Upload folder using huggingface_hub

Dockerfile

@@ -9,6 +9,14 @@ COPY . .
 # Install dependencies from the requirements file without using cache to reduce image size
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Create a directory for models
+RUN mkdir -p /app/models
+
+# Copy the XGBoost model files (these should be in the same directory as Dockerfile)
+COPY best_xgboost_model_20250821_220910.pkl /app/
+COPY xgboost_preprocessor_20250821_220910.pkl /app/
+COPY xgboost_metadata_20250821_220910.pkl /app/
+
 # Define the command to start the application using Gunicorn with 4 worker processes
 # - `-w 4`: Uses 4 worker processes for handling requests
 # - `-b 0.0.0.0:7860`: Binds the server to port 7860 on all network interfaces

app.py

@@ -5,66 +5,137 @@ from flask import Flask, request, jsonify
 # Initialize Flask app with a name
 app = Flask("SuperKart Sales Forecaster")
 
-# Load the trained sales forecasting model
-model = joblib.load("best_tuned_random_forest_model.pkl")
+# Load the trained XGBoost sales forecasting model
+try:
+    model = joblib.load("best_xgboost_model_20250821_220910.pkl")
+    preprocessor = joblib.load("xgboost_preprocessor_20250821_220910.pkl")
+    metadata = joblib.load("xgboost_metadata_20250821_220910.pkl")
+    print("✅ XGBoost model loaded successfully!")
+    print(f"Model: {metadata.get('model_name', 'XGBoost')}")
+    print(f"Training Date: {metadata.get('training_date', 'Unknown')}")
+except Exception as e:
+    print(f"❌ Error loading model: {e}")
+    model = None
+    preprocessor = None
+    metadata = None
 
 # Define a route for the home page
 @app.get('/')
 def home():
-    return "Welcome to the SuperKart Sales Forecasting API"
+    return "Welcome to the SuperKart Sales Forecasting API (XGBoost)"
 
 # Define an endpoint to predict sales for a single product
 @app.post('/v1/sales')
 def predict_sales():
-    # Get JSON data from the request
-    product_data = request.get_json()
-
-    # Extract relevant product features from the input data
-    sample = {
-        'Product_Weight': product_data['Product_Weight'],
-        'Product_Allocated_Area': product_data['Product_Allocated_Area'],
-        'Product_MRP': product_data['Product_MRP'],
-        'Store_Size': product_data['Store_Size'],
-        'Store_Location_City_Type': product_data['Store_Location_City_Type'],
-        'Store_Type': product_data['Store_Type'],
-        'Product_Type': product_data['Product_Type'],
-        'Product_Sugar_Content': product_data['Product_Sugar_Content']
-    }
-
-    # Convert the extracted data into a DataFrame
-    input_data = pd.DataFrame([sample])
-
-    # Make a sales prediction using the trained model
-    prediction = model.predict(input_data).tolist()[0]
-
-    # Return the prediction as a JSON response
-    return jsonify({'Predicted_Sales': f"₹{prediction:.2f}"})
+    if model is None:
+        return jsonify({'error': 'Model not loaded'}), 500
+    
+    try:
+        # Get JSON data from the request
+        product_data = request.get_json()
+
+        # Extract relevant product features from the input data
+        sample = {
+            'Product_Weight': product_data['Product_Weight'],
+            'Product_Allocated_Area': product_data['Product_Allocated_Area'],
+            'Product_MRP': product_data['Product_MRP'],
+            'Store_Size': product_data['Store_Size'],
+            'Store_Location_City_Type': product_data['Store_Location_City_Type'],
+            'Store_Type': product_data['Store_Type'],
+            'Product_Type': product_data['Product_Type'],
+            'Product_Sugar_Content': product_data['Product_Sugar_Content']
+        }
+
+        # Convert the extracted data into a DataFrame
+        input_data = pd.DataFrame([sample])
+
+        # Apply preprocessing if available
+        if preprocessor is not None:
+            processed_data = preprocessor.transform(input_data)
+        else:
+            processed_data = input_data
+
+        # Make a sales prediction using the trained XGBoost model
+        prediction = model.predict(processed_data).tolist()[0]
+
+        # Return the prediction as a JSON response
+        return jsonify({'Predicted_Sales': f"₹{prediction:.2f}"})
+    
+    except Exception as e:
+        return jsonify({'error': f'Prediction failed: {str(e)}'}), 500
 
