[MRG] Support for list of lists or list of arrays multilabel indicator (continuation) #14865

leonardbinet · 2019-09-01T22:30:08Z

Reference Issues/PRs

Rebase of #9158 to solve #7931

What does this implement/fix? Explain your changes.

Any other comments?

sklearn/utils/estimator_checks.py

jnothman · 2019-09-01T23:38:17Z

Yes, let's do this!

jnothman

Test failures...

sklearn/utils/multiclass.py

leonardbinet · 2019-09-02T21:40:20Z

@jnothman, fixed it

jnothman

Thanks, it's looking pretty good.

Please add an Enhancement entry to the change log at doc/whats_new/v0.22.rst. Like the other entries there, please reference this pull request with :pr: and credit yourself (and other contributors if applicable) with :user:

sklearn/preprocessing/tests/test_label.py

sklearn/metrics/tests/test_common.py

sklearn/tree/tests/test_tree.py

sklearn/utils/tests/test_estimator_checks.py

jnothman · 2019-09-08T00:26:57Z

Please append commits rather than force pushing. Much easier to see changes between reviews that way

doc/whats_new/v0.22.rst

jnothman · 2019-09-08T00:30:01Z

doc/whats_new/v0.22.rst

+  :class:`preprocessing.KernelCenterer`
  :pr:`14336` by :user:`Gregory Dexter <gdex1>`.

+- |Enhancement| label_binarize now supports list of lists for multilabel data


Might be more helpful to advertise the changes in terms of how they affect metrics and other users of type_of_target

how would you put it? one entry per impacted function or a more generic explanation on what is considered as multiclass-multioutput/multilabel-indicator?

jnothman · 2019-09-08T00:31:47Z

I've not looked at the code or codecov's complaint this time

jnothman · 2019-09-08T20:53:02Z

I think this mostly affects sklearn.metrics, so commenting there that "multilabel metrics now accept ..." would do; or add a "multiple modules" section to the change log.

sklearn/preprocessing/tests/test_label.py

sklearn/metrics/tests/test_common.py

doc/whats_new/v0.22.rst

leonardbinet · 2019-09-09T22:41:50Z

thanks for your help and patience @jnothman, you're the victim of my first contribution to scikit-learn 😅

jnothman · 2019-09-09T23:18:03Z

I'm very happy you're finishing this off. It's something we should have done years ago, having removed sequence of sequence support around 2015

jnothman · 2019-09-10T01:48:59Z

sklearn/preprocessing/tests/test_label.py:27:1: F401 'sklearn.utils.testing.assert_raises' imported but unused
from sklearn.utils.testing import assert_raises, assert_raise_message

jnothman

Otherwise lgtm

sklearn/preprocessing/tests/test_label.py

jnothman · 2019-09-10T08:13:53Z

It looks like you've introduced a syntax error. And it's worth looking at the codecov error

jnothman · 2019-09-10T08:15:04Z

Maybe merging in the latest master will fix the codecov issue

NicolasHug

Thanks @leonardbinet , only super nitpicks from me

The PR has merge conflicts but apart from that LGTM

sklearn/utils/multiclass.py

sklearn/utils/estimator_checks.py

NicolasHug · 2019-10-16T15:40:57Z

sklearn/utils/multiclass.py

    'multiclass-multioutput'
    >>> type_of_target([[1, 2]])
-    'multiclass-multioutput'
+    'multilabel-indicator'


question: is this a bugfix?

It is more-or-less a bug fix. Either way, type_of_target is expected to return "the most specific type that can be inferred". Multilabel is a specific subtype of multiclass-multioutput. So this is correct

jnothman · 2019-10-23T01:45:07Z

@leonardbinet please resolve conflicts with master and address @NicolasHug's teeny weeny comments so that we can merge this for release. If that's looking unlikely in the next week or so, please let us know.

leonardbinet · 2019-10-24T20:26:29Z

@jnothman @NicolasHug sure, thanks for your feedbacks I'll fix this this week-end

…x_is_multilabel

jnothman · 2019-10-25T00:35:01Z

@leonardbinet, force pushing makes it much harder for reviewers to track what changes have occurred since last review.

jnothman · 2019-10-25T00:37:43Z

Thank you @leonardbinet, @herilalaina and @srivatsan-ramesh!!

