import os import joblib import pandas as pd from flask import Flask, request, jsonify sales_predictor_api = Flask("Sales Predictor") # Load the trained model pipeline MODEL_PATH = os.path.join("deployment_files", "sales_predictor.joblib") model = joblib.load(MODEL_PATH) @sales_predictor_api.get("/") def home(): return "Welcome to the Sales Prediction API!" def validate_and_make_dataframe(sales_data: dict) -> pd.DataFrame: """ Validate required fields and convert a single JSON payload to a DataFrame. """ required_fields = [ "Product_Weight", "Product_Sugar_Content", "Product_Allocated_Area", "Product_Type", "Product_MRP", "Store_Establishment_Year_Before_2005", "Store_Size", "Store_Location_City_Type", "Store_Type", ] missing = [f for f in required_fields if f not in sales_data] if missing: raise KeyError(f"Missing required fields: {', '.join(missing)}") 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_Establishment_Year_Before_2005": sales_data["Store_Establishment_Year_Before_2005"], "Store_Size": sales_data["Store_Size"], "Store_Location_City_Type": sales_data["Store_Location_City_Type"], "Store_Type": sales_data["Store_Type"], } return pd.DataFrame([sample]) @sales_predictor_api.post("/v1/sales") def predict_sale(): try: sales_data = request.get_json(force=True, silent=False) input_df = validate_and_make_dataframe(sales_data) prediction = model.predict(input_df).tolist()[0] return jsonify({"Predicted_Sales": prediction}) except Exception as e: return jsonify({"error": str(e)}), 400 @sales_predictor_api.post("/v1/salesbatch") def predict_sale_batch(): try: file = request.files.get("file") if file is None: return jsonify({"error": "No file provided with form field 'file'"}), 400 input_df = pd.read_csv(file) predictions = model.predict(input_df).tolist() return jsonify({"Predicted_Sales": predictions}) except Exception as e: return jsonify({"error": str(e)}), 400 if __name__ == "__main__": port = int(os.environ.get("PORT", "7860")) sales_predictor_api.run(host="0.0.0.0", port=port, debug=False)