Code for predicting loan status

# importing important libraries

import numpy as np
import pandas as pd

#reading the data

data=pd.read_csv(‘train_u6lujuX_CVtuZ9i.csv’)

#understanding the data

print(data.head())
print (data.shape)

Output Window:

# removing uneccessary columns

data.drop([‘Loan_ID’],axis=1,inplace=True)

#converting the data into catogorical variables

c_data=pd.get_dummies(data=data, columns=[‘Gender’, ‘Married’, ‘Dependents’, ‘Education’,’Self_Employed’, ‘ApplicantIncome’, ‘CoapplicantIncome’, ‘LoanAmount’,’Loan_Amount_Term’, ‘Credit_History’, ‘Property_Area’, ‘Loan_Status’])
print(c_data.head())

Output Window:

#visualising feature importance

from sklearn.ensemble import ExtraTreesClassifier
import matplotlib as plt
x_1=c_data.iloc[:,0:1022]

#independent columns

y_1=c_data.iloc[:,-1]#target column i.e loanstatus
clf=ExtraTreesClassifier()
clf.fit(x_1,y_1)
print(clf.feature_importances_)

Output Window:

[0.00669548 0.00604855 0.00583627 ... 0.00706081 0.010482   0.0048473 ]

#finding out feature with more weightage

feat_imp=pd.Series(clf.feature_importances_,index=x_1.columns).sort_values(ascending=False,inplace=False)
top_imp_features=feat_imp.nlargest(5)
print(top_imp_features)
feat_imp.nlargest(5).plot(kind=’barh’)

Output Window:

Credit_History_0.0         0.128080
Credit_History_1.0         0.089349
CoapplicantIncome_0.0      0.010859
Property_Area_Semiurban    0.010482
ApplicantIncome_10000      0.009264
dtype: float64

#splitting the data into train test split

x=c_data[[‘Credit_History_0.0′,’Credit_History_1.0′,’Property_Area_Semiurban’,’ApplicantIncome_10000′]]
y=c_data[‘Loan_Status_Y’]
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=5)
from sklearn.ensemble import RandomForestClassifier
rfc=RandomForestClassifier(n_estimators=100)
rfc.fit(x_train,y_train)

#reading test data

data_2=pd.read_csv(r”C:\Users\Shanmukh\.spyder-py3\test_Y3wMUE5_7gLdaTN.csv”)

#removing unneccesary columns from test data

data_2.drop([‘Loan_ID’],axis=1,inplace=True)

#converting the test data into catogorical data

c2_data=pd.get_dummies(data=data_2, columns=[‘Credit_History’,’CoapplicantIncome’,’Property_Area’,’ApplicantIncome’])
x2=c2_data[[‘Credit_History_0.0′,’Credit_History_1.0′,’Property_Area_Semiurban’,’ApplicantIncome_10000′]]

#predicting loan status based on our train data for the test data

y_pred=rfc.predict(x_test)
print(y_pred)
predicted_y=rfc.predict(x2)
print(predicted_y)

Output Window:

#finding out accuracy

from sklearn.metrics import accuracy_score
accuracy=accuracy_score(y_test,y_pred)
print(‘Accuracy’,accuracy*100)

Output Window:

Accuracy 84.32432432432432