DOC add additional pointers in Forest regarding how to use warm_start
#24579
Conversation
|
Hi @choo8, thanks for your pull request. |
|
Hi @cmarmo, I synchronized the pull request with the latest commits on the main branch and passed the tests. Please let me know if there is anything else I should do to get this pull request merged. Thanks! |
|
Thanks @choo8, all is green! :) |
sklearn/ensemble/_forest.py
Outdated
| # Only perform OOB computation again if there are newly grown trees or | ||
| # if it was not previously calculated due to oob_score parameter being False | ||
| if self.oob_score and ( | ||
| n_more_estimators > 0 or not hasattr(self, "oob_score_") | ||
| ): |
There was a problem hiding this comment.
Could we isolate this change in another PR? It will require writing some new unit tests. Since the documentation would be mergeable as-it, it would be the best way to go.
There was a problem hiding this comment.
Hi @glemaitre, thank you for your comments. As suggested, I have updated my PR to remove the changes to the OOB computation and only kept the changes to the documentation.
warm_start
|
Thanks @choo8 LGTM. Merging. |
I created the fixes according to the comments made by @NicolasHug #20435 (comment). It's my first time contributing to an open source project, so please let me know if there is anything I can do to improve my pull request. Thank You!
Reference Issues/PRs
Fixes #20435
What does this implement/fix? Explain your changes.
Removes unnecessary OOB computation when
n_more_estimators == 0and updates the docs for BaseForest-derived classes to link to https://scikit-learn.org/stable/modules/ensemble.html#fitting-additional-weak-learnersAny other comments?