DOC move d2_log_loss_score in the classification metrics section #28938
Conversation
doc/modules/model_evaluation.rst
Outdated
| D2 log loss score | ||
| ----------------- | ||
|
|
||
| The :func:`d2_log_loss_score` function implements the special case | ||
| of D² with the log loss, see :ref:`log_loss`, i.e.: | ||
|
|
||
| .. math:: | ||
|
|
||
| \text{dev}(y, \hat{y}) = \text{log_loss}(y, \hat{y}). | ||
|
|
||
| The :math:`y_{\text{null}}` for the :func:`log_loss` is the per-class | ||
| proportion. |
There was a problem hiding this comment.
In regression metrics, we have a section about D2 scores in general and then subsections for each specific D2 score, see doc. Since we plan to add D2_brier_score_loss as well (#20943 (comment)), I think we should adopt the same structure.
There was a problem hiding this comment.
Wouldn't that look better when we also add D2_brier_score_loss? So maybe we could integrate the structure with that PR?
There was a problem hiding this comment.
The thing is that in its current form, this PR adds a section with terms that are not defined yet, like what's a D2 score, how's y_null related to the score. That's why I think we need to have the D2 introduction paragraph in this PR
There was a problem hiding this comment.
Okay sounds good.
|
@OmarManzoor please also add |
…kit-learn#28938) Co-authored-by: Omar Salman <omar.salman@arbisoft>
Reference Issues/PRs
Fixes #28933
What does this implement/fix? Explain your changes.
Any other comments?
CC: @jeremiedbb @lorentzenchr