search

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
465 stars 52 forks source link

Fix BOHB, change n_iter -> n_trials, fix up early stopping #81

Closed richardliaw closed 4 years ago

richardliaw commented 4 years ago

You can see the diff here: https://github.com/inventormc/tune-sklearn-1/compare/warm-start...richardliaw:bobh

Here are the main changes incorporated in the PR:

  1. We fix the BOHB example to actually leverage bohb properly.
  2. early stopping was not calling partial_fit, so that is also fixed.
  3. we rename n_iter to n_trials because n_iter is incredibly confusing (closes #80)
Yard1 commented 4 years ago

@richardliaw So just to confirm, BOHB requires partial fit or warm start to work correctly, yes?

richardliaw commented 4 years ago

Yeah, same with other early stopping mechanisms.