WIP: Calibration plot #882
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def brier_score(y_true, y_prob): | ||
| """Compute the Brier score. | ||
|
|
||
| The smaller the Brier score, the better. |
There was a problem hiding this comment.
that's a bit against the scikit convention where all score functions want to be maximized
There was a problem hiding this comment.
Hum, you're right. I didn't pay attention because the name of this metric is Brier score.
|
Could you add some documentation on what a calibration plot is and what is it useful for? |
|
My commits have been included in @agramfort 's PR #1176, so I'm closing this one. |
Thanks for cleaning up. I find that it's challenging to keep track of |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I've started to play with calibration plots. I added an example:
examples/plot_calibration.py.I would like to get feedback from people who are familiar with the subject (especially @alextp and maybe @paolo-losi?).
On small and even not so small datasets, I'm often getting empty bins (meaning that some ranges of probabilities are never predicted by the classifier). To get smoother plots, I was thinking I could use polynomial regression but there's a risk of overfitting since the number of bins is small.
This is a work-in-progress: I won't have time to add tests and documentation anytime soon.