.
-        :param focus_atts: lisf of str, each str represents the name of a column in above data that will be pre-processed.
-        :param fit_flag: boolean, whether to initialize step object here.
-        :param weight_flag: boolean, whether to output extra sample weight after fair-preprocessor.
-        :param sensitive_att: str, the name of a sensitive attribute.
-        :param target_col: str, the name of the target attribute.
-        :param fair_aware: boolean, whether the preprocessor is fair-aware. Default is False. If true, sensitive_att and target_col can not be null.
-        """
-        super().__init__(step_name=step_name, df=df, focus_atts=focus_atts, sensitive_att=sensitive_att, target_col=target_col)
-
-        if len(focus_atts) > 0 and fit_flag:
-            fitted_step = {}
-            for idx, ai in enumerate(focus_atts):
-                fitted_step[ai] = step[ai].fit(np.array(df[ai]).reshape(-1, 1))
-            self.step = fitted_step
-        elif fair_aware and fit_flag: # for fair-preprocessors
-            aif_df = BinaryLabelDataset(df=df, label_names=[target_col], protected_attribute_names=[sensitive_att])
-            self.step = step.fit(aif_df)
-        else:
-            if step is not None:
-                self.step = step
-
-        # address different encoding of missing values
-        if na_mark is not None:
-            self.na_mark = na_mark
-        else:
-            self.na_mark = None
-        self.fair_aware = fair_aware
-        self.fit_flag = fit_flag
-        self.weight_flag = weight_flag
-
-
-    def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the learned discretizer.
-        :return: pandas dataframe, stores the data after discretize.
-        """
-        if self.na_mark:
-            df = df.replace({self.na_mark:np.nan})
-        if self.fair_aware: # fair-preprocessor
-            aif_df = BinaryLabelDataset(df=df, label_names=[self.target_col], protected_attribute_names=[self.sensitive_att])
-            if self.fit_flag: # fit has been initialized
-                after_aif_df = self.step.transform(aif_df)
-            else: # fit and transform is combined, e.g. DisparateImpactRemover
-                after_aif_df = self.step.fit_transform(aif_df)
-
-            after_df, _ = after_aif_df.convert_to_dataframe(de_dummy_code=True, sep='=', set_category=True)
-            if self.weight_flag:
-                preprocessed_weights = after_aif_df.instance_weights
-
-        else: # regular preprocessor
-            after_df = df.copy()
-            for ai in self.focus_atts:
-                after_df[ai] = self.step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
-
-        if self.weight_flag: # for the preprocessor that updates weights, e.g. Reweighing
-            return after_df, preprocessed_weights
-        else:
-            return after_df
\ No newline at end of file
diff --git a/pipeline/preprocess/samplers.py b/pipeline/preprocess/samplers.py
index e95ea2d..d19e390 100644
--- a/pipeline/preprocess/samplers.py
+++ b/pipeline/preprocess/samplers.py
@@ -4,58 +4,92 @@
 
 import numpy as np
 import pandas as pd
-from pipeline.preprocess.preprocessor import Preprocessor
+from pipeline.step import Step
 
-class RandomSampler(Preprocessor):
-    def __init__(self, df, sample_n, random_state=0):
+
+class NoSampler(Step):
+    def __init__(self):
+        self.fitted_step = None
+
+    def fit(self, df):
+        pass
+
+    def apply(self, df):
+        return df
+
+    def name(self):
+        return "NoSampler"
+
+    def abbr_name(self):
+        return "NS"
+
+    def step_name(self):
+        return "Sampler"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+class RandomSampler(Step):
+    def __init__(self, sample_n, seed):
         """
         :param sample_n: integer, the size of the sampled subset of data
+        :param seed: integer, the seed for random process.
         """
-        if not sample_n:
-            print("Need to specify a size greater than 0!")
-            raise ValueError
         self.sample_n = sample_n
-        self.random_state = random_state
-        super().__init__("RandomSampler@"+str(sample_n), df=df, fit_flag=False)
+        self.seed = seed
+
+    def fit(self, df):
+        pass
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to be sampled.
-        :return: pandas dataframe, stores the data after sample.
-        """
-        return df.sample(n=self.sample_n, random_state=self.random_state)
+        return df.sample(n=self.sample_n, random_state=self.seed)
+
+    def name(self):
+        return "RandomSampler"
+
+    def abbr_name(self):
+        return "RS"
+
+    def step_name(self):
+        return "Sampler"
 
