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

import sklearn.preprocessing

from sklearn.linear_model import LogisticRegression

from sklearn.model_selection import train_test_split

df = pd.read_csv('bank.csv', header =0,delimiter=";")

df.head(10)

df.describe()

print(df.shape)

print(list(df.columns))

import seaborn as sns

sns.pairplot(df[['age', 'job', 'marital', 'education', 'default', 'balance', 'housing', 'loan', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays', 'previous', 'poutcome', 'y']])

df.columns

df_new = df.drop(df.columns[[0, 3, 7, 8, 9, 10, 11, 12, 13, 14]], axis =1)

df_new

data = pd.get_dummies(df_new, columns=['job','marital','default','housing','poutcome'],drop_first=False)

data.columns[[13,22,24]]

len(data.columns)

data = data.drop(data.columns[[13,22,24]],axis=1)

Y = data.iloc[:,1]

Y.head()

X = data.drop(data.columns[[1]],axis=1)

X.head()

X_train, X_test, Y_train, Y_test = train_test_split(X, Y, random_state = 0)

classifier = LogisticRegression(solver = "lbfgs", random_state= 0) classifier.fit(X_train, Y_train)

predicted_y = classifier.predict(X_test)

sum(predicted_y.tolist() == Y_test)/len(Y_test)

classifier.score(X_test, Y_test)

df["y"].value_counts() print(df["y"].value_counts()[0]/(df["y"].value_counts()[0]+df["y"].value_counts()[1]))