app.py

import gradio as gr
import pandas as pd
import numpy as np
import pickle
import json
import tensorflow as tf
from tensorflow.keras.models import model_from_json
import os

# Initialize model components
model = None
scaler = None
metadata = {}
feature_names = []

def load_model():
    global model, scaler, metadata, feature_names
    
    try:
        # Load model architecture
        with open('model_architecture.json', 'r') as json_file:
            model_json = json_file.read()
        model = model_from_json(model_json)
        
        # Load weights
        model.load_weights('final_model.h5')
        
        # Load scaler
        with open('scaler.pkl', 'rb') as f:
            scaler = pickle.load(f)
        
        # Load metadata
        with open('metadata.json', 'r') as f:
            metadata = json.load(f)
            feature_names = metadata['feature_names']
            
        print("✅ Model loaded successfully!")
        print(f"Using features: {feature_names}")
    except Exception as e:
        print(f"❌ Error loading model: {str(e)}")

# Load model at startup
load_model()

def predict(*args):
    try:
        if model is None or scaler is None:
            raise Exception("Model not loaded. Please check the model files.")
            
        # Create input dictionary
        input_data = {feature_names[i]: float(args[i]) for i in range(len(feature_names))}
        input_df = pd.DataFrame([input_data])
        
        # Scale features
        scaled_input = scaler.transform(input_df)
        
        # Predict
        probability = float(model.predict(scaled_input)[0][0])
        prediction = "Eligible" if probability > 0.5 else "Not Eligible"
        confidence = abs(probability - 0.5) * 2
        
        return {
            "Prediction": prediction,
            "Probability": f"{probability:.4f}",
            "Confidence": f"{confidence:.4f}"
        }
    except Exception as e:
        return {"Error": str(e)}

# Create Gradio interface
iface = gr.Interface(
    fn=predict,
    inputs=[gr.Number(label=name) for name in feature_names],
    outputs=[
        gr.Textbox(label="Prediction"),
        gr.Textbox(label="Probability"), 
        gr.Textbox(label="Confidence")
    ],
    title="🎓 Student Eligibility Predictor",
    description="Predict student eligibility based on academic performance metrics",
    examples=[[75, 80, 85] if len(feature_names) >= 3 else [75, 80]]  # Example inputs
)

if __name__ == "__main__":
    iface.launch()