from flask import Flask, request, jsonify
import joblib
import numpy as np
import pandas as pd

# Initialize Flask app
store_sales_predictor_api = Flask("Super Kart Store Sales Predictor Application")

# Load the trained model
try:
    superkart_model = joblib.load("superkart_storesales_prediction_model_v1_0.joblib")
    print("Model loaded successfully.")
except FileNotFoundError:
    print("Error: 'superkart_storesales_prediction_model_v1_0.joblib' not found. Please train and save the model first.")
    superkart_model = None

# Define home page for app
@store_sales_predictor_api.get('/')
def home():
    return "Welcome to Super Kart Store Sales Predictor Application",200

# Health check endpoint
@store_sales_predictor_api.get('/healthcheck')
def health_check():
    """
    Returns a 200 status code and a JSON response to indicate the service is healthy.
    """
    return jsonify({"status": "healthy"}), 200
    
# Define prediction form page for app
@store_sales_predictor_api.post('/v1/predict')
def predict_sales_price():

    """
    Handles prediction requests.
    Expects a JSON payload with 'features'.
    """
    try:
        # Get data from the POST request
        payload = request.get_json()
    
        # Extract Relevant Features from Payload
        app_features = {
            "Product_Weight": payload["Product_Weight"],
            "Product_Sugar_Content": payload["Product_Sugar_Content"],
            "Product_Allocated_Area": payload["Product_Allocated_Area"],
            "Product_Type": payload["Product_Type"],
            "Product_MRP": payload["Product_MRP"],
            "Store_Establishment_Year": payload["Store_Establishment_Year"],
            "Store_Location_City_Type": payload["Store_Location_City_Type"],
            "Store_Id": payload["Store_Id"],  
            "Store_Type": payload["Store_Type"],
            "Store_Size": payload["Store_Size"]}    
        # store app_features in dataframe
        input_data = pd.DataFrame([app_features])
    
        # Make prediction and get store sales
        predicted_sales = superkart_model.predict(input_data)[0]
    
        # calculate actual value
        #predicted_sales_value = np.exp(predicted_sales)
    
        # convert value to python float
        predicted_sales_value = round(float(predicted_sales),2)
    
        return jsonify({"predicted store sales total": predicted_sales_value}), 200

    except Exception as e:
        return jsonify({"error": str(e)}), 500
  
    # Run the Flask app in debug mode
    if __name__ == '__main__':
      app.run(debug=True)