Projects

Projects let you organize your team's model-related resources, i.e. Models, Experiments, and Model Evaluations. Create a separate project for each use case or ML application you are working on. Whenever you register a new Experiment or Model Evaluation with MarkovML, you'll need to specify which project it belongs to.

Projects are public by default, but can be made private. Public projects are visible to all workspace members, whereas private projects are only visible to the user who created them. The Models and related resources associated with a project inherit the visibility of the project.

Datasets registered with MarkovML exist outside of any single project. This way, you can register a dataset once and reuse it across multiple projects, which may focus on different use cases.