Data version control (DVC) is open-source, Git version control for machine learning projects. Benefits include:

• Reproducible and shareable machine learning models and pipelines
• Git version large datasets and models without Git-LFS
• Git diffs for model and data metrics across commits, tags and branches

The treeverse/setup-dvc action is a JavaScript action that sets up DVC in your workflow.

This action can be run on ubuntu-latest, macos-latest, windows-latest.

Basic usage:

steps:
  - uses: actions/checkout@v5

  - uses: treeverse/setup-dvc@v2

A specific version can be pinned to your workflow using the version argument.

steps:
  - uses: actions/checkout@v5

  - uses: treeverse/setup-dvc@v2
    with:
      version: '3.62.0'

The following inputs are supported.

• version - (optional) The version of DVC to install. A value of latest will install the latest version of DVC. Defaults to latest.

Setup DVC has no outputs.