🚀 Rebuilt backend with Flask + Docker

app.py

@@ -1,23 +1,68 @@
+# SuperKart - Sales Forecasting Flask API
 
 from flask import Flask, request, jsonify
 import pandas as pd
 import numpy as np
 import joblib
 import os
+import traceback
 
 app = Flask(__name__)
-model = joblib.load("best_model.pkl")
 
+# === Load Trained Model ===
+MODEL_PATH = "best_model.pkl"
+
+try:
+    model = joblib.load(MODEL_PATH)
+    print("✅ Model loaded successfully.")
+except Exception as e:
+    print("❌ Failed to load model from:", MODEL_PATH)
+    traceback.print_exc()
+
+# === Health Check ===
 @app.route("/", methods=["GET"])
 def health_check():
-    return "✅ SuperKart backend is up!", 200
+    return "✅ SuperKart backend is up and running!", 200
 
+# === Single Forecast Prediction ===
 @app.route("/v1/forecast", methods=["POST"])
-def predict():
-    data = request.get_json()
-    df = pd.DataFrame([data])
-    df["Store_Age"] = 2025 - df["Store_Establishment_Year"]
-    df["Price_per_kg"] = df["Product_MRP"] / df["Product_Weight"]
-    df["MRP_Band"] = pd.qcut(df["Product_MRP"], 3, labels=["Low", "Mid", "High"])
-    pred = np.expm1(model.predict(df)[0])
-    return jsonify({"Predicted_Sales": round(float(pred), 2)}), 200
+def predict_single():
+    try:
+        data = request.get_json()
+        df = pd.DataFrame([data])
+
+        df["Store_Age"] = 2025 - df["Store_Establishment_Year"]
+        df["Price_per_kg"] = df["Product_MRP"] / df["Product_Weight"]
+        df["MRP_Band"] = pd.cut(df["Product_MRP"], bins=[0, 100, 200, float("inf")], labels=["Low", "Mid", "High"])
+
+        pred_log = model.predict(df)[0]
+        pred = np.expm1(pred_log)
+
+        return jsonify({"Predicted_Sales": round(float(pred), 2)}), 200
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+# === Batch Forecast Prediction ===
+@app.route("/v1/forecastbatch", methods=["POST"])
+def predict_batch():
+    try:
+        file = request.files.get("file")
+        if file is None:
+            return jsonify({"error": "No file uploaded"}), 400
+
+        df = pd.read_csv(file)
+        df["Store_Age"] = 2025 - df["Store_Establishment_Year"]
+        df["Price_per_kg"] = df["Product_MRP"] / df["Product_Weight"]
+        df["MRP_Band"] = pd.cut(df["Product_MRP"], bins=[0, 100, 200, float("inf")], labels=["Low", "Mid", "High"])
+
+        preds = model.predict(df)
+        results = [round(float(np.expm1(p)), 2) for p in preds]
+
+        return jsonify({"Predictions": results}), 200
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+# === Entrypoint for Docker / HF Space ===
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 7860))
+    app.run(host="0.0.0.0", port=port)