2

I was doing some experiments with MLflow using Python 3.7, and I was wondering if I can rename a stage or add a new one to the pre-existing

None | Staging | Production | Archived

Currently, I registered a model obtained from an experiment, but I would like to add multiple developing phases. I read the docs concerning that part, but the only thing I've found is the following:

Each distinct model version can be assigned one stage at any given time. MLflow provides predefined stages for common use-cases such as Staging, Production or Archived. You can transition a model version from one stage to another stage.

devilteo911
  • 51
  • 5
  • https://github.com/mlflow/mlflow/issues/2165 closed for similar request for now, waiting for new mlflow version to support your ask – punam Jan 18 '22 at 13:19

1 Answers1

0

Currently this is not possible but has been submitted as a feature request.

As a workaround you could add custom tags to your registered models and apply some logic in your code regarding these tags, but on the MLflow Model Registry they will still appear altogether on the same pre-defined stages.

Javier Beltrán
  • 128
  • 2
  • 10