Upload 4 files

Dockerfile

@@ -0,0 +1,26 @@
+# Use an official Python runtime as a parent image
+FROM python:3.9-slim-buster
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy the requirements file into the container at /app
+COPY requirements.txt .
+
+# Install any needed packages specified in requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the Flask application and the trained model into the container at /app
+COPY app.py .
+COPY best_sales_forecasting_model.joblib .
+
+# Expose the port that the Flask app will run on
+EXPOSE 5000
+
+# Define environment variables for Flask
+ENV FLASK_APP=app.py
+ENV FLASK_RUN_HOST=0.0.0.0
+ENV FLASK_RUN_PORT=5000
+
+# Run the Flask application
+CMD ["flask", "run"]

app.py

@@ -0,0 +1,80 @@
+from flask import Flask, request, jsonify
+import joblib
+import pandas as pd
+import numpy as np
+
+# Import necessary sklearn components for the pipeline to be reconstructed
+# These were part of the pipeline during training and are needed to load the model
+from sklearn.pipeline import Pipeline
+from sklearn.compose import ColumnTransformer
+from sklearn.preprocessing import OneHotEncoder
+from sklearn.ensemble import RandomForestRegressor # Or whichever regressor was used
+
+app = Flask(__name__)
+
+# Define the path to your serialized model
+MODEL_PATH = 'best_sales_forecasting_model.joblib'
+
+# Load the model when the application starts
+try:
+    model = joblib.load(MODEL_PATH)
+    print(f"Model loaded successfully from {MODEL_PATH}")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    model = None # Set model to None if loading fails
+
+@app.route('/predict', methods=['POST'])
+def predict():
+    if model is None:
+        return jsonify({'error': 'Model not loaded.'}), 500
+
+    try:
+        # Get JSON data from the request
+        data = request.get_json(force=True)
+
+        # Convert incoming data to a Pandas DataFrame
+        # Ensure the column order and names match the training data's original features
+        # before preprocessing.
+        # X_train_raw (from your notebook) had these columns originally:
+        # ['Product_Id', 'Product_Weight', 'Product_Sugar_Content', 'Product_Allocated_Area',
+        #  'Product_Type', 'Product_MRP', 'Store_Id', 'Store_Establishment_Year',
+        #  'Store_Size', 'Store_Location_City_Type', 'Store_Type', 'Store_Age']
+        # Ensure the input data has these columns.
+
+        # Example of expected input data structure:
+        # {
+        #     "Product_Id": ["FD6114"],
+        #     "Product_Weight": [12.66],
+        #     "Product_Sugar_Content": ["Low Sugar"],
+        #     "Product_Allocated_Area": [0.027],
+        #     "Product_Type": ["Frozen Foods"],
+        #     "Product_MRP": [117.08],
+        #     "Store_Id": ["OUT004"],
+        #     "Store_Establishment_Year": [2009],
+        #     "Store_Size": ["Medium"],
+        #     "Store_Location_City_Type": ["Tier 2"],
+        #     "Store_Type": ["Supermarket Type2"],
+        #     "Store_Age": [17] # Or calculate in app if Store_Establishment_Year is provided
+        # }
+
+        input_df = pd.DataFrame(data)
+
+        # Make prediction using the loaded model pipeline
+        prediction = model.predict(input_df)
+
+        # Return the prediction as JSON
+        return jsonify({'predicted_sales': prediction.tolist()})
+
+    except Exception as e:
+        return jsonify({'error': str(e)}), 400
+
+@app.route('/', methods=['GET'])
+def home():
+    return "Welcome to the SuperKart Sales Forecasting API! Send POST requests to /predict with your data."
+
+if __name__ == '__main__':
+    # In a Colab environment, you might want to use ngrok to expose your local server.
+    # For local testing, you can run:
+    # app.run(debug=True)
+    # For deployment, remove debug=True and configure a production-ready WSGI server.
+    app.run(host='0.0.0.0', port=5000)

best_sales_forecasting_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a1c15b842d2f51860705e60695f09787c082bb685c6b797bf5cb085e279b3f9
+size 37543978

requirements.txt

@@ -0,0 +1,6 @@
+flask
+pandas
+scikit-learn
+joblib
+numpy
+pyngrok