app.py

import numpy as np
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
import gradio as gr

df = pd.read_csv("credit_risk_dataset.csv")

df = df.dropna()

df.columns

X =df.drop("loan_status", axis = 1)
y = df['loan_status']

categorical_features = ["person_home_ownership", "loan_intent", "loan_grade", "cb_person_default_on_file"]
X = pd.get_dummies(X, categorical_features)
X.columns



from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2)

X_train.head()

from sklearn.preprocessing import StandardScaler
scaler_normal = StandardScaler()
def scaler(data, runtime = False):
    normal_col = ['person_income','person_age','person_emp_length', 'loan_amnt','loan_int_rate','cb_person_cred_hist_length','loan_percent_income']
    if(runtime == False):
        data.loc[:,normal_col] = scaler_normal.fit_transform(data.loc[:,normal_col])
    else:
        data.loc[:,normal_col] = scaler_normal.transform(data.loc[:,normal_col])
    return data

X_train = scaler(X_train)
X_test = scaler(X_test, True)

rf_model = RandomForestClassifier(max_depth = 5)
rf_model.fit(X_train, y_train)

features = {
    "person_home_ownership": ['MORTGAGE', 'OTHER','OWN', 'RENT',],
    "loan_intent": ['DEBTCONSOLIDATION', 'EDUCATION', 'HOMEIMPROVEMENT', 'MEDICAL', 'PERSONAL', 'VENTURE'],
    "loan_grade": ['A','B', 'C', 'D', 'E','F', 'G'],
    "cb_person_default_on_file": ['N', 'Y']
}
def preprocess(model_input):
    for feature in features:
        for option in features[feature]:
            selection = model_input[feature]
            if option is selection:
                model_input[f'{feature}_{option}'] = 1
            else:
                model_input[f'{feature}_{option}'] = 0
                
    model_input.drop([_ for _ in features], inplace = True, axis = 1)
    return model_input

    

def credit_run(person_age, person_income, person_home_ownership,
       person_emp_length, loan_intent, loan_grade, loan_amnt,
       loan_int_rate, cb_person_default_on_file, cb_person_cred_hist_length):
    model_input = preprocess(
    pd.DataFrame( { 'person_age': person_age,
                   'person_income': person_income,
                   'person_home_ownership': person_home_ownership,
                   'person_emp_length': person_emp_length,
                   'loan_intent': loan_intent,
                   'loan_grade': loan_grade,
                   'loan_amnt': loan_amnt,
                   'loan_int_rate': loan_int_rate,
                   'loan_percent_income': loan_amnt / person_income,
                   'cb_person_default_on_file': cb_person_default_on_file,
                   'cb_person_cred_hist_length': cb_person_cred_hist_length
                  }, index = [0]
    ))
    out = rf_model.predict(model_input)
    return "High risk of defaulting" if out[0] == 1 else "Low risk of defaulting" 

demo = gr.Interface(
    fn = credit_run,
    inputs = [
        gr.Slider(label="Person Age(In Years)", minimum=18, maximum=90, step=1),
        gr.Number(label="Person Income(per month)"),
        gr.Radio(['MORTGAGE', 'OTHER','OWN', 'RENT'],label="Home Ownership Status"),
        gr.Slider(label="Pererson Emp Length(In Years)", minimum=0, maximum=60, step=1),
        gr.Radio(['DEBTCONSOLIDATION', 'EDUCATION', 'HOMEIMPROVEMENT', 'MEDICAL', 'PERSONAL', 'VENTURE'],label="Credit Intent"),
        gr.Radio(['A','B', 'C', 'D', 'E','F', 'G'],label="Type Of Credit"),
        gr.Number(label="Credit Amount"),
        gr.Number(label="Credit Interest Rate"),
        gr.Radio(['N', 'Y'],label="Person Defaulted in History"),
        gr.Number(label="Person's Credit History Length"),
             ],
    outputs = "text",
    title = gr.Image(type="pil", value=Non_payment_logo.jpg),
    examples = [[23,25000,'RENT',2,'EDUCATION','A',30000,8.9,'N',6],
[34,50000,'OWN',1,'MEDICAL','B',62000,10.65,'N',3],
[42,30000,"MORTGAGE",12,'HOMEIMPROVEMENT','C',800000,7.9,'Y',8],
[42,20000,"MORTGAGE",10,'PERSONAL','F',100000,15.25,'Y',5]]
)
demo.launch(debug=True)