Upload folder using huggingface_hub

app.py

@@ -1,97 +1,39 @@
 # app.py — SuperKart Sales Forecaster Backend
 
-# ---------------------------------------------------------
-# Flask App Setup
-# ---------------------------------------------------------
-app = Flask(__name__)
-app.name = "SuperKart Sales Forecaster"
-
-# ---------------------------------------------------------
-# Model Files (relative paths inside Docker)
-# ---------------------------------------------------------
-MODEL_PKL = os.path.join("backend_files", "tuned_xgb_sales_forecaster.pkl")
-MODEL_JSON = os.path.join("backend_files", "tuned_xgb_sales_forecaster.json")
-
-# ---------------------------------------------------------
-# Feature Columns expected by the model pipeline
-# ---------------------------------------------------------
-FEATURE_COLS = [
-    'Product_Weight', 'Product_Sugar_Content', 'Product_Allocated_Area',
-    'Product_Type', 'Product_MRP', 'Store_Size',
-    'Store_Location_City_Type', 'Store_Type', 'Store_Age',
-    'Product_Category_Simplified'
-]
+from flask import Flask, request, jsonify
+import joblib
+import numpy as np
+import pandas as pd
 
-# ---------------------------------------------------------
-# Load the Serialized Model on Startup
-# ---------------------------------------------------------
-model_pipeline = None
+# Initialize Flask app
+app = Flask(__name__)
 
-try:
-    if os.path.exists(MODEL_PKL):
-        model_pipeline = joblib.load(MODEL_PKL)
-        print(f"Model pipeline loaded successfully from {MODEL_PKL}")
-    elif os.path.exists(MODEL_JSON):
-        xgb_model = XGBRegressor()
-        xgb_model.load_model(MODEL_JSON)
-        model_pipeline = xgb_model
-        print(f"XGBoost model loaded from {MODEL_JSON}")
-    else:
-        raise FileNotFoundError("No model file found in backend_files/")
-except Exception as e:
-    print(f"CRITICAL ERROR: Unable to load model → {e}")
+# === Load model ===
+model = joblib.load("tuned_xgb_sales_forecaster.pkl")
 
-# ---------------------------------------------------------
-# Root Route (Required by Hugging Face Spaces)
-# ---------------------------------------------------------
-@app.route("/", methods=["GET"])
+@app.route("/")
 def home():
-    return jsonify({
-        "message": f"Welcome to {app.name}! 🚀",
-        "available_endpoints": ["/health", "/predict"],
-        "status": "running"
-    })
-
-# ---------------------------------------------------------
-# Health Check Endpoint (for Docker + Hugging Face)
-# ---------------------------------------------------------
-@app.route("/health", methods=["GET"])
-def health_check():
-    status = "healthy" if model_pipeline is not None else "model_not_loaded"
-    return jsonify({"app": app.name, "status": status}), 200 if model_pipeline else 500
-
-# ---------------------------------------------------------
-# Prediction Endpoint
-# ---------------------------------------------------------
-@app.route('/predict', methods=['POST'])
-def predict_sales():
-    if model_pipeline is None:
-        return jsonify({'error': 'Model not loaded on server.'}), 500
+    return jsonify({"message": "SuperKart Sales Forecasting API is running!"})
 
+@app.route("/predict", methods=["POST"])
+def predict():
     try:
-        data = request.get_json(force=True)
-
-        if not isinstance(data, list) or len(data) == 0:
-            return jsonify({'error': 'Input data must be a non-empty list of objects.'}), 400
-
-        input_df = pd.DataFrame(data, columns=FEATURE_COLS)
-
-        log_prediction = model_pipeline.predict(input_df)
-        predicted_sales = np.expm1(log_prediction)
-
+        # Expecting JSON input with "features" list
+        data = request.get_json()
+        features = np.array(data["features"]).reshape(1, -1)
+        
+        prediction_log = model.predict(features)[0]
+        prediction_original = float(np.expm1(prediction_log))
+        
         return jsonify({
-            "status": "success",
-            "predicted_sales_revenue": round(float(predicted_sales[0]), 2)
+            "predicted_sales": prediction_original,
+            "status": "success"
         })
-
     except Exception as e:
-        print(f"Prediction error: {e}")
-        return jsonify({'error': f'Prediction failed: {str(e)}'}), 400
+        return jsonify({
+            "error": str(e),
+            "status": "failed"
+        }), 400
 
-# ---------------------------------------------------------
-# Docker Entrypoint
-# ---------------------------------------------------------
-if __name__ == '__main__':
-    port = int(os.environ.get("PORT", 7860))  # Hugging Face uses 7860
-    print(f"Starting {app.name} on port {port} ...")
-    app.run(host='0.0.0.0', port=port)
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=7860)