-class BalancePopulationSampler(Preprocessor):
-    def __init__(self, df, sample_n, balance_col, random_state=0):
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
+
+class BalancePopulationSampler(Step):
+    def __init__(self, sample_n, balance_col, seed):
         """
         :param sample_n: integer, the size of the sampled subset of data
         :param balance_col: str, the name of a categorical column that the population of groups within this column are balanced in the sampled subset.
-        :param random_state: integer, the seed for random process, same as random_state in pandas.DataFrame.sample.
+        :param seed: integer, the seed for random process.
 
         """
-        if not sample_n:
-            print("Need to specify a size greater than 0!")
-            raise ValueError
-        if not balance_col:
-            print("Need to specify the name of a column to perform balance sampling within this column!")
-            raise ValueError
-        if balance_col not in df.columns:
-            print("Need to specify a valid column to perform balance sampling within this column!")
-            raise ValueError
+
         self.sample_n = sample_n
         self.balance_col = balance_col
-        self.random_state = random_state
-        super().__init__("@".join(["BalanceSampler", balance_col, str(sample_n)]), df=df, fit_flag=False)
+        self.seed = seed
+
+    def fit(self, df):
+        pass
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to be sampled.
-        :return: pandas dataframe, stores the data after sample.
-        """
-        # TODO: update to minimum sample set
+        # TODO: update to minimum sample set and remove print statement
         balance_groups = list(df[self.balance_col].unique())
         n_group = int(np.ceil(self.sample_n/len(balance_groups)))
-        # print(n_group)
+
         sampled_df = {}
         small_groups = []
         for gi in balance_groups:
@@ -64,7 +98,7 @@ def apply(self, df):
                 sampled_df[gi] = gi_data
                 small_groups.append(gi)
             else:
-                sampled_df[gi] = df[df[self.balance_col]==gi].sample(n=n_group, random_state=self.random_state)
+                sampled_df[gi] = df[df[self.balance_col]==gi].sample(n=n_group, random_state=self.seed)
 
         after_df = pd.DataFrame()
         if not self.sample_n % len(balance_groups): # for even groups
@@ -86,11 +120,31 @@ def apply(self, df):
         print(after_df.groupby(self.balance_col).count())
         return after_df
 
+    def name(self):
+        return "BalanceSampler"
+
+    def abbr_name(self):
+        return "BS"
+
+    def step_name(self):
+        return "Sampler"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return False
+
 if __name__ == '__main__':
-    # cur_o = BalancePopulationSampler(1000, "marital-status")
-    data = pd.read_csv("../../data/adult.csv")
-    cur_o = RandomSampler(data, 1000)
 
+    data = pd.read_csv("../../data/german_AIF.csv")
+    # cur_o = RandomSampler(200, 0)
+
+    cur_o = BalancePopulationSampler(200, "sex", 0)
+
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_" + cur_o.get_name() + ".csv", index=False)
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/german_AIF_" + cur_o.name() + ".csv", index=False)
\ No newline at end of file
diff --git a/pipeline/preprocess/scalers.py b/pipeline/preprocess/scalers.py
index 7a1e80f..8ad585a 100644
--- a/pipeline/preprocess/scalers.py
+++ b/pipeline/preprocess/scalers.py
@@ -2,46 +2,113 @@
     Classes to scale data.
 """
 import pandas as pd
+import numpy as np
 from sklearn.preprocessing import StandardScaler, MinMaxScaler
-from pipeline.preprocess.preprocessor import Preprocessor
+from pipeline.step import Step
 
-class SK_StandardScaler(Preprocessor):
+class SK_StandardScaler(Step):
 
