DOC Add quantile loss to user guide on HGBT regression #29063
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
glemaitre
reviewed
May 21, 2024
Co-authored-by: Guillaume Lemaitre <[email protected]>
ogrisel
reviewed
Jun 11, 2024
There was a problem hiding this comment.
Some feedback. I noticed that GradientBoostingRegressor has the "huber" loss which HistGradientBoostingRegressor does not provide and on the contrary, GradientBoostingRegressor is lacking "poisson" and "gamma".
The latter is kind of related to #16668 but I don't think we have an issue to add the huber loss to HistGradientBoostingRegressor (although I am not sure how useful it would be in practice because it does not really have any meaningful probablistic interpretation as far as I know.
Co-authored-by: Olivier Grisel <[email protected]>
glemaitre
approved these changes
Jul 23, 2024
|
Since the different comments have been addressed, I'm going to merge. Thanks @ArturoAmorQ |
MarcBresson
pushed a commit
to MarcBresson/scikit-learn
that referenced
this pull request
Sep 2, 2024
…29063) Co-authored-by: ArturoAmorQ <[email protected]> Co-authored-by: Guillaume Lemaitre <[email protected]> Co-authored-by: Olivier Grisel <[email protected]>
glemaitre
added a commit
to glemaitre/scikit-learn
that referenced
this pull request
Sep 9, 2024
…29063) Co-authored-by: ArturoAmorQ <[email protected]> Co-authored-by: Guillaume Lemaitre <[email protected]> Co-authored-by: Olivier Grisel <[email protected]>
glemaitre
added a commit
that referenced
this pull request
Sep 11, 2024
Co-authored-by: ArturoAmorQ <[email protected]> Co-authored-by: Guillaume Lemaitre <[email protected]> Co-authored-by: Olivier Grisel <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Reference Issues/PRs
Follows #28652.
What does this implement/fix? Explain your changes.
As mentioned in #28652 (comment), the description of HGBT regression losses was missing the quantile loss. This PR fixes the issue.
Any other comments?
In the same comment it is mentioned that
For a follow-up PR maybe we can consider refactoring the narrative to first introduce common mathematical aspects of GBT models and then the HGBT version. WDYT?