Use MLflow experiment name instead of id for Aim experiment name

[stedn]

This would be just a "nice-to-have" if it's possible. Currently, the Aim experiment name is set to the experiment ID from mlflow. This makes it a little bit confusing because it's not always top-of-mind which experiment ID is associated with a given run.

Looking at the aim repo I couldn't see a way to set the experiment name separate from the experiment ID so this might not be possible, but thought I would check.

[stedn]

Also just wanted to add that I think this project is some great work and will be very helpful to my team using MLflow and wanting to incorporate some of the visually attractive cross-run comparison tools like you'd find in W&B etc. Thanks!

[alberttorosyan]

Hey @stedn!
I think it should be possible to set the experiment name during the conversion.
mlflow Python SDK has a method for getting the Experiment object by it's ID.

[mihran113]

Hey @stedn! Thanks for the suggestion, makes sense. This should be relatively easy to add, will include in the next release.

[ositaunus]

maybe stating the obvious: instead of 'instead', it would be more helpful to have both available - the name as the first thing you see, because it is more intuitive, and then with little inspection the ID to confirm it is the right experiment (sometimes names can be similar)

[sirykd]

with little inspection the ID to confirm it is the right experiment (sometimes names can be similar)

Should be covered by #14, which makes the experiment id accessible again via 'mlflow_experiment_id'