Upload folder using huggingface_hub

Dockerfile

@@ -0,0 +1,19 @@
+FROM python:3.9-slim
+
+# Set working directory inside the container
+WORKDIR /app
+
+# Copy requirements first (to leverage Docker cache for faster builds)
+COPY requirements.txt .
+
+# Install dependencies without cache
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+# Copy the rest of the application files
+COPY . .
+
+# Expose the application port
+EXPOSE 7860
+
+# Start the application with Gunicorn (4 workers)
+CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:7860", "app:sales_predictor_api"]

app.py

@@ -0,0 +1,76 @@
+import joblib
+import pandas as pd
+from flask import Flask, request, jsonify
+
+# Initialize flask app
+sales_predictor_api = Flask("Sales Predictor")
+
+# Load the trained sales prediction model
+model = joblib.load("sales_prediction_model_v1_0.joblib")
+
+# Home route
+@sales_predictor_api.get('/')
+def home():
+    return "Welcome to the Sales Prediction API"
+
+# Predict for a single product
+@sales_predictor_api.post('/v1/product')
+def predict_sales():
+    try:
+        # Get JSON data from the request
+        customer_data = request.get_json()
+
+        # Extract relevant features
+        sample = {
+            'Product_Weight': customer_data['Product_Weight'],
+            'Product_Sugar_Content': customer_data['Product_Sugar_Content'],
+            'Product_Allocated_Area': customer_data['Product_Allocated_Area'],
+            'Product_Type': customer_data['Product_Type'],
+            'Product_MRP': customer_data['Product_MRP'],
+            'Store_Id': customer_data['Store_Id'],
+            'Store_Establishment_Year': customer_data['Store_Establishment_Year'],
+            'Store_Size': customer_data['Store_Size'],
+            'Store_Location_City_Type': customer_data['Store_Location_City_Type'],
+            'Store_Type': customer_data['Store_Type'],
+            'Product_Id': customer_data['Product_Id']
+        }
+
+        # Convert to DataFrame
+        input_data = pd.DataFrame([sample])
+
+        # Optional: extract ID prefix if needed
+        input_data["Id"] = input_data["Product_Id"].str[:2]
+        input_data.drop("Product_Id", axis=1, inplace=True)
+
+        # Predict
+        prediction = model.predict(input_data).tolist()[0]
+
+        return jsonify({"Prediction": prediction})
+    
+    except Exception as e:
+        return jsonify({"error": str(e)}), 400
+
+# Predict for a batch of products
+@sales_predictor_api.post('/v1/productbatch')
+def predict_sales_batch():
+    try:
+        # Get uploaded CSV
+        file = request.files['file']
+        input_data = pd.read_csv(file)
+
+        input_data["Id"] = input_data["Product_Id"].str[:2]
+        cust_id_list = input_data["Product_Id"].tolist()
+        input_data.drop("Product_Id", axis=1, inplace=True)
+
+        # Predict
+        predictions = model.predict(input_data).tolist()
+        output_dict = dict(zip(cust_id_list, predictions))
+
+        return jsonify(output_dict)
+    
+    except Exception as e:
+        return jsonify({"error": str(e)}), 400
+
+# Run Flask app
+if __name__ == "__main__":
+    sales_predictor_api.run(debug=True)

requirements.txt

@@ -0,0 +1,11 @@
+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==2.28.1
+uvicorn[standard]
+streamlit==1.43.2

sales_prediction_model_v1_0.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84e263c9b1b8b97eb8fc23c58312e29906a3c10b25d7306d7f1d797adca82865
+size 13931434