-    def __init__(self, df, num_atts, copy=True, with_mean=True, with_std=True):
+    def __init__(self, num_atts, copy=True, with_mean=True, with_std=True):
         """
-        :param df: pandas dataframe, stores the data to fit the scaler.
         :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
         :param copy: same parameter with sklearn StandardScaler
         :param with_mean: same parameter with sklearn StandardScaler
         :param with_std: same parameter with sklearn StandardScaler
         """
-        cur_step = {}
-        for ai in num_atts:
-            cur_step[ai] = StandardScaler(copy=copy, with_mean=with_mean, with_std=with_std)
+        self.focus_atts = num_atts
+        self.copy = copy
+        self.with_mean = with_mean
+        self.with_std = with_std
 
-        super().__init__("@".join(["StandardScaler"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+        self.fitted_step = None
 
+    def fit(self, df):
+        fitted_step = {}
+        for ai in self.focus_atts:
+            fitted_step[ai] = StandardScaler(copy=self.copy, with_mean=self.with_mean, with_std=self.with_std).fit(np.array(df[ai]).reshape(-1, 1))
 
-class SK_MinMaxScaler(Preprocessor):
-    def __init__(self, df, num_atts, feature_range=(0, 1), copy=True):
+        self.fitted_step = fitted_step
+
+        return self
+
+    def apply(self, df):
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
+
+    def name(self):
+        return "StandardScaler"
+
+    def abbr_name(self):
+        return "SS"
+
+    def step_name(self):
+        return "Scaler"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+
+class SK_MinMaxScaler(Step):
+    def __init__(self, num_atts, feature_range=(0, 1), copy=True):
         """
-        :param df: pandas dataframe, stores the data to fit the scaler.
         :param num_atts: list of str, each str represents the name of a numerical attribute in above data.
         :param feature_range: same parameter with sklearn MinMaxScaler
         :param copy: same parameter with sklearn MinMaxScaler
         """
-        cur_step = {}
-        for ai in num_atts:
-            cur_step[ai] = MinMaxScaler(feature_range=feature_range, copy=copy)
-        super().__init__("@".join(["MinMaxScaler"]+num_atts), df, step=cur_step, focus_atts=num_atts, fit_flag=True)
+        self.focus_atts = num_atts
+        self.feature_range = feature_range
+        self.copy = copy
+
+        self.fitted_step = None
 
+    def fit(self, df):
+        fitted_step = {}
+        for ai in self.focus_atts:
+            fitted_step[ai] = MinMaxScaler(feature_range=self.feature_range, copy=self.copy).fit(np.array(df[ai]).reshape(-1, 1))
+
+        self.fitted_step = fitted_step
+
+        return self
+
+    def apply(self, df):
+        after_df = df.copy()
+        for ai in self.focus_atts:
+            after_df[ai] = self.fitted_step[ai].transform(np.array(after_df[ai]).reshape(-1, 1))
+
+        return after_df
+
+    def name(self):
+        return "MinMaxScaler"
+
+    def abbr_name(self):
+        return "MS"
+
+    def step_name(self):
+        return "Scaler"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
-    cur_o = SK_StandardScaler(data, ["fnlwgt", "age"])
-    # cur_o = SK_MinMaxScaler(data, ["fnlwgt", "age"])
+    data = pd.read_csv("../../data/german_AIF.csv")
+    # cur_o = SK_StandardScaler(["month", "credit_amount"])
+    cur_o = SK_MinMaxScaler(["month", "credit_amount"])
 
+    cur_o.fit(data)
     after_data = cur_o.apply(data)
-    after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
-
-    print(cur_o.get_name())
\ No newline at end of file
+    after_data.to_csv("../../data/german_AIF_"+cur_o.name()+".csv", index=False)
diff --git a/pipeline/preprocess/splitters.py b/pipeline/preprocess/splitters.py
index d1ed315..9a725dc 100644
--- a/pipeline/preprocess/splitters.py
+++ b/pipeline/preprocess/splitters.py
@@ -1,23 +1,40 @@
 """
     Classes to split data into train, validation, and test set.
 """
-from pipeline.preprocess.preprocessor import Preprocessor
 import numpy as np
 import pandas as pd
 from sklearn.model_selection import StratifiedShuffleSplit
-
-def valid_split_ratio(input_ratios):
-    if input_ratios is None:
+from pipeline.step import Step
+
+# TODO: optimize the below function
+def split_to_df(splitter, input_df, y_label):
+    after_df_1 = pd.DataFrame()
+    after_df_2 = pd.DataFrame()
+    X = np.array(input_df.drop(columns=[y_label]))
+    y = np.array(input_df[y_label])
+    for index_1, index_2 in splitter.split(X, y):
+        X_1, X_2 = X[index_1], X[index_2]
+        y_1, y_2 = y[index_1], y[index_2]
+
+        after_df_1 = pd.concat(
+            [after_df_1, pd.DataFrame(data=np.hstack((X_1, y_1.reshape(-1, 1))), columns=input_df.columns)])
+        after_df_2 = pd.concat(
+            [after_df_2, pd.DataFrame(data=np.hstack((X_2, y_2.reshape(-1, 1))), columns=input_df.columns)])
+    return after_df_1, after_df_2
+
+def valid_split_ratio(split_ratio):
+    if split_ratio is None:
         print("Need to specify split_ratio!")
         raise ValueError
     else:
-        if len(input_ratios) == 1:  # not valid, at least two for train and test
-            print("split_ratio should have at least 2 values for train and test sets and at most 3 values for train, validation and test sets!")
+        if len(split_ratio) == 1:  # not valid, at least two for train and test
+            print(
+                "split_ratio should have at least 2 values for train and test sets and at most 3 values for train, validation and test sets!")
             raise ValueError
-        if sum([not isinstance(x, float) for x in input_ratios]) > 0:
+        if sum([not isinstance(x, float) for x in split_ratio]) > 0:
             print("split_ratio includes non float value!")
             raise ValueError
-        for x in input_ratios:
+        for x in split_ratio:
             if not isinstance(x, float):
                 print("split_ratio includes non float value!")
                 raise ValueError
@@ -25,108 +42,125 @@ def valid_split_ratio(input_ratios):
                 if x < 0 or x > 1:
                     print("split_ratio includes not valid value! Value should between 0 and 1.")
                     raise ValueError
-                if sum(input_ratios) != 1:
+                if sum(split_ratio) != 1:
                     print("The sum of split_ratio does not equal to 1!")
                     raise ValueError
 
     return True
 
-class BalanceTargetSplitter(Preprocessor):
-    def __init__(self, df, split_ratio, target_col, seed=0):
+class BalanceTargetSplitter(Step):
+    # TODO: fix the bug of not returning enough item
+    def __init__(self, split_ratio, target_col, seed):
         """
-        :param df: pandas dataframe, stores the data to split.
         :param split_ratio: list of float, each float represents the size-ratio of splitted data. Corresponding order maps to the size of the train, [validataion], and test set.
                             Value ranges in [0,1]. Sum of the values in this list should be equal to 1.
                             e.g. [0.7, 0.2, 0.1] means 70% train, 20% validation, and 10% test set.
         :param target_col: str, the name of the target variable in above data.
         :param seed: integer, seed to be used to generate random state. Same as 'random_state' in sklearn.model_selection.StratifiedKFold. Default is 0.
         """
-        super().__init__("@".join(["BalanceTargetSplitter", str(len(split_ratio))]), df=df, fit_flag=False, target_col=target_col)
-        if valid_split_ratio(split_ratio):
-            if len(split_ratio) == 2: # train and test
-                train_size = split_ratio[0]
-                test_size = split_ratio[1]
-                self.splitters = [StratifiedShuffleSplit(n_splits=1, test_size=test_size, train_size=train_size, random_state=seed)]
-            else: # train, validation and test
-                train_size = split_ratio[0]
-                validation_size = split_ratio[1]
-                test_size = split_ratio[2]
-
-                self.splitters = [StratifiedShuffleSplit(n_splits=1, test_size=test_size+validation_size, train_size=train_size, random_state=seed),
-                                  StratifiedShuffleSplit(n_splits=1, test_size=test_size, train_size=validation_size, random_state=seed)]
-
         self.split_ratio = split_ratio
+        self.target_col = target_col
         self.seed = seed
 
