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.

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

Export AUC per class in json evaluation reports #545

Open
marcboulle opened this issue Jan 23, 2025 · 0 comments
Open

Export AUC per class in json evaluation reports #545

marcboulle opened this issue Jan 23, 2025 · 0 comments

Comments

@marcboulle
Copy link
Collaborator

Description

En classification supervisée, le critère d'évaluation AUC est calculé en train et en test et exporté dans le rapport d'evaluation

AUC: area under the ROC curve (AUC); evaluates the ordering of the predicted scores per target value. Note that the value calculated is the AUC ROC, not the AUC PR.
In case of multiple class values, the AUC is computed for each class value against all the others, then the results are weighted by the proportions of each class value.

En interne, on calcule donc l'AUC pour chaque classe.
Il serait pertinent d'ajouter dans les rapports l'effectif et l'AUC par classe.
A voir ensuite si on fait évoluer l'outil de visualisation pour exploiter ces nouveaux indicateurs.

Context

  • Khiops version: V11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant