feature_analysis.py

import numpy as np
import pandas as pd
import missingno as msno
import seaborn as sn
import matplotlib.pyplot as plt

data = pd.read_csv('Credit_Card_Applications.csv')
data = data.drop(["CustomerID"],axis=1)

# Missing data detection
msno.matrix(data,figsize=(10,3))

# Outliners detection and class imbalance
continiousData = pd.DataFrame()
continousVariableList  = ["A2", "A3", "A7", "A10", "A13", "A14"]
for var in continousVariableList:
    continiousData[var] = data[var].astype("float32")

fig, axes = plt.subplots(nrows=1,ncols=1)
fig.set_size_inches(10, 20)
sn.boxplot(data=continiousData,orient="v",ax=axes)

# Correlation analysis
corrMatt = data.corr()
mask = np.array(corrMatt)
mask[np.tril_indices_from(mask)] = False
fig,ax= plt.subplots()
fig.set_size_inches(20,10)
sn.heatmap(corrMatt, mask=mask,vmax=.8, square=True,annot=True)