Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix gain calculation in multi-target tree. #9978

Merged
merged 9 commits into from
Jan 17, 2024

Conversation

trivialfis
Copy link
Member

Close #9960 .

@trivialfis
Copy link
Member Author

@rongou I think the allgather is not working correctly for vector-leaf, will look into it after replacing the rabit module (still trying to get back to it).

@trivialfis trivialfis marked this pull request as draft January 11, 2024 09:43
@trivialfis
Copy link
Member Author

@rongou I have disabled the test for now. We need to refactor the evaluation to serialize the entries before synchronization.

@trivialfis trivialfis marked this pull request as ready for review January 15, 2024 11:34
@@ -149,6 +149,9 @@ class MultiTargetHistBuilder {
}

void InitData(DMatrix *p_fmat, RegTree const *p_tree) {
if (collective::IsDistributed()) {
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@hcho3 Added a check to disable the distributed training.

@trivialfis trivialfis merged commit cacb4b1 into dmlc:master Jan 17, 2024
29 checks passed
@trivialfis trivialfis deleted the multi-fix-gain branch January 17, 2024 05:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Mistake in vector leaf train
2 participants