app.py

# Import necessary libraries
import numpy as np
import joblib  # For loading the serialized model
import pandas as pd  # For data manipulation
from flask import Flask, request, jsonify  # For creating the Flask API

# Initialize the Flask app
app = Flask(__name__)

model = joblib.load("backend_files/superkart_sales_prediction_model_v1_0.joblib")


# Define a route for the home page (GET request)
@store_sales_api.post('/v1/sale')
def predict_store_sales():
    """
    Handles POST requests for predicting sales of a single store.
    Expects JSON input with store details.
    """
    # Get JSON input
    store_data = request.get_json()

    # Extract relevant features (match your model training columns)
    sample = {
        'Product_Weight': store_data['Product_Weight'],
        'Product_Allocated_Area': store_data['Product_Allocated_Area'],
        'Product_MRP': store_data['Product_MRP'],
        'Store_Age': store_data['Store_Age'],
        'Product_Sugar_Content': store_data['Product_Sugar_Content'],
        'Product_Type': store_data['Product_Type'],
        'Store_Size': store_data['Store_Size'],
        'Store_Location_City_Type': store_data['Store_Location_City_Type'],
        'Store_Type': store_data['Store_Type'],
        'Store_Id': store_data['Store_Id']
    }

    # Convert into DataFrame
    input_data = pd.DataFrame([sample])

    # Make prediction
    predicted_sales = model.predict(input_data)[0]

    # Convert to Python float and round
    predicted_sales = round(float(predicted_sales), 2)

    # Return JSON response
    return jsonify({'Predicted Store Sales': predicted_sales})

# ...existing code...
# Run Flask app
if __name__ == '__main__':
    store_sales_api.run(debug=True)