Upload folder using huggingface_hub

Dockerfile

@@ -0,0 +1,16 @@
+FROM python:3.9-slim
+
+# Set the working directory inside the container
+WORKDIR /app
+
+# Copy all files from the current directory to the container's working directory
+COPY . .
+
+# Install dependencies from the requirements file without using cache to reduce image size
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose API port
+EXPOSE 7860
+
+# Start Flask API with Gunicorn (4 workers, bind to 7860)
+CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:7860", "superkart_backend.app:sales_prediction_api"]

app.py

@@ -0,0 +1,98 @@
+import joblib
+import pandas as pd
+import numpy as np
+from flask import Flask, request, jsonify
+
+# -------------------------------
+# Initialize Flask App
+# -------------------------------
+sales_prediction_api = Flask('Superkart Sales Prediction')
+
+# -------------------------------
+# Load the trained Gradient Boosting model
+# (Make sure the path matches where you saved the .joblib file)
+# -------------------------------
+model = joblib.load("gradient_tuned.joblib")
+
+# -------------------------------
+# Root endpoint (Health check)
+# -------------------------------
+@sales_prediction_api.get("/")
+def home():
+    """Simple health check endpoint"""
+    return 'Welcome to the SuperKart Sales Prediction API'
+
+# -------------------------------
+# Single Prediction Endpoint
+# -------------------------------
+@sales_prediction_api.post("/v1/salesdata")
+def predict_sales():
+    """
+    This endpoint accepts a single JSON object with product & store details
+    and returns the predicted sales total.
+    """
+    # Parse incoming JSON
+    sales_data = request.get_json()
+
+    # Create a sample dictionary with only the features used by the model
+    sample = {
+        'Product_Weight': sales_data['Product_Weight'],
+        'Product_Allocated_Area': sales_data['Product_Allocated_Area'],
+        'Product_MRP': sales_data['Product_MRP'],
+        'Product_Sugar_Content': sales_data['Product_Sugar_Content'],  # fixed key
+        'Product_Type': sales_data['Product_Type'],
+        'Store_Id': sales_data['Store_Id'],
+        'Store_Size': sales_data['Store_Size'],
+        'Store_Location_City_Type': sales_data['Store_Location_City_Type'],
+    }
+
+    # Convert the dictionary into a pandas DataFrame (model expects a DataFrame)
+    input_data = pd.DataFrame([sample])
+
+    # Predict sales (model is trained on log of sales, so output is log-scale)
+    predicted_log_sales = model.predict(input_data)[0]
+
+    # Convert back from log scale to original sales units
+    predicted_sales = np.exp(predicted_log_sales)
+
+    # Round the sales value to 2 decimal places for readability
+    predicted_sales = round(float(predicted_sales), 2)
+
+    # Return prediction as JSON
+    return jsonify({'Product_Store_Sales_Total': predicted_sales})
+
+# -------------------------------
+# Batch Prediction Endpoint
+# -------------------------------
+@sales_prediction_api.post("/v1/salesdatabatch")
+def predict_sales_batch():
+    """
+    This endpoint accepts a CSV file with multiple product records
+    and returns predicted sales totals for all rows.
+    """
+    # Get uploaded CSV file from request
+    file = request.files['file']
+
+    # Load the CSV into a pandas DataFrame
+    input_data = pd.read_csv(file)
+
+    # Predict sales for each row (drop unused columns before prediction)
+    predicted_log_sales = model.predict(
+        input_data.drop(['Product_Id', 'Store_Establishment_Year', 'Store_Type'], axis=1)
+    ).tolist()
+
+    # Convert predictions back to original sales units
+    predicted_sales = [round(float(np.exp(log_price)), 2) for log_price in predicted_log_sales]
+
+    # Attach predictions to Product IDs for clarity
+    product_ids = input_data['Product_Id'].tolist()
+    output_dict = dict(zip(product_ids, predicted_sales))
+
+    # Return predictions as JSON
+    return jsonify(output_dict)
+
+# -------------------------------
+# Run Flask App
+# -------------------------------
+if __name__ == '__main__':
+    sales_prediction_api.run(debug=True)

gradient_tuned.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42b4d70fdf4fd9609e15aec0d5d8426f4a6b1d04b8d958f98de8c2227cfd097a
+size 1650258

requirements.txt

@@ -0,0 +1,10 @@
+pandas==2.2.2
+numpy==2.0.2
+scikit-learn==1.6.1
+xgboost==2.1.4
+joblib==1.4.2
+Werkzeug==2.2.2
+flask==2.2.2
+gunicorn==20.1.0
+requests==2.32.3
+uvicorn[standard]==0.32.0