# Import necessary libraries
import joblib
import pandas as pd
from flask import Flask, request, jsonify

print("Starting SuperKart Flask API...")

# Initialize Flask app
app = Flask(__name__)

# Load trained model
try:
    model = joblib.load("superkart_sales_model.pkl")
    print("Model loaded successfully")
except Exception as e:
    print("Error loading model:", e)
    raise

# Home route (health check)
@app.route("/", methods=["GET"])
def home():
    return "✅ SuperKart Sales Prediction API is running!"

# ---------- SINGLE PREDICTION ----------
@app.route("/predict", methods=["POST"])
def predict_sales():
    data = request.get_json()

    sample = {
        "Product_Id": data["Product_Id"],
        "Product_Weight": data["Product_Weight"],
        "Product_Sugar_Content": data["Product_Sugar_Content"],
        "Product_Allocated_Area": data["Product_Allocated_Area"],
        "Product_Type": data["Product_Type"],
        "Product_MRP": data["Product_MRP"],
        "Store_Id": data["Store_Id"],
        "Store_Type": data["Store_Type"],
        "Store_Size": data["Store_Size"],
        "Store_Location_City_Type": data["Store_Location_City_Type"],
        "Store_Current_Age": data["Store_Current_Age"],
    }

    input_df = pd.DataFrame([sample])
    prediction = model.predict(input_df)[0]

    return jsonify({
        "Predicted_Sales": round(float(prediction), 2)
    })

# ---------- BATCH PREDICTION ----------
@app.route("/predict-batch", methods=["POST"])
def predict_sales_batch():
    file = request.files["file"]
    df = pd.read_csv(file)

    predictions = model.predict(df)
    predictions = [round(float(p), 2) for p in predictions]

    return jsonify({
        "Predicted_Sales": predictions
    })

# Run locally (Hugging Face ignores this but keeps it safe)
if __name__ == "__main__":
    app.run(host="0.0.0.0", port=7860)