Get Started: Experiment Management
Chapters
-
Experiment tracking - Instrument your code to quickly start tracking experiments. Manage changes to code, data, metrics, parameters and plots associated with each experiment without bloating your Git repo.
-
Collaborating on experiments - Manage experiments the same way developers manage code. Share experiments with your team over git, create branches and pull requests, and compare results.
-
Experimenting using pipelines - Leverage DVC data pipelines as an experiment management system. Split your workflow into stages, track dependencies and outputs, and natively inject parameters and hyper-parameters to experiment runs.
These are captured in our example-get-started-experiments repo (see its tags).
Where To Go Next
Pick a page from the list above, the left-side navigation bar, or just click
NEXT below!
Click here to jump back to the Get Started landing page.