app.py

import gradio as gr
import joblib
import numpy as np

# Load trained model
model = joblib.load("driving_score_model.pkl")

# Define the prediction function
def predict_driving_score(Acceleration_X, Acceleration_Y, Acceleration_Z, Gyroscope_X, Gyroscope_Y, Gyroscope_Z, Speed_kmh):
    try:
        # Prepare input as a numpy array
        sample_input = np.array([[Acceleration_X, Acceleration_Y, Acceleration_Z, Gyroscope_X, Gyroscope_Y, Gyroscope_Z, Speed_kmh]])
        
        # Predict scores
        prediction = model.predict(sample_input)
        safety_score = prediction[0][0]
        eco_score = prediction[0][1]

        return round(safety_score, 2), round(eco_score, 2)
    
    except Exception as e:
        return f"Error: {str(e)}", f"Error: {str(e)}"

# Gradio Interface (expose POST method)
def gradio_interface():
    return gr.Interface(
        fn=predict_driving_score,
        inputs=[
            gr.Number(label="Acceleration_X"), 
            gr.Number(label="Acceleration_Y"), 
            gr.Number(label="Acceleration_Z"),
            gr.Number(label="Gyroscope_X"),
            gr.Number(label="Gyroscope_Y"),
            gr.Number(label="Gyroscope_Z"),
            gr.Number(label="Speed_kmh")
        ],
        outputs=[
            gr.Number(label="Predicted Safety Score"),
            gr.Number(label="Predicted EcoScore")
        ],
        title="Driving Safety & Eco Score Predictor",
        description="Enter sensor values to predict Safety Score and EcoScore.",
        allow_flagging="never",  # Optionally remove flagging for a more streamlined interface
        live=True  # Automatically update results as inputs change
    )

# Launch the Gradio interface and expose a public link
gradio_interface().launch(share=True)  # share=True will provide a public link