app.py

import joblib
import pandas as pd
from flask import Flask, request, jsonify
import traceback

# Initialize Flask app
product_sales_predictor_api = Flask("Product Sales Predictor")

# Manual CORS headers
'''
@product_sales_predictor_api.after_request
def after_request(response):
    response.headers.add('Access-Control-Allow-Origin', '*')
    response.headers.add('Access-Control-Allow-Headers', 'Content-Type,Authorization')
    response.headers.add('Access-Control-Allow-Methods', 'GET,PUT,POST,DELETE')
    return response
'''

# Load the trained model
model = joblib.load("model/product_sales_prediction_model_v1_0.joblib")

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

# Single product prediction with debug
@product_sales_predictor_api.post('/v1/product')
def predict_single_product():
    try:
        # Get JSON data from request
        product_info = request.get_json()
        print(f"Received data: {product_info}")

        # Extract features
        prod_input = {
            'Product_Weight': product_info['Product_Weight'],
            'Product_Sugar_Content': product_info['Product_Sugar_Content'],
            'Product_Allocated_Area': product_info['Product_Allocated_Area'],
            'Product_Type': product_info['Product_Type'],
            'Product_MRP': product_info['Product_MRP'],
            'Store_Id': product_info['Store_Id'],
            'Store_Establishment_Year': product_info['Store_Establishment_Year'],
            'Store_Size': product_info['Store_Size'],
            'Store_Location_City_Type': product_info['Store_Location_City_Type'],
            'Store_Type': product_info['Store_Type']
        }

        print(f"Processed input: {prod_input}")

        # Convert to DataFrame and predict
        input_data = pd.DataFrame([prod_input])
        print(f"DataFrame shape: {input_data.shape}")
        print(f"DataFrame columns: {input_data.columns.tolist()}")

        prediction = model.predict(input_data)[0]
        print(f"Prediction: {prediction}")

        return jsonify({'predicted_sales': float(round(prediction, 2))})

    except Exception as e:
        print(f"Error: {str(e)}")
        print(f"Traceback: {traceback.format_exc()}")
        return jsonify({'error': str(e)}), 500

# Batch prediction
@product_sales_predictor_api.post('/v1/products')
def predict_multiple_products():
    try:
        # Get uploaded CSV file
        input_file = request.files['file']
        input_data = pd.read_csv(input_file)

        # Remove Product_Id if present
        prediction_data = input_data.drop(['Product_Id'], axis=1, errors='ignore')

        # Make predictions
        predictions = model.predict(prediction_data).tolist()
        predictions = [round(pred, 2) for pred in predictions]

        # Create output dictionary
        if 'Product_Id' in input_data.columns:
            prod_id_list = input_data['Product_Id'].tolist()
            result = dict(zip(prod_id_list, predictions))
        else:
            result = {f"product_{i+1}": pred for i, pred in enumerate(predictions)}

        return jsonify(result)

    except Exception as e:
        return jsonify({'error': str(e)}), 500