app.py

# Import necessary libraries
import joblib
import pandas as pd
from flask import Flask, request, jsonify

# Initialize the Flask application
superkart_sales_api = Flask("SuperKart Sales Revenue Predictor")

# Load the trained machine learning model
model = joblib.load("superkart_prediction_model_v1_0.joblib")

# Home endpoint
@superkart_sales_api.get('/')
def home():
    return "Welcome to the SuperKart Sales Revenue Prediction API!"

# Single record prediction
@superkart_sales_api.post('/v1/sales')
def predict_sales_single():
    """
    Handles single prediction for SuperKart product revenue.
    """
    try:
        product_data = request.get_json()

        # Validate required fields
        required_fields = [
            'Product_Weight', 'Product_Sugar_Content', 'Product_Allocated_Area',
            'Product_Type', 'Product_MRP', 'Store_Size',
            'Store_Location_City_Type', 'Store_Type', 'Store_Establishment_Year'
        ]
        for field in required_fields:
            if field not in product_data:
                return jsonify({"error": f"Missing required field: {field}"}), 400

        # Compute derived features
        store_age = 2025 - int(product_data['Store_Establishment_Year'])
        mrp_per_sqm = float(product_data['Product_MRP']) / float(product_data['Product_Allocated_Area'])
        mrp_x_weight = float(product_data['Product_MRP']) * float(product_data['Product_Weight'])

        # Prepare input sample
        sample = {
            'Product_Weight': product_data['Product_Weight'],
            'Product_Sugar_Content': product_data['Product_Sugar_Content'],
            'Product_Allocated_Area': product_data['Product_Allocated_Area'],
            'Product_Type': product_data['Product_Type'],
            'Product_MRP': product_data['Product_MRP'],
            'Store_Size': product_data['Store_Size'],
            'Store_Location_City_Type': product_data['Store_Location_City_Type'],
            'Store_Type': product_data['Store_Type'],
            'Store_Age': store_age,
            'MRP_per_sqm': mrp_per_sqm,
            'MRP_x_Weight': mrp_x_weight
        }

        input_df = pd.DataFrame([sample])
        predicted_sales = round(float(model.predict(input_df)[0]), 2)

        return jsonify({'Predicted Sales (in dollars)': predicted_sales})

    except Exception as e:
        print("Error in /v1/sales:", e)
        return jsonify({"error": str(e)}), 500

# Batch prediction endpoint
@superkart_sales_api.post('/v1/salesbatch')
def predict_sales_batch():
    """
    Handles batch prediction for SuperKart product revenue from uploaded CSV.
    """
    try:
        file = request.files['file']
        input_data = pd.read_csv(file)

        # Derive features
        input_data["Store_Age"] = 2025 - input_data["Store_Establishment_Year"]
        input_data["MRP_per_sqm"] = input_data["Product_MRP"] / input_data["Product_Allocated_Area"]
        input_data["MRP_x_Weight"] = input_data["Product_MRP"] * input_data["Product_Weight"]

        # Drop unused columns if they exist
        input_data.drop(columns=["Product_Id", "Store_Id", "Store_Establishment_Year"], errors='ignore', inplace=True)

        # Model prediction
        predicted_sales = [round(float(p), 2) for p in model.predict(input_data)]

        # Use index as fallback key
        row_indices = input_data.index.tolist()
        output_dict = dict(zip(row_indices, predicted_sales))

        return jsonify(output_dict)

    except Exception as e:
        print("Error in /v1/salesbatch:", e)
        return jsonify({"error": str(e)}), 500

# Run the Flask app
if __name__ == '__main__':
    superkart_sales_api.run(debug=True)

# Hugging Face expects `app` by default
app = superkart_sales_api