Upload folder using huggingface_hub

Dockerfile

@@ -0,0 +1,25 @@
+
+Use a stable and slim Python base image
+FROM python:3.9-slim
+
+Set the working directory inside the container to /app
+WORKDIR /app
+
+Copy the requirements file first to leverage Docker's layer caching.
+This way, dependencies are only re-installed if requirements.txt changes.
+COPY requirements.txt .
+
+Install the Python dependencies specified in the requirements file
+RUN pip install --no-cache-dir -r requirements.txt
+
+Copy all other application files (app.py, model .joblib file, etc.) into the container
+COPY . .
+
+Expose the port that Hugging Face Spaces uses to run the application
+EXPOSE 7860
+
+Define the command to run the application using Gunicorn, a production-ready web server.
+-w 2: Starts 2 worker processes to handle incoming requests.
+-b 0.0.0.0:7860: Binds the server to port 7860, making it accessible from outside the container.
+app:app: Specifies to run the 'app' Flask instance found in the 'app.py' file.
+CMD ["gunicorn", "-w", "2", "-b", "0.0.0.0

app.py

@@ -0,0 +1,69 @@
+
+import os
+import joblib
+import pandas as pd
+from flask import Flask, request, jsonify
+
+# Initialize Flask app
+app = Flask("SuperKart Sales Predictor")
+
+# Load the trained model pipeline from the file
+# This is loaded only once when the application starts
+try:
+    model = joblib.load("superkart_sales_pipeline.joblib")
+    print("Model loaded successfully.")
+except FileNotFoundError:
+    print("Model file not found. Make sure 'superkart_sales_pipeline.joblib' is in the same directory.")
+    model = None
+except Exception as e:
+    print(f"An error occurred while loading the model: {e}")
+    model = None
+
+# Define a root endpoint for a health check
+@app.route('/', methods=['GET'])
+def home():
+    """A simple endpoint to confirm the API is running."""
+    return "Welcome to the SuperKart Sales Prediction API!"
+
+# Define the main endpoint for making sales predictions
+@app.route('/predict', methods=['POST'])
+def predict_sales():
+    """
+    Receives a JSON object with features and returns a sales prediction.
+    """
+    if model is None:
+        return jsonify({'error': 'Model is not loaded or failed to load.'}), 500
+
+    # Get the JSON data from the request body
+    input_data = request.get_json()
+    if not input_data:
+        return jsonify({'error': 'No input data provided.'}), 400
+
+    try:
+        # Convert the JSON data into a pandas DataFrame
+        # The `index=[0]` is crucial for creating a single-row DataFrame
+        features_df = pd.DataFrame(input_data, index=[0])
+
+        # Make a prediction using the full pipeline
+        # The pipeline handles all preprocessing steps (scaling, encoding, etc.)
+        prediction = model.predict(features_df)
+
+        # The prediction is a numpy array, so we extract the single value
+        predicted_value = prediction[0]
+
+        # Return the prediction in a JSON response, rounded to 2 decimal places
+        return jsonify({'predicted_sales': round(predicted_value, 2)})
+
+    except (KeyError, TypeError) as e:
+        # This catches errors if the input JSON is missing keys or malformed
+        return jsonify({'error': f'Invalid input data format: {str(e)}'}), 400
+    except Exception as e:
+        # A general catch-all for any other unexpected errors
+        return jsonify({'error': f'An unexpected error occurred: {str(e)}'}), 500
+
+# This block is for deployment environments like Hugging Face Spaces
+if __name__ == '__main__':
+    # The port is determined by the environment variable, defaulting to 7860
+    port = int(os.environ.get("PORT", 7860))
+    # Running on 0.0.0.0 makes the app accessible from outside the container
+    app.run(host='0.0.0.0', port=port)

requirements.txt

@@ -0,0 +1,10 @@
+# Flask Web Framework and Gunicorn Server
+Flask==2.2.2
+gunicorn==20.1.0
+
+# Core Data Science and ML Libraries
+pandas==2.2.2
+numpy==2.0.2
+scikit-learn==1.6.1
+xgboost==2.1.4
+joblib==1.4.2

superkart_sales_pipeline.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4576bc43ad42b7c3ece2fbe8c3f19293f9d2914e3f5f36e67a1abbc795086ff8
+size 24069091