import pandas as pd import numpy as np

df_data = pd.read_csv('/work/WA_Fn-UseC_-Telco-Customer-Churn.csv')

df_data.head(5)

df_data.info()

df_data.TotalCharges = pd.to_numeric(df_data.TotalCharges, errors='coerce')

df_data.info()

df_data.isnull().sum()

df_data.dropna(inplace=True)

df_data.drop('customerID', axis=1, inplace=True)

df_data.head(5)

df_data['Churn'].replace(to_replace="Yes", value=1, inplace=True) df_data['Churn'].replace(to_replace="No", value=0, inplace=True)

df_data['Churn']

df_data_processing = df_data.copy()

df_data_processing = pd.get_dummies(df_data_processing) df_data_processing.head(5)

import matplotlib.pyplot as plt

fig = plt.figure(figsize=(15,9)) df_data_processing.corr()['Churn'].sort_values(ascending=True).plot(kind='bar') plt.show()

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler() df_data_processing_scaled = scaler.fit_transform(df_data_processing)

df_data_processing_scaled = pd.DataFrame(df_data_processing_scaled)

df_data_processing_scaled.columns = df_data_processing.columns

df_data_processing_scaled.head(5)

import seaborn as sns

sns.countplot(data=df_data, x='gender', hue='Churn') plt.show()

def plot_categorical(column): fig = plt.figure(figsize=(7,7)) sns.countplot(data=df_data, x=column, hue='Churn') plt.show()

column_cat = df_data.select_dtypes(include='object').columns

for _ in column_cat: plot_categorical(_)

fig = plt.figure(figsize=(7,7)) sns.pairplot(data=df_data, hue='Churn') plt.show()

X = df_data_processing_scaled.drop('Churn', axis=1) y = df_data_processing_scaled['Churn'].values

from sklearn.model_selection import train_test_split X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.3, random_state=42)

from sklearn.linear_model import LogisticRegression model = LogisticRegression() result = model.fit(X_train, y_train)

from sklearn import metrics prediction_test = model.predict(X_test) print(metrics.accuracy_score(y_test, prediction_test))

model.predict_log_proba(X_test)

model.coef_

model.feature_names_in_

weights = pd.Series(model.coef_[0], index=X.columns.values) print(weights.sort_values(ascending=False)[:10].plot(kind='bar'))

print(weights.sort_values(ascending=False)[-10:].plot(kind='bar'))

from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay

fig = plt.figure(figsize=(11,11)) cm = confusion_matrix(y_test, prediction_test, labels=model.classes_) disp = ConfusionMatrixDisplay(confusion_matrix = cm, display_labels=model.classes_) disp.plot(cmap='gray') plt.show()

lasso = LogisticRegression(max_iter=10000, penalty='l1',solver='saga',C=0.5) lasso.fit(X_train, y_train) lasso.score(X_test,y_test)

cm=confusion_matrix(lasso.predict(X_test),y_test) sns.heatmap( cm, annot=True, cmap='gray', cbar=False, square=True, fmt="d" ) plt.ylabel('Real Label') plt.xlabel('Predicted Label');

weights = pd.Series(lasso.coef_[0], index=X.columns.values).sort_values(ascending=False) fig = plt.figure(figsize=(15,5)) weights.plot(kind='bar');

weights[weights==0]

ridge=LogisticRegression(max_iter=10000, penalty='l2',solver='saga',C=0.5) ridge.fit(X_train, y_train) ridge.score(X_test,y_test)

cm=confusion_matrix(ridge.predict(X_test),y_test) sns.heatmap( cm, annot=True, cmap='gray', cbar=False, square=True, fmt="d" ) plt.ylabel('Real Label') plt.xlabel('Predicted Label');

weights = pd.Series(ridge.coef_[0], index=X.columns.values).sort_values(ascending=False) fig = plt.figure(figsize=(15,5)) weights.plot(kind='bar');