Build CI/CD Pipeline for ML Models using Docker and GitHub Actions

This repository contains code and files for building and deploying a Machine Learning model for predicting the survival of Titanic passengers using Docker and GitHub Actions.

Overview

The project consists of the following components:

titanic_model: A Python package that contains the data processing and model training scripts for the Titanic survival prediction task.
titanic_model_api: A Flask web application that serves the trained model as a RESTful API endpoint.
Dockerfile: A file that defines the steps and commands for creating a Docker image of the web application.
.github/workflows: A folder that contains the GitHub Actions workflow file for automating the Docker image building and pushing process.
tests: A folder that contains unit tests and integration tests for the Python package and the web application.
requirements: A file that lists the Python dependencies for the project.
setup.py: A file that defines the metadata and installation instructions for the Python package.
MANIFEST.in: A file that specifies the non-code files to be included in the Python package distribution.
mypy.ini: A file that configures the type checking tool mypy for the project.
pyproject.toml: A file that defines the formatting tool black for the project.

Installation

To install the Python package, run the following command from the root directory of the repository:

pip install .

To build the Docker image, run the following command from the root directory of the repository:

docker build -t titanic-model-api .

To run the Docker container, run the following command:

docker run -p 5000:5000 titanic-model-api

Usage

To use the web application, send a POST request to the /predict endpoint with a JSON payload containing the features of a Titanic passenger. For example:

curl -X POST -H "Content-Type: application/json" -d '{"Pclass": 3, "Sex": "male", "Age": 22, "SibSp": 1, "Parch": 0, "Fare": 7.25, "Embarked": "S"}' http://localhost:5000/predict

The response will be a JSON object containing the predicted probability of survival for the passenger. For example:

{"survival_probability": 0.087}

Testing

To run the unit tests and integration tests, run the following command from the root directory of the repository:

pytest

To run the type checking and formatting tools, run the following commands:

mypy titanic_model titanic_model_api
black titanic_model titanic_model_api

Contributing

If you have a Data Science mini-project that you'd like to share, please follow the guidelines in CONTRIBUTING.md.

Code of Conduct

Please adhere to our Code of Conduct in all your interactions with the project.

License

This project is licensed under the MIT License.

Contact

For questions or inquiries, feel free to contact me on Linkedin.

About Me:

I’m a seasoned Data Scientist and founder of TowardsMachineLearning.Org. I've worked on various Machine Learning, NLP, and cutting-edge deep learning frameworks to solve numerous business problems.

Praveen76/Build-CI-CD-pipeline-for-ML-models-using-Docker-and-GitHub-Actions