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.

Sign up for GitHub

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

Jump to bottom

Automatically detect binary classification tasks and default to roc_auc #424

Open
Aylr opened this issue Nov 3, 2017 · 0 comments
Open

Automatically detect binary classification tasks and default to roc_auc #424

Aylr opened this issue Nov 3, 2017 · 0 comments
Assignees
@Aylr
Labels
bug med

Comments

@Aylr
Copy link
Contributor

Aylr commented Nov 3, 2017

Background

  • With the multiclass MVP all metrics were changed to default to accuracy. This isn't ideal.
  • The trainer should detect what kind of classification task and use accuracy if number_of_classes > 2 else use roc_auc
@Aylr Aylr added the bug med label Nov 3, 2017
@Aylr Aylr mentioned this issue Nov 6, 2017
Closed
@mmastand mmastand mentioned this issue Nov 10, 2017
Open
10 tasks
@Aylr Aylr modified the milestones: Sprint 37, Sprint 38 Dec 13, 2017
@Aylr Aylr modified the milestones: Sprint 38, Sprint 39 Jan 5, 2018
@Aylr Aylr removed this from the Sprint 39 milestone Jan 22, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@Aylr Aylr

Labels
bug med
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@Aylr