+        self.fitted_step = None
+
+
+    def fit(self, df):
+        if valid_split_ratio(self.split_ratio):
+            train_size = self.split_ratio[0]
+            validation_size = self.split_ratio[1]
+            test_size = self.split_ratio[2]
+
+            self.fitted_step = [StratifiedShuffleSplit(n_splits=1, test_size=test_size+validation_size, train_size=train_size, random_state=self.seed),
+                              StratifiedShuffleSplit(n_splits=1, test_size=test_size, train_size=validation_size, random_state=self.seed)]
+        else:
+            print("Invalid inputs!")
+            raise ValueError
+        return self
+
     def apply(self, df):
+        after_train_df, rest_df = split_to_df(self.fitted_step[0], df, self.target_col)
+        after_val_df, after_test_df = split_to_df(self.fitted_step[1], rest_df, self.target_col)
+        return after_train_df, after_val_df, after_test_df
+
+    def name(self):
+        return "BalanceTargetSplitter"
+
+    def abbr_name(self):
+        return "BS"
+
+    def step_name(self):
+        return "Splitter"
+
+    def input_encoded_data(self):
+        return False
+
+    def output_encoded_data(self):
+        return False
+
+    def fit_only_on_train(self):
+        return True
+
+class RandomSplitter(Step):
+    def __init__(self, split_ratio, seed):
         """
-        :param df: pandas dataframe, stores the data to apply the learned splitter.
-        :return: pandas dataframe, stores the data after split.
-        """
-        def split_to_df(splitter, input_df, y_label):
-            after_df_1 = pd.DataFrame()
-            after_df_2 = pd.DataFrame()
-            X = np.array(df.drop(columns=[y_label]))
-            y = np.array(df[y_label])
-            for index_1, index_2 in splitter.split(X, y):
-                X_1, X_2 = X[index_1], X[index_2]
-                y_1, y_2 = y[index_1], y[index_2]
-
-                after_df_1 = pd.concat([after_df_1, pd.DataFrame(data=np.hstack((X_1, y_1.reshape(-1,1))) ,columns=input_df.columns)])
-                after_df_2 = pd.concat([after_df_2, pd.DataFrame(data=np.hstack((X_2, y_2.reshape(-1,1))), columns=input_df.columns)])
-            return after_df_1, after_df_2
-
-        if len(self.split_ratio) == 2: # without validation set
-            return split_to_df(self.splitters[0], df, self.target_col)
-        else: # with validation set
-            after_train_df, rest_df = split_to_df(self.splitters[0], df, self.target_col)
-            after_val_df, after_test_df = split_to_df(self.splitters[1], rest_df, self.target_col)
-            return after_train_df, after_val_df, after_test_df
-
-class RandomSplitter(Preprocessor):
-    def __init__(self, df, split_ratio, seed=0):
-        """
-        :param df: pandas dataframe, stores the data to split.
         :param split_ratio: list of float, each float represents the size-ratio of splitted data. Corresponding order maps to the size of the train, [validataion], and test set.
                             Value ranges in [0,1]. Sum of the values in this list should be equal to 1.
                             e.g. [0.7, 0.2, 0.1] means 70% train, 20% validation, and 10% test set.
         :param seed: integer, seed to be used to generate random state. Same as 'random_state' in sklearn.model_selection.StratifiedKFold. Default is 0.
         """
-        super().__init__("@".join(["RandomSplitter", str(len(split_ratio))]), df=df, fit_flag=False)
-
-        if valid_split_ratio(split_ratio):
-            self.split_ratio = split_ratio
+        self.split_ratio = split_ratio
         self.seed = seed
+        self.fitted_step = None
+
+    def fit(self, df):
+        if valid_split_ratio(self.split_ratio):
+            pass
+        else:
+            print("Invalid inputs!")
+            raise ValueError
 
     def apply(self, df):
-        """
-        :param df: pandas dataframe, stores the data to apply the learned splitter.
-        :return: pandas dataframe, stores the data after split.
-        """
         df = df.sample(frac=1, random_state=self.seed).reset_index(drop=True)
-        if len(self.split_ratio) == 2:  # without validation set
-            split_idx = int(self.split_ratio[0]*df.shape[0])
-            return df[:split_idx], df[split_idx:]
-        else:  # with validation set
-            split_idx_1 = int(self.split_ratio[0] * df.shape[0])
-            split_idx_2 = split_idx_1 + int(self.split_ratio[1] * df.shape[0])
+        split_idx_1 = int(self.split_ratio[0] * df.shape[0])
+        split_idx_2 = split_idx_1 + int(self.split_ratio[1] * df.shape[0])
+
+        return df.iloc[:split_idx_1], df.iloc[split_idx_1:split_idx_2], df.iloc[split_idx_2:]
+
+    def name(self):
+        return "RandomSplitter"
+
+    def abbr_name(self):
+        return "RS"
+
+    def step_name(self):
+        return "Splitter"
+
+    def input_encoded_data(self):
+        return False
 
