[Refactor] Classification 6/n #1163
Merged
Conversation
This file contains 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
SkafteNicki
added
enhancement
Codecov Report
@@ Coverage Diff @@
## devel/classification #1163 +/- ##
=====================================================
- Coverage 90% 90% -1%
=====================================================
Files 187 189 +2
Lines 9774 10543 +769
=====================================================
+ Hits 8843 9469 +626
- Misses 931 1074 +143 |
justusschock
approved these changes
Aug 18, 2022
4 tasks
Borda
pushed a commit
that referenced
this pull request
Aug 26, 2022
* some idea * implementation stuff * error message * somethings working * working * working multilabel precision recall * docstrings * working average precision * working roc * working auroc * working precision at recall * docs * init files * beginning doctest * more docs * more docs * more docs * more docs * more docs * more docs * correction to math * fix mypy * add suggestions * change default from 100 to None * fix * fix * some fixes * suggestions for stancld * try fixing * try fix * try again * more fixing * another fix * skip half + cpu test for old versions * fix link * another fix * nan safety * another fix * skip non working cpu half tests * skip non working cpu + half tests
4 tasks
Borda
pushed a commit
that referenced
this pull request
Sep 12, 2022
* some idea * implementation stuff * error message * somethings working * working * working multilabel precision recall * docstrings * working average precision * working roc * working auroc * working precision at recall * docs * init files * beginning doctest * more docs * more docs * more docs * more docs * more docs * more docs * correction to math * fix mypy * add suggestions * change default from 100 to None * fix * fix * some fixes * suggestions for stancld * try fixing * try fix * try again * more fixing * another fix * skip half + cpu test for old versions * fix link * another fix * nan safety * another fix * skip non working cpu half tests * skip non working cpu + half tests
Borda
pushed a commit
that referenced
this pull request
Sep 12, 2022
* some idea * implementation stuff * error message * somethings working * working * working multilabel precision recall * docstrings * working average precision * working roc * working auroc * working precision at recall * docs * init files * beginning doctest * more docs * more docs * more docs * more docs * more docs * more docs * correction to math * fix mypy * add suggestions * change default from 100 to None * fix * fix * some fixes * suggestions for stancld * try fixing * try fix * try again * more fixing * another fix * skip half + cpu test for old versions * fix link * another fix * nan safety * another fix * skip non working cpu half tests * skip non working cpu + half tests
Borda
pushed a commit
that referenced
this pull request
Sep 13, 2022
* some idea * implementation stuff * error message * somethings working * working * working multilabel precision recall * docstrings * working average precision * working roc * working auroc * working precision at recall * docs * init files * beginning doctest * more docs * more docs * more docs * more docs * more docs * more docs * correction to math * fix mypy * add suggestions * change default from 100 to None * fix * fix * some fixes * suggestions for stancld * try fixing * try fix * try again * more fixing * another fix * skip half + cpu test for old versions * fix link * another fix * nan safety * another fix * skip non working cpu half tests * skip non working cpu + half tests
Borda
pushed a commit
that referenced
this pull request
Sep 13, 2022
* some idea * implementation stuff * error message * somethings working * working * working multilabel precision recall * docstrings * working average precision * working roc * working auroc * working precision at recall * docs * init files * beginning doctest * more docs * more docs * more docs * more docs * more docs * more docs * correction to math * fix mypy * add suggestions * change default from 100 to None * fix * fix * some fixes * suggestions for stancld * try fixing * try fix * try again * more fixing * another fix * skip half + cpu test for old versions * fix link * another fix * nan safety * another fix * skip non working cpu half tests * skip non working cpu + half tests
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.
What does this PR do?
Continues work on classification refactor #1001
Prior work:
This PR takes care of
precision_recall_curve,roc,auroc,average_precision,recall_at_fixed_precision. Importantly this PR brings a significant improvement since all these metrics can now be calculated using constant memory instead of relying on saving all predictions and targets in memory. It is controlled through the newthresholdsargument that can either be set toNonewhich corresponds to the old behaviour or to anint(or an list if user want to specify the thresholds themself) that will calculate the metric using an approach similar to binned metrics we have.Before submitting
PR review
Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.
Did you have fun?
Make sure you had fun coding 🙃