leonardbinet mentioned this pull request Sep 1, 2019

[MRG] Support for list of lists or list of arrays multilabel indicator (continuation) #9158

Closed

leonardbinet commented Sep 1, 2019

View reviewed changes

sklearn/utils/estimator_checks.py Outdated Show resolved Hide resolved

jnothman reviewed Sep 2, 2019

View reviewed changes

sklearn/utils/multiclass.py Outdated Show resolved Hide resolved

leonardbinet force-pushed the fix_is_multilabel branch from 064ce44 to 04c4265 Compare September 2, 2019 19:40

leonardbinet changed the title ~~[WIP] Support for list of lists or list of arrays multilabel indicator (continuation)~~ [MRG] Support for list of lists or list of arrays multilabel indicator (continuation) Sep 2, 2019

jnothman added High Priority High priority issues and pull requests Waiting for Reviewer labels Sep 5, 2019

jnothman reviewed Sep 5, 2019

View reviewed changes

leonardbinet force-pushed the fix_is_multilabel branch 3 times, most recently from 0d3797a to e18e836 Compare September 7, 2019 10:14

jnothman mentioned this pull request Sep 8, 2019

Precision Recall and F-score: behavior when all negative #14876

Closed

jnothman reviewed Sep 8, 2019

View reviewed changes

leonardbinet force-pushed the fix_is_multilabel branch from 589c621 to 42bd1f5 Compare September 8, 2019 17:58

jnothman reviewed Sep 9, 2019

View reviewed changes

sklearn/preprocessing/tests/test_label.py Outdated Show resolved Hide resolved

sklearn/metrics/tests/test_common.py Show resolved Hide resolved

doc/whats_new/v0.22.rst Outdated Show resolved Hide resolved

jnothman approved these changes Sep 10, 2019

View reviewed changes

sklearn/preprocessing/tests/test_label.py Outdated Show resolved Hide resolved

srivatsan-ramesh added 4 commits September 10, 2019 23:53

added list of list and list of array support for multilabel indicator

358909d

tests for multilabel list-list or list-array indicator

79e09a5

changed an example in type_of_target

b15bb17

Test for multilabel-indicator representation invariance in OvR

742def3

alk-lbinet and others added 11 commits September 10, 2019 23:53

remove deprecated test

09f1f3e

remove multioutput tag

e819614

remove deprecated test skip

0df8ab1

rst doc

06177ec

is_multilabel tests for lists of lists

1c0e094

include other contributors in credits

4f10aac

update changelog

2997d1e

update changelog

ad73ccd

pytest preferred

98afb43

reintroduce test, without first now-valid sample

ff529ab

update test_label to use preferred pytest.raises

51b127a

leonardbinet force-pushed the fix_is_multilabel branch from 321f598 to 51b127a Compare September 10, 2019 21:58

NicolasHug approved these changes Oct 16, 2019

View reviewed changes

leonardbinet force-pushed the fix_is_multilabel branch 7 times, most recently from f0d8b94 to a71cac1 Compare October 24, 2019 22:34

Merge branch 'master' of github.com:scikit-learn/scikit-learn into fi…

84f2681

…x_is_multilabel

leonardbinet force-pushed the fix_is_multilabel branch from a71cac1 to 84f2681 Compare October 24, 2019 22:37

remove unnecessary line

9f73bf2

jnothman merged commit 846e6a3 into scikit-learn:master Oct 25, 2019

jnothman mentioned this pull request Oct 27, 2019

Multilabel as list of lists or list of arrays should be supported #7931

Closed

cmarmo mentioned this pull request Mar 23, 2022

BUG: sklearn/utils/multiclass.type_of_target inconsistent for similar inputs #14671

Closed

Uh oh!

[MRG] Support for list of lists or list of arrays multilabel indicator (continuation) #14865

[MRG] Support for list of lists or list of arrays multilabel indicator (continuation) #14865

Uh oh!

Conversation

leonardbinet commented Sep 1, 2019

Reference Issues/PRs

What does this implement/fix? Explain your changes.

Any other comments?

Uh oh!

Uh oh!

jnothman commented Sep 1, 2019

Uh oh!

jnothman left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

leonardbinet commented Sep 2, 2019

Uh oh!

jnothman left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jnothman commented Sep 8, 2019

Uh oh!

Uh oh!

jnothman Sep 