-            return df[:split_idx_1], df[split_idx_1:split_idx_2], df[split_idx_2:]
+    def output_encoded_data(self):
+        return False
 
+    def fit_only_on_train(self):
+        return False
 
 if __name__ == '__main__':
-    data = pd.read_csv("../../data/adult_pre_RandomSampler_1000.csv")
-    # cur_o = BalanceTargetSplitter(data, [0.7, 0.3], "income-per-year")
-    cur_o = RandomSplitter(data, [0.5, 0.3, 0.2])
+    data = pd.read_csv("../../data/german_AIF.csv")
+    print(data.shape)
+    # cur_o = BalanceTargetSplitter([0.5, 0.3, 0.2], "credit", 0) # bug: returned splitted data's total size does not equal to the input data
+    cur_o = RandomSplitter([0.5, 0.3, 0.2], 0)
 
+    cur_o.fit(data)
     after_train, after_val, after_test = cur_o.apply(data)
     # after_train, after_test = cur_o.apply(data)
     print(after_train.shape)
     print(after_val.shape)
     print(after_test.shape)
-    # after_data.to_csv("../../data/adult_"+cur_o.get_name()+".csv", index=False)
 
-    print(cur_o.get_name())
+    after_train.to_csv("../../data/german_AIF_train_"+cur_o.name()+".csv", index=False)
+    after_val.to_csv("../../data/german_AIF_val_"+cur_o.name()+".csv", index=False)
+    after_test.to_csv("../../data/german_AIF_test_"+cur_o.name()+".csv", index=False)
 
 
 
diff --git a/pipeline/step.py b/pipeline/step.py
index afd4aa5..e353bfa 100644
--- a/pipeline/step.py
+++ b/pipeline/step.py
@@ -1,49 +1,55 @@
 """
     Base abstract class for every step supported in this system.
 """
-STEP_NAMES = {"Sampler": "SA"}
+METHOD_NAME_MAPPING = {"RandomSplitter": "RSP", "BalanceTargetSplitter": "BTSP",
+             "RandomSampler": "RSM", "BalancePopulationSampler": "BPSM",
+             "DropNAImputer": "DNIM", "ModeImputer": "MIM", "DatawigImputer": "DWIM",
+             "SK_StandardScaler": "SSC", "SK_MinMaxScaler": "MMSC",
+             "SK_Discretizer": "DCA", "SK_Binarizer": "BCA",
+             "SK_OrdinalEncoder": "OREN", "SK_OneHotEncoder": "OHEN",
+             "CustomCateAttsEncoder": "CCSN",
+             "AIF_Reweighing": "RWFB", "AIF_DIRemover": "DIRFB",
+             "SK_LogisticRegression": "LRRM", "SK_DecisionTree": "DTRM", "OPT_LogisticRegression": "OLRRM", "OPT_DecisionTree": "ODTRM", "AIF_AdversarialDebiasing": "ADFM",
+             "AIF_EqOddsPostprocessing": "EQFA", "AIF_CalibratedEqOddsPostprocessing": "CEQFA"}
+
+SUPPORT_STEPS = {"SP": "Splitter", "SM": "Sampler", "IM": "Imputer", "SC": "Scaler",
+                 "CA": "Categorizer", "EN": "Encoder", "SN": "SensitiveAttEncoder",
+                 "FB": "FairPreprocessor", "RM": "model", "FM": "model", "RA": "FairPostprocessor"}
 
