Using DVC Commands

New! DVC is also available for the VS Code IDE, which adds many DVC operations to the Command Palette.

The core of DVC is a command line tool. These pages contain the specifications, self-contained descriptions, and comprehensive usage examples for dvc commands. Use dvc -h to list them.

To run DVC in a specific directory, use dvc --cd <path> ... before the command and its options/arguments (does not change directories in your terminal).

Typical DVC workflow

Initialize a DVC project in a Git repo with dvc init.
Copy data files or dataset directories for modeling into the project and use dvc add to tell DVC to cache and track them.
Create a simple dvc.yaml file to codify a data processing pipeline. It uses your own source code and specifies further data outputs for DVC to control.
Execute or restore any version of your pipeline using dvc repro, or experiment on it with dvc exp features.
Sharing the repository will not include locally cached data. Use remote storage with dvc push and dvc pull to share data artifacts.

Using DVC Commands

Typical DVC workflow

Content