app.py

import gradio as gr
import joblib  # Use pickle if your model is in .pkl format
import numpy as np

# Load the trained model
model_path = "mobile_price_model.joblib"  # Change to your file name if using .pkl
model = joblib.load(model_path)

# Define the prediction function
def predict_price(battery_power, ram, px_width, px_height):
    """Predicts the mobile price category based on input features."""
    features = np.array([[battery_power, ram, px_width, px_height]])  # Adjust as needed
    prediction = model.predict(features)
    return f"Predicted Price Category: {prediction[0]}"

# Create Gradio Interface
inputs = [
    gr.Number(label="Battery Power (mAh)"),
    gr.Number(label="RAM (MB)"),
    gr.Number(label="Pixel Width"),
    gr.Number(label="Pixel Height")
]

output = gr.Textbox(label="Price Category")

app = gr.Interface(fn=predict_price, inputs=inputs, outputs=output, title="📱 Mobile Price Prediction")

# Run the app
app.launch()