Giters

databricks / mlops-stacks

This repo provides a customizable stack for starting new ML projects on Databricks that follow production best-practices out of the box.

Home Page:https://docs.databricks.com/en/dev-tools/bundles/mlops-stacks.html

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Add Google Cloud Support

michael-chi opened this issue · comments

commented

Description

Today the MLOps Stack supports Azure and AWS, I'd like to add support to Google Cloud Platform.

Features

Users generate ML project from the MLOps Stack project templates which integrated with Databricks workspaces running on Google Cloud. Initial features including:

  • GitHub Actions as the CI/CD platform.
  • Ability to generate ML project from the cookiecutter template. Which allows the user to configure:
    • ML Project
    • CI/CD platform
    • Choose Google Cloud as the cloud provider
    • Enable Feature Store
    • Production / Staging workspaces
    • ML experiment parent directory
    • Github branches
  • Integration between GitHub Actions and Google Cloud.
    • Authentication through Workload Identity Pool.
    • Setup and configure required Google Cloud permissions / roles through Terraform.
  • Add required configuration and setup instructions to the readme.
commented

Hi @michael-chi, I saw you opened a PR to add this, thank you! Are you still interested in following up on this?