app.py

import gradio as gr
import joblib
import numpy as np

# Load the saved model and scaler
model = joblib.load('best_random_forest_model.pkl')
scaler = joblib.load('scaler.pkl')

def predict_house_price(feature1, feature2, feature3, feature4, feature5):
    # Combine all features into a single list
    features = [feature1, feature2, feature3, feature4, feature5]
    
    # Scale the input features using the saved scaler
    scaled_features = scaler.transform([features])
    
    # Make predictions using the loaded model
    prediction = model.predict(scaled_features)
    
    return f"Predicted House Price: ${prediction[0]:,.2f}"

# Create the Gradio interface
interface = gr.Interface(
    fn=predict_house_price,
    inputs=[
        gr.Number(label="Feature 1: (e.g., Average number of rooms per dwelling)"),
        gr.Number(label="Feature 2: (e.g., Total rooms)"),
        gr.Number(label="Feature 3: (e.g., Total bedrooms)"),
        gr.Number(label="Feature 4: (e.g., Median age of houses)"),
        gr.Number(label="Feature 5: (e.g., Population)"),
    ],
    outputs=gr.Textbox(label="Predicted House Price"),
    description="Predict California house prices based on input features.",
    title="California House Price Predictor"
)

# Launch the Gradio app
interface.launch()