import joblib
import numpy as np
import pandas as pd
from fastapi import FastAPI, Request, UploadFile, File

superkart_api = FastAPI(title="Superkart Sales Prediction",
              description="API for predicting Superkart sales",
              version="0.116.1"
)
model = joblib.load('model_1.joblib')

@superkart_api.get('/')
def home():
  return "Welcome to SuperKart Sales Prediction API!"

@superkart_api.post('/v1/superkart_single')
async def salepred_single(request: Request):
  sales_data = await request.json()

  # Read input 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_Id':sales_data['Store_Id'],
      'Store_Size':sales_data['Store_Size'],
      'Store_Location_City_Type':sales_data['Store_Location_City_Type'],
      'Store_Type':sales_data['Store_Type'],

  }
  input_data = pd.DataFrame([sample])

  # Make predictions
  predicted_sale = model.predict(input_data)[0]

  # Create response
  response = {'Store_Outlet':sample['Store_Id'],"Sale":round(float(predicted_sale), 2)}
  return response

@superkart_api.post('/v1/superkart_batch')
async def salepred_batch(file: UploadFile = File(...)):

    # Read input data
    input_data = pd.read_csv(file.file)

    # Make predictions
    predicted_sale = model.predict(input_data).tolist()

    # Add predictions to input data
    input_data['Predicted_Sale'] = predicted_sale

    # Group by Store_Id and sum the predicted sales
    grouped_sales = input_data.groupby('Store_Id')['Predicted_Sale'].sum().to_dict()

    # Create response
    response = {
        'store_sales': {store_id: round(float(sale), 2) for store_id, sale in grouped_sales.items()}
    }
    print("Final Response:", response)

    return response