dradecic

import math
import numpy as np
import pandas as pd

from bokeh.embed import components
from bokeh.models import ColumnDataSource, HoverTool, PrintfTickFormatter
from bokeh.plotting import figure
from bokeh.transform import factor_cmap

from flask import Flask, render_template, request

dradecic

%matplotlib inline

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

import warnings
warnings.filterwarnings('ignore')
pd.options.display.max_columns = None

dradecic

data.drop(['Ticket', 'PassengerId'], axis=1, inplace=True)

gender_mapper = {'male': 0, 'female': 1}
data['Sex'].replace(gender_mapper, inplace=True)

data['Title'] = data['Name'].apply(lambda x: x.split(',')[1].strip().split(' ')[0])
data['Title'] = [0 if x in ['Mr.', 'Miss.', 'Mrs.'] else 1 for x in data['Title']]
data = data.rename(columns={'Title': 'Title_Unusual'})
data.drop('Name', axis=1, inplace=True)

dradecic

print('Optimal number of features: {}'.format(rfecv.n_features_))

dradecic

print(np.where(rfecv.support_ == False)[0])

X.drop(X.columns[np.where(rfecv.support_ == False)[0]], axis=1, inplace=True)

dradecic

dset = pd.DataFrame()
dset['attr'] = X.columns
dset['importance'] = rfecv.estimator_.feature_importances_

dset = dset.sort_values(by='importance', ascending=False)


plt.figure(figsize=(16, 14))
plt.barh(y=dset['attr'], width=dset['importance'], color='#1976D2')
plt.title('RFECV - Feature Importances', fontsize=20, fontweight='bold', pad=20)

dradecic

plt.figure(figsize=(16, 9))
plt.title('Recursive Feature Elimination with Cross-Validation', fontsize=18, fontweight='bold', pad=20)
plt.xlabel('Number of features selected', fontsize=14, labelpad=20)
plt.ylabel('% Correct Classification', fontsize=14, labelpad=20)
plt.plot(range(1, len(rfecv.grid_scores_) + 1), rfecv.grid_scores_, color='#303F9F', linewidth=3)

plt.show()

dradecic

data.drop(['RowNumber', 'CustomerId', 'Surname'], axis=1, inplace=True)

data['CreditScore_Bins'] = pd.qcut(data['CreditScore'], 5, 
                                   labels=['CS_lt_566', 'CS_556_to_627', 'CS_627_to_678', 'CS_678_to_735', 'CS_gt_735'])
data['Age_Bins'] = pd.qcut(data['Age'], 5, 
                           labels=['Age_lt_31', 'Age_31_to_35', 'Age_35_to_40', 'Age_40_to_46', 'Age_gt_46'])
data['Balance_Bins'] = pd.qcut(data['Balance'], 5, 
                               labels=['Bal_lt_73080', 'Bal_73080_to_110138', 'Bal_110138_to_133710', 'Bal_gt_133710'],
                               duplicates='drop')
data['Salary_Bins'] = pd.qcut(data['EstimatedSalary'], 5, 

dradecic

for col in data.columns:
    if col == 'Exited': continue
    else:
        print('WoE and IV for column: {}'.format(col))
        df, iv = calculate_woe_iv(data, col, 'Exited')
        print(df)
        print('IV score: {:.2f}'.format(iv))
        print('\n')

dradecic

def coarse_classer(df, indexloc_1, indexloc_2):
    mean_val = pd.DataFrame(np.mean(pd.DataFrame([df.iloc[indexloc_1], df.iloc[indexloc_2]]))).T
    original = df.drop([indexloc_1, indexloc_2])
    
    coarsed_df = pd.concat([original, mean_val])
    coarsed_df = coarsed_df.sort_values(by='WoE', ascending=False).reset_index(drop=True)
    
    return coarsed_df