app.py

import gradio as gr
import joblib
import pandas as pd

# Load the model
model = joblib.load('loan_RFmodel.joblib')

def predict_loan(gender, married, dependents, education, self_employed, 
                 income, co_income, amount, term, history, area):
    
    # Model still receives the numbers (0, 1) from the dropdown values
    features = pd.DataFrame([[
        gender, married, dependents, education, self_employed, 
        income, co_income, amount, term, history, area
    ]], columns=['Gender', 'Married', 'Dependents', 'Education', 'Self_Employed', 
                 'ApplicantIncome', 'CoapplicantIncome', 'LoanAmount', 
                 'Loan_Amount_Term', 'Credit_History', 'Property_Area'])
    
    prediction = model.predict(features)[0]
    return "✅ Loan Approved" if prediction == 1 else "❌ Loan Rejected"

# The Interface using the ("Label", Value) format
demo = gr.Interface(
    fn=predict_loan,
    inputs=[
    # The FIRST item is what the user sees, the SECOND is the number the model gets
    gr.Dropdown(choices=[("Male", 1), ("Female", 0)], label="Gender"),
    gr.Dropdown(choices=[("Yes", 1), ("No", 0)], label="Married"),
    gr.Dropdown(choices=[("0", 0), ("1", 1), ("2", 2), ("3+", 3)], label="Dependents"),
    gr.Dropdown(choices=[("Graduate", 0), ("Not Graduate", 1)], label="Education"),
    gr.Dropdown(choices=[("Yes", 1), ("No", 0)], label="Self Employed"),
    gr.Number(label="Applicant Income"),
    gr.Number(label="Co-applicant Income"),
    gr.Number(label="Loan Amount"),
    gr.Number(label="Term"),
    gr.Dropdown(choices=[("Good", 1), ("Bad", 0)], label="Credit History"),
    gr.Dropdown(choices=[("Rural", 0), ("Semiurban", 1), ("Urban", 2)], label="Property Area")
    ],
    outputs="text",
    title="🏦 Loan Approval Predictor Final"
)

# Crucial for Hugging Face deployment
demo.launch(server_name="0.0.0.0", server_port=7860)