Add the missing max_delta_step #3668
Merged
Conversation
This file contains 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
|
@henrygouk Can you take a look? |
|
Looks good to me. In future, constraints like this should probably be put into new SplitEvaluator subclasses to prevent a labyrinth of if statements from clogging up ElasticNet. That said, I think this particular constraint should be present in ElasticNet, since it is fixing backwards compatibility. |
khotilov
force-pushed
the
add_max_delta_step
branch
from
05cd504 to
3ada9b7
Compare
CodingCat
pushed a commit
to CodingCat/xgboost
that referenced
this pull request
Sep 18, 2018
* add max_delta_step to SplitEvaluator * test for max_delta_step * missing x2 factor for L1 term * remove gamma from ElasticNet
alois-bissuel
pushed a commit
to criteo-forks/xgboost
that referenced
this pull request
Dec 4, 2018
* add max_delta_step to SplitEvaluator * test for max_delta_step * missing x2 factor for L1 term * remove gamma from ElasticNet
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
The
max_delta_stepparameter fell through the cracks during the recent change toSplitEvaluatorfor thegrow_colmakerandgrow_fast_histmakerupdaters. It can be very useful for noisy datasets. I didn't see any noticeable timing performance change for the defaultmax_delta_step=0case.Also I've removed the unused reg_gamma parameter from ElasticNet, since it's only to be used during pruning.