[feature request] Support `resume` (and potentially other `tune.run` args)

ray-project / tune-sklearn

A drop-in replacement for Scikit-Learn’s GridSearchCV / RandomizedSearchCV -- but with cutting edge hyperparameter tuning techniques.

https://docs.ray.io/en/master/tune/api_docs/sklearn.html

Apache License 2.0

467 stars 51 forks source link

[feature request] Support `resume` (and potentially other `tune.run` args) #211

Closed Yard1 closed 3 years ago

Yard1 commented 3 years ago

From: https://discuss.ray.io/t/tunesearchcv-resume/2414

Is there a way to resume TuneSearchCV like with tune.run? I am using named log folders, but it seems like the resume parameter is missing–Is it possible to add? or would you recommend I try to find a workaround, like using the vanilla tune.run?

I think we could just have a tune_run_kwargs argument somewhere, probably in fit that would allow the user to pass any parameter they want to tune.run, including resume.