diff --git a/process_data.py b/process_data.py
index c5ab027..fdc5e2a 100644
--- a/process_data.py
+++ b/process_data.py
@@ -1,13 +1,12 @@
 import numpy as np
 import theano
-import cPickle
+import pickle as cPickle
 from collections import defaultdict
 import sys, re
 import pandas as pd
 import csv
 import getpass
 
-
 def build_data_cv(datafile, cv=10, clean_string=True):
     """
     Loads data and split into 10 folds.
@@ -15,7 +14,7 @@ def build_data_cv(datafile, cv=10, clean_string=True):
     revs = []
     vocab = defaultdict(float)
 
-    with open(datafile, "rb") as csvf:
+    with open(datafile, "r") as csvf:
         csvreader=csv.reader(csvf,delimiter=',',quotechar='"')
         first_line=True
         for line in csvreader:
@@ -93,7 +92,7 @@ def load_bin_vec(fname, vocab):
         header = f.readline()
         vocab_size, layer1_size = map(int, header.split())
         binary_len = np.dtype(theano.config.floatX).itemsize * layer1_size
-        for line in xrange(vocab_size):
+        for line in range(vocab_size):
             word = []
             while True:
                 ch = f.read(1)
@@ -116,7 +115,7 @@ def add_unknown_words(word_vecs, vocab, min_df=1, k=300):
     for word in vocab:
         if word not in word_vecs and vocab[word] >= min_df:
             word_vecs[word] = np.random.uniform(-0.25,0.25,k)
-            print word
+            print(word)
 
 def clean_str(string, TREC=False):
     """
@@ -159,18 +158,18 @@ def get_mairesse_features(file_name):
     w2v_file = sys.argv[1]
     data_folder = sys.argv[2]
     mairesse_file = sys.argv[3]
-    print "loading data...",
+    print("loading data...",end ='')
     revs, vocab = build_data_cv(data_folder, cv=10, clean_string=True)
     num_words=pd.DataFrame(revs)["num_words"]
     max_l = np.max(num_words)
-    print "data loaded!"
-    print "number of status: " + str(len(revs))
-    print "vocab size: " + str(len(vocab))
-    print "max sentence length: " + str(max_l)
-    print "loading word2vec vectors...",
+    print ("data loaded!")
+    print ("number of status: " + str(len(revs)))
+    print ("vocab size: " + str(len(vocab)))
+    print ("max sentence length: " + str(max_l))
+    print ("loading word2vec vectors..."),
     w2v = load_bin_vec(w2v_file, vocab)
-    print "word2vec loaded!"
-    print "num words already in word2vec: " + str(len(w2v))
+    print ("word2vec loaded!")
+    print ("num words already in word2vec: " + str(len(w2v)))
     add_unknown_words(w2v, vocab)
     W, word_idx_map = get_W(w2v)
     rand_vecs = {}
@@ -178,5 +177,5 @@ def get_mairesse_features(file_name):
     W2, _ = get_W(rand_vecs)
     mairesse = get_mairesse_features(mairesse_file)
     cPickle.dump([revs, W, W2, word_idx_map, vocab, mairesse], open("essays_mairesse.p", "wb"))
-    print "dataset created!"
+    print ("dataset created!")