preprocess.py

import pandas as pd
import numpy as np

df3 = pd.read_csv('datasets/CustomerChurn.csv')

df3 = df3.drop('customerID', axis=1)
df3 = df3.replace(r'^\s+$', np.nan, regex=True) 
df3['TotalCharges'] = df3['TotalCharges'].astype(float)
df3 = df3.fillna(df3.mean())
df3.isnull().values.any()

for key in df3.keys():
#    print(np.unique(df3[key]))
    print(key+' '+str(len(np.unique(df3[key]))))