EXPERIMENT TRACKING

[marirapellini]

Hi,

I have created an experiment and inside I have different runs  

Is there the possibility to delete from a run a parameter that I logged previously?
thank you

[fsergot]

Hello,

It is not possible to change runs after their creation. From a UI perspective, if the parameter is annoying, you can unselect it from the view. But it will remain associated to the run.

[marirapellini]

Hi

thank you

do you know if I can delete a model?

I would like to delete s3_custom_model and mlflowwrapper from the view

thank you

[fsergot]

Unfortunately no either. You can delete the actual model files in the backing Managed folder if you want to reclaim space. But the reference to the model will remain here.

[marirapellini]

Thanks!

is there a way to retrieve the list of classes saved with set_run_inference_info

?

[fsergot]

Unfortunately this public API does not exist. I will create it for our backlog though.