8000 Restore previous sample_weight validation behaviour in BaseWeightBoosting by ryuwd · Pull Request #18935 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content

Restore previous sample_weight validation behaviour in BaseWeightBoosting #18935

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

Closed
wants to merge 1 commit into from

Conversation

ryuwd
Copy link
@ryuwd ryuwd commented Nov 27, 2020

Reference Issues/PRs

Fixes #18934

What does this implement/fix? Explain your changes.

Described in #18934

Support (or rather don't explicitly disallow) negative sample weights as long as their sum is not negative.

Any other comments?

I opened this PR in case this proposed trivial change is enough (i.e. restoring previous behaviour) to close the above issue.

Thank you all for this library, it's really neat.

@ryuwd
Copy link
Author
ryuwd commented Nov 27, 2020

It seems that based on the failing tests quite a few things depend on the behaviour currently in place.

On further digging, the approach proposed in #15531 (using a config parameter to allow negative weights) is a better and more mature alternative to the changes I have proposed here.

@ryuwd ryuwd closed this Nov 27, 2020
@ryuwd ryuwd deleted the patch-1 branch November 27, 2020 21:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

AdaBoostClassifier / BaseWeightBoosting no longer supports negative sample weights
1 participant
0