Skip to content

[ML] Reduce variability in regression and classification results across our target platforms #1127

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

Merged
merged 13 commits into from
Apr 9, 2020

Conversation

tveasey
Copy link
Contributor

@tveasey tveasey commented Apr 7, 2020

This addresses some of the underlying causes for variation in regression and classification results for different platforms: std::log and std::exp don't always agree in the last decimal place and our split order predicate wasn't a total order.

Copy link
Contributor

@droberts195 droberts195 left a comment

Choose a reason for hiding this comment

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

Looks good as long as there are no nasty edge cases. Maybe there aren't but a unit test of the extremes would prove it.

Copy link
Contributor

@droberts195 droberts195 left a comment

Choose a reason for hiding this comment

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

LGTM

@tveasey tveasey changed the title [ML] Regression and classification should produce identical results for all target OS [ML] Reduce differences in regression and classification results between target platforms Apr 9, 2020
@tveasey tveasey merged commit 12b5014 into elastic:master Apr 9, 2020
@tveasey tveasey deleted the os-differences branch April 9, 2020 12:57
@tveasey tveasey changed the title [ML] Reduce differences in regression and classification results between target platforms [ML] Reduce variability in regression and classification results across our target platforms Apr 9, 2020
tveasey added a commit to tveasey/ml-cpp-1 that referenced this pull request Apr 17, 2020
tveasey added a commit that referenced this pull request Apr 17, 2020
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.

2 participants