 # Define an endpoint to predict sales for a batch of products
 @app.post('/v1/salesbatch')
 def predict_sales_batch():
-    # Get the uploaded CSV file from the request
-    file = request.files['file']
-
-    # Read the file into a DataFrame
-    input_data = pd.read_csv(file)
-
-    # Make predictions for the batch data
-    predictions = model.predict(input_data).tolist()
-
-    # Convert predictions to formatted strings
-    formatted_predictions = [f"₹{pred:.2f}" for pred in predictions]
-
-    # Create output dictionary with product IDs and predictions
-    if 'Product_Id' in input_data.columns:
-        product_ids = input_data['Product_Id'].values.tolist()
-    else:
-        product_ids = [f"Product_{i+1}" for i in range(len(predictions))]
-
-    output_dict = dict(zip(product_ids, formatted_predictions))
-
-    return output_dict
+    if model is None:
+        return jsonify({'error': 'Model not loaded'}), 500
+    
+    try:
+        # Get the uploaded CSV file from the request
+        file = request.files['file']
+
+        # Read the file into a DataFrame
+        input_data = pd.read_csv(file)
+
+        # Apply preprocessing if available
+        if preprocessor is not None:
+            processed_data = preprocessor.transform(input_data)
+        else:
+            processed_data = input_data
+
+        # Make predictions for the batch data
+        predictions = model.predict(processed_data).tolist()
+
+        # Convert predictions to formatted strings
+        formatted_predictions = [f"₹{pred:.2f}" for pred in predictions]
+
+        # Create output dictionary with product IDs and predictions
+        if 'Product_Id' in input_data.columns:
+            product_ids = input_data['Product_Id'].values.tolist()
+        else:
+            product_ids = [f"Product_{i+1}" for i in range(len(predictions))]
+
+        output_dict = dict(zip(product_ids, formatted_predictions))
+
+        return output_dict
+    
+    except Exception as e:
+        return jsonify({'error': f'Batch prediction failed: {str(e)}'}), 500
+
+# Health check endpoint
+@app.get('/health')
+def health_check():
+    if model is None:
+        return jsonify({
+            'status': 'unhealthy', 
+            'model_loaded': False,
+            'model_type': 'XGBoost'
+        }), 500
+    
+    return jsonify({
+        'status': 'healthy', 
+        'model_loaded': True,
+        'model_type': 'XGBoost',
+        'metadata': metadata
+    })
+
+# Model info endpoint
+@app.get('/model-info')
+def model_info():
+    if model is None:
+        return jsonify({'error': 'XGBoost model not loaded'}), 500
+    
+    return jsonify({
+        'model_type': 'XGBoost Regressor',
+        'model_loaded': True,
+        'preprocessor_loaded': preprocessor is not None,
+        'metadata': metadata
+    })
 
 # Run the Flask app in debug mode
 if __name__ == '__main__':
-    app.run(debug=True)
+    if model is not None:
+        print("�� Starting SuperKart Sales Forecasting API with XGBoost...")
+        app.run(debug=True, host='0.0.0.0', port=5000)
+    else:
+        print("❌ Cannot start API: Model not loaded")

best_xgboost_model_20250821_220910.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8431281a8287cd895c36a1942e1525c589a44b596abb2b79eefd93c86c749cda
+size 1416350

xgboost_metadata_20250821_220910.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe4a86554a6ec16fe2ad52eee8ac69418f6ea9cf3dbbe88fbb295f46d5d3f9c
+size 1197

xgboost_preprocessor_20250821_220910.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44de370800e33315d314c49a51096d1b76bfc248162eebe229c587585c534f8b
+size 70173