DEV Community

Gusti Triandi Winata
Gusti Triandi Winata

Posted on

Your Getting Started Template Repository for Continuous and Automated Machine Learning Project with GitHub Actions

A starting template for continuous ML projects where data kept growing, models kept learning and services deployed continuously.

You can use this template for:

  • Initiate your ML projects
  • Set .dvc to track data & save models
  • Set cml to automate training and track experiment by pull requests
  • Set model serving to endpoint and frontend example then deploy your models continuously

My Workflow

Motivation

! Mention:

  • model stuck in research
  • model didn't get to production, repetitive task on modelling
  • machine assignment for training, mention cheaphead machine that burns when using their own machine for training
  • experiment tracking for reliability
  • data and models that grows time to time that needs its own version control

Image description

Had to have a reliable getting started template for machine learning project since most of project that had data version control, experiment management, deployment as a service, serving to sample frontend and collaboration with a yaml hyperparameter tuning file to ease experiment tracking.

Image description

Deliver actions diagrams, how it works specifically and how it would help with automating the project on specific events on repository.

Image description

Using github script, definition, and how it compares to another framework

Image description

Submission Category:

DIY Deployments/Wacky Wildcards

Yaml File or Link to Code

GitHub logo sanggusti / cml-streamlit-template

A Template Repository for Continuous Machine Learning Project with GitHub Actions and display it live with deployment on Streamlit

cml-streamlit-template

A Template Repository for Continuous Machine Learning Project with GitHub Actions and display it live with deployment on Streamlit




Additional Resources / Info

[Note:] # (Be sure to link to any open source projects that are using your workflow!)

[Reminder]: # (Submissions are due on December 8th, 2021 (11:59 PM PT or 2 AM ET/6 AM UTC on December 9th).

Image of Timescale

🚀 pgai Vectorizer: SQLAlchemy and LiteLLM Make Vector Search Simple

We built pgai Vectorizer to simplify embedding management for AI applications—without needing a separate database or complex infrastructure. Since launch, developers have created over 3,000 vectorizers on Timescale Cloud, with many more self-hosted.

Read full post →

Top comments (0)

Billboard image

The Next Generation Developer Platform

Coherence is the first Platform-as-a-Service you can control. Unlike "black-box" platforms that are opinionated about the infra you can deploy, Coherence is powered by CNC, the open-source IaC framework, which offers limitless customization.

Learn more

đź‘‹ Kindness is contagious

Please leave a ❤️ or a friendly comment on this post if you found it helpful!

Okay