Commit Live PR

q01_outlier_removal/build.py

@@ -1,8 +1,31 @@
+# %load q01_outlier_removal/build.py
 # Default imports
 import pandas as pd
-
+import numpy as np
 loan_data = pd.read_csv('data/loan_prediction_uncleaned.csv')
 loan_data = loan_data.drop('Loan_ID', 1)
 
 
 # Write your Solution here:
+def outlier_removal(data):
+    App_inc_UQR = data.quantile(0.95,interpolation='nearest')[0]
+    Coo_app_inc_UQR = data.quantile(0.95,interpolation='nearest')[1]
+    Loan_amount_UQR = data.quantile(0.95,interpolation='nearest')[2]
+    Loan_amount_term_UQR = data.quantile(0.95,interpolation='nearest')[3]
+
+    data = data[data['ApplicantIncome']< App_inc_UQR]
+    #loan_data = loan_data[loan_data['CoapplicantIncome']<= Coo_app_inc_UQR]
+    data = data[data['LoanAmount'] < Loan_amount_UQR]
+    #loan_data = loan_data[loan_data['Loan_Amount_Term']<= Loan_amount_term_UQR]
+
+    return data
+
+
+
+
+
+
+
+
+
+

q02_data_cleaning_all/build.py

@@ -1,5 +1,7 @@
+# %load q02_data_cleaning_all/build.py
 # Default Imports
 import sys, os
+import random
 sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname('__file__'))))
 import pandas as pd
 import numpy as np
@@ -10,5 +12,41 @@
 loan_data = loan_data.drop('Loan_ID', 1)
 loan_data = outlier_removal(loan_data)
 
+random.seed(9)
 
 # Write your solution here :
+def data_cleaning(loan_data):
+    columns = np.array(loan_data.columns)
+
+    null_values_columns=[]
+    for i in range(len(columns)):
+        print('-----------------')
+        print(columns[i])
+        print(loan_data.loc[:,columns[i]].isnull().values.any())
+        if (loan_data.loc[:,columns[i]].isnull().values.any() == True):
+            null_values_columns.append(columns[i])
+
+
+    for i in range(len(null_values_columns)):
+        dtype = loan_data.loc[:,null_values_columns[i]].get_dtype_counts().index[0]
+        if dtype == 'float64':
+            mean = loan_data.loc[:,null_values_columns[i]].mean()
+            loan_data.loc[:,null_values_columns[i]].fillna(mean,inplace=True)
+        elif dtype=='object':
+            mode = loan_data.loc[:,null_values_columns[i]].mode()[0]
+            loan_data.loc[:,null_values_columns[i]].fillna(mode,inplace=True)
+
+    X_train,X_test,y_train,y_test = train_test_split(loan_data.iloc[:,:-1],loan_data.iloc[:,-1],test_size=0.25)
+    X , y = loan_data.iloc[:,:-1],loan_data.iloc[:,-1]
+
+    return X,y,X_train,X_test,y_train,y_test
+
+
+
+
+
+
+
+
+
+