commit-live-students · akshaylakade · Jan 18, 2019 · Jan 18, 2019 · Jan 18, 2019
diff --git a/__pycache__/__init__.cpython-36.pyc b/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_missing_value/__pycache__/__init__.cpython-36.pyc b/q01_missing_value/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_missing_value/__pycache__/build.cpython-36.pyc b/q01_missing_value/__pycache__/build.cpython-36.pyc
diff --git a/q01_missing_value/build.py b/q01_missing_value/build.py
@@ -1,5 +1,9 @@
+# %load q01_missing_value/build.py
 # Default imports
 import pandas as pd
+import numpy as np
+from sklearn.preprocessing import Imputer
+
 
 # Data loading
 ny_housing = pd.read_csv('data/train.csv')
@@ -8,3 +12,27 @@
 
 
 # Write your code here:
+def imputation(housing_data):
+    ''' Function to impute numerical data with mean
+        impute categorical data with mode '''
+    #Seperate numercial and categorical columns
+    num_col = housing_data._get_numeric_data().columns
+    cat_col = housing_data.columns.difference(num_col)
+
+    #impute numerical data
+    imput_num = Imputer(missing_values=np.NaN , strategy= 'mean')
+    df_num = imput_num.fit_transform(housing_data.loc[: , num_col])
+    df_num = pd.DataFrame(df_num)
+    df_num.columns = num_col
+
+    #impute categorial data
+    for col in cat_col:
+        housing_data.loc[: , col].fillna(housing_data.loc[:,col].mode()[0] , inplace = True)
+    df_cat = housing_data.loc[:,cat_col]
+
+    return df_num,df_cat
+
+imputation(housing_data)
+
+
+
diff --git a/q01_missing_value/tests/__pycache__/__init__.cpython-36.pyc b/q01_missing_value/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_missing_value/tests/__pycache__/test_q01_imputation.cpython-36.pyc b/q01_missing_value/tests/__pycache__/test_q01_imputation.cpython-36.pyc
diff --git a/q02_outlier_removal/__pycache__/__init__.cpython-36.pyc b/q02_outlier_removal/__pycache__/__init__.cpython-36.pyc
diff --git a/q02_outlier_removal/__pycache__/build.cpython-36.pyc b/q02_outlier_removal/__pycache__/build.cpython-36.pyc
diff --git a/q02_outlier_removal/build.py b/q02_outlier_removal/build.py
@@ -1,5 +1,14 @@
+# %load q02_outlier_removal/build.py
 # Default imports
+
 import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+from scipy import stats
+import warnings
+warnings.filterwarnings('ignore')
+
 
 # Data
 ny_housing = pd.read_csv('data/train.csv')
@@ -8,3 +17,13 @@
 
 
 # Write your code here:
+def outlier_removal(housing_data):
+    num_col = housing_data._get_numeric_data().columns
+    for col in num_col:
+        housing_data = housing_data.drop(housing_data[housing_data[col] > housing_data[col].quantile(0.95)].index)
+    return housing_data
+
+outlier_removal(housing_data)
+print(housing_data.shape)
+
+
diff --git a/q02_outlier_removal/tests/__pycache__/__init__.cpython-36.pyc b/q02_outlier_removal/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q02_outlier_removal/tests/__pycache__/test_q02_outlier_removal.cpython-36.pyc b/q02_outlier_removal/tests/__pycache__/test_q02_outlier_removal.cpython-36.pyc
diff --git a/q03_skewness_log/__pycache__/__init__.cpython-36.pyc b/q03_skewness_log/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_skewness_log/__pycache__/build.cpython-36.pyc b/q03_skewness_log/__pycache__/build.cpython-36.pyc
diff --git a/q03_skewness_log/build.py b/q03_skewness_log/build.py
@@ -1,3 +1,4 @@
+# %load q03_skewness_log/build.py
 from scipy.stats import skew
 import pandas as pd
 import numpy as np
@@ -6,3 +7,9 @@
 
 
 # Write code here:
+def skewness_log(data):
+    return skew(np.log(data['GrLivArea'])), skew(np.log(data['SalePrice']))
+
+skewness_log(data)
+
+
diff --git a/q03_skewness_log/tests/__pycache__/__init__.cpython-36.pyc b/q03_skewness_log/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_skewness_log/tests/__pycache__/test_q03_skewness_log.cpython-36.pyc b/q03_skewness_log/tests/__pycache__/test_q03_skewness_log.cpython-36.pyc
diff --git a/q03_skewness_sqrt/__pycache__/__init__.cpython-36.pyc b/q03_skewness_sqrt/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_skewness_sqrt/__pycache__/build.cpython-36.pyc b/q03_skewness_sqrt/__pycache__/build.cpython-36.pyc
diff --git a/q03_skewness_sqrt/build.py b/q03_skewness_sqrt/build.py
@@ -1,3 +1,4 @@
+# %load q03_skewness_sqrt/build.py
 # Default imports
 from scipy.stats import skew
 import pandas as pd
@@ -8,3 +9,12 @@
 
 # Write your Solution Here:
 
+def skewness_sqrt(ny_housing):
+    skewed_sqrt_val2 = skew(np.sqrt(ny_housing['SalePrice']))
+    skewed_sqrt_val1 = skew(np.sqrt(ny_housing['GrLivArea']))
+    return skewed_sqrt_val1,skewed_sqrt_val2
+
+skewness_sqrt(ny_housing)
+
+
+
diff --git a/q03_skewness_sqrt/tests/__pycache__/__init__.cpython-36.pyc b/q03_skewness_sqrt/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_skewness_sqrt/tests/__pycache__/test-q03_skewness_sqrt.cpython-36.pyc b/q03_skewness_sqrt/tests/__pycache__/test-q03_skewness_sqrt.cpython-36.pyc