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 application
sales_predictor_api = Flask("SuperKart Sales Predictor")

# Load the trained machine learning model
model = joblib.load("sales_prediction_model_v1_0.joblib")

# Define a route for the home page (GET request)
@sales_predictor_api.get('/')
def home():
    """
    This function handles GET requests to the root URL ('/') of the API.
    It returns a simple welcome message.
    """
    return "Welcome to SalesKart Sales Prediction API!"

# Define an endpoint for single property prediction (POST request)
@sales_predictor_api.post('/v1/sales')
def predict_sales():
    """
    This function handles POST requests to the '/v1/sales' endpoint.
    It expects a JSON payload containing property details and returns
    the predicted rental price as a JSON response.
    """
    # Get the JSON data from the request body
    sales_data = request.get_json()

    # Extract relevant features from the JSON data
    sample = {
        'Product_Weight': sales_data['Product_Weight'],
        'Product_Sugar_Content': sales_data['Product_Sugar_Content'],
        'Product_Allocated_Area': sales_data['Product_Allocated_Area'],
        'Product_Type': sales_data['Product_Type'],
        'Product_MRP': sales_data['Product_MRP'],
        'Store_Size': sales_data['Store_Size'],
        'Store_Location_City_Type': sales_data['Store_Location_City_Type'],
        'Store_Type': sales_data['Store_Type'],
        'Store_Age': sales_data['Store_Age'],
        'Store_id': sales_data['Store_id']
    }

    # Convert the extracted data into a Pandas DataFrame
    input_data = pd.DataFrame([sample])

    # Make prediction (get log_sales)
    predictions = np.exp(model.predict(input_data)[0])

    # Round predictions
    predicted_sales = round(float(predictions), 2)
    # The conversion above is needed as we convert the model prediction (log price) to actual price using np.exp, which returns predictions as NumPy float32 values.
    # When we send this value directly within a JSON response, Flask's jsonify function encounters a datatype error

    # Return the actual price
    return jsonify({'Predicted Sales (in dollars)': predicted_sales})


# Define an endpoint for batch prediction (POST request)
#sales_predictor_api = Flask("SuperKart Sales Predictor")
@sales_predictor_api.post('/v1/salesbatch')
def sales_price_batch():
    """
    Endpoint to handle batch predictions from uploaded CSV.
    Returns predicted sales for each (Product_Id, Store_Id) pair.
    """
    try:
        # Get the uploaded CSV file
        file = request.files['file']

        # Load and process data
        input_data_batch = pd.read_csv(file)
        input_data_batch['Store_Age'] = 2025 - input_data_batch['Store_Establishment_Year']

        # Extract identifiers
        product_ids = input_data_batch['Product_Id'].tolist()
        store_ids = input_data_batch['Store_Id'].tolist()

        # Drop unused columns
        input_data_batch = input_data_batch.drop(['Product_Id', 'Store_Establishment_Year'], axis=1)

        # Apply preprocessing if needed
        # input_data_transformed = preprocessor.transform(input_data_batch)
        # predictions = model.predict(input_data_transformed)

        predictions = np.exp(model.predict(input_data_batch))  # Assuming already preprocessed or numeric

        # Round predictions
        predicted_sales = [round(float(x), 2) for x in predictions]

        # Structure output as a list of dicts
        output = [
            {
                "Product_Id": pid,
                "Store_Id": sid,
                "Predicted_Sales": psale
            }
            for pid, sid, psale in zip(product_ids, store_ids, predicted_sales)
        ]

        return jsonify({"predictions": output})

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


if __name__ == '__main__':
    sales_predictor_api.run(debug=True)