 class Step():
-    def __init__(self, step_name, df=None, focus_atts=[], sensitive_att=None, target_col=None):
-
-        if df is None:
-            print("Input data is empty!")
-            raise ValueError
-
-        if sensitive_att:
-            if sensitive_att not in df.columns:
-                print("Need to specify a valid sensitive attribute!")
-                raise ValueError
-            self.sensitive_att = sensitive_att
-        if target_col is not None:
-            if target_col not in df.columns:
-                print("Need to specify a valid target attribute to be predicted!")
-                raise ValueError
-            if len(df[target_col].unique()) != 2:
-                print("Only support binary target feature now!")
-                raise ValueError
-            self.target_col = target_col
-            self.pred_target_col = "pred_" + target_col  # store the predicted score (probability) column using this fixed name
-        if len(focus_atts) > 0:
-            if sum([x not in df.columns for x in focus_atts]) > 0:
-                print("Some specified attributes do not appear in the data!")
-                raise ValueError
-            self.focus_atts = focus_atts
-        self.name = step_name
-        # self.input_data = df
-
-
+    def fit(self, df):
+        """
+        :param df: pandas dataframe, stores the data to learn the step.
+        :return: self: the fitted step is updated.
+        """
+        raise NotImplementedError
 
     def apply(self, df):
         """
-        :param df: pandas dataframe, stores the data to apply the learned discretizer.
-        :return: pandas dataframe, stores the data after discretize.
+        :param df: pandas dataframe, stores the data to apply the step.
+        :return: pandas dataframe, stores the data after the step.
         """
         raise NotImplementedError
 
+    def name(self):
+        raise NotImplementedError
+
+    def abbr_name(self):
+        raise NotImplementedError
+
+    def step_name(self):
+        raise NotImplementedError
+
+    def input_encoded_data(self):
+        # use to detect the change of the dimension of the dataset
+        raise NotImplementedError
+
+    def output_encoded_data(self):
+        # use to detect the change of the dimension of the dataset
+        raise NotImplementedError
+
+    def fit_only_on_train(self):
+        # indicate whether the step fit on the input data (return False) or use the fitted model (return True)
+        raise NotImplementedError
 
-    def get_name(self): # return full name to print out
-        return self.name
 
-    def get_abbr_name(self): # return abbreviated name used in the file name of data
-        return STEP_NAMES[self.name]

From 46e1f3f093a63ac2f611ea2c9a3f54795b56193e Mon Sep 17 00:00:00 2001
From: Ke Yang <ky630@nyu.edu>
Date: Fri, 31 Jul 2020 09:26:47 -0400
Subject: [PATCH 7/7] minor for generator function

---
 pipeline/fairprep.py | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/pipeline/fairprep.py b/pipeline/fairprep.py
index a4dcc61..32f629b 100644
--- a/pipeline/fairprep.py
+++ b/pipeline/fairprep.py
@@ -181,20 +181,17 @@ def get_meta_information(self):
 
         return self
 
-    def iter_steps(self, steps, input_df):
+    def iter_steps(self, steps):
+        # TODO: optimize with the input data.
         """
         Generator function to iterate steps.
         :param steps: list of objects that represent the steps user want to perform on the input data.
                       Supported steps are listed in STEPS.md.
         :return:  the pandas dataframes that are returned by applying a step on the input data.
         """
-        for idx, stepi in enumerate(islice(steps, 0, len(steps))):
-            # if the current step is the encoder, feed inputdata's metadata so that the encoded data can have the domain of the raw dataset
-            stepi.fit(input_df)
-            return_df = stepi.apply(input_df)
-            if len(return_df) == 2: # special heck for the step that return weights
-                return_df = return_df[0]
-            yield idx, stepi.fitted_step, return_df
+        # islice(steps, 0, len(steps))
+        for idx, stepi in enumerate(steps):
+            yield idx, stepi
 
     def validate_input_steps(self, steps):
 
@@ -241,7 +238,8 @@ def run_pipeline(self, steps, save_interdata=False):
         print(PRINT_SPLIT)
 
         # for step_idx, train_fitted_step, train_df in self.iter_steps(steps, train_df):
-        for step_idx, stepi in enumerate(steps):
+        # for step_idx, stepi in enumerate(steps):
+        for step_idx, stepi in self.iter_steps(steps):
             if step_idx == 0:
                 continue