-
-
Notifications
You must be signed in to change notification settings - Fork 8.7k
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
[spark] Make spark model have the same UID with its estimator #9022
Conversation
Signed-off-by: Weichen Xu <[email protected]>
CC @wbo4958 @trivialfis Would you mind taking a look ? Thank you! |
Can we backport this fix to xgboost 1.7.6 version ? |
I don't have any plan for 1.7.6 at the moment, is this fix significant? In what scenario this can cause issues? |
Only some very edge case (a pyspark pipeline contains xgboost estimator, and crossvalidator tunes the pipeline, and we tunes some params related to xgboost model prediction). It is not an critical bug so 1.7.6 release is not urgent for . :) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@trivialfis Could you trigger all CI runs and then merge the PR once CI passed? Thankyou! |
Reminder: Backport this commit to 1.7 branch :) |
…tor (dmlc#9022) Signed-off-by: Weichen Xu <[email protected]>
…tor (dmlc#9022) Signed-off-by: Weichen Xu <[email protected]>
…tor (#9022) (#9285) Signed-off-by: Weichen Xu <[email protected]> Co-authored-by: WeichenXu <[email protected]>
Make spark model have the same UID with its estimator.
Because all other pyspark models obey the rule that uses the same UID with estimator, and the UID is used in param copying (e.g. CrossValidator copies param list to estimator, and UID is used for param owner validation), so we need to make the UID correct.