from flask import Flask, request, jsonify
import pandas as pd
import joblib
import os
from backend_files.routes import welcome_message  # ✅ Corrected import

# Initialize Flask app
superkart_sales_predictor_api = Flask(__name__)

# Load model
model_path = os.path.join("backend_files", "superkart_model_prediction_model_v1_0.joblib")
model = joblib.load(model_path)

# Home route
@superkart_sales_predictor_api.get('/')
def home():
    return welcome_message()

# Single prediction route
@superkart_sales_predictor_api.post('/v1/superkart')
def predict_sales_total():
    product_data = request.get_json()

    sample = {
        'Product Type': product_data['Product Type'],
        'Product ID': product_data['Product ID'],
        'Product Weight': product_data['Product Weight'],
        'Product Sugar Content': product_data['Product Sugar Content'],
        'Product Allocated Area': product_data['Product Allocated Area'],
        'Product MRP': product_data['Product MRP'],
        'Store ID': product_data['Store ID'],
        'Store Establishment Year': product_data['Store Establishment Year'],
        'Store Size': product_data['Store Size'],
        'Store Location': product_data['Store Location'],
        'City Size': product_data['City Size'],
        'Store Type': product_data['Store Type']
    }

    input_df = pd.DataFrame([sample])
    prediction = model.predict(input_df)[0]
    return jsonify({'Predicted Product Store Sales Total': round(float(prediction), 2)})

# Batch prediction route
@superkart_sales_predictor_api.post('/v1/superkartbatch')
def predict_sales_total_batch():
    file = request.files['file']
    input_df = pd.read_csv(file)
    predictions = model.predict(input_df).tolist()
    product_ids = input_df['Product ID'].tolist()
    results = dict(zip(product_ids, [round(float(p), 2) for p in predictions]))
    return jsonify(results)

# Start app
if __name__ == '__main__':
    superkart_sales_predictor_api.run(debug=True, host='0.0.0.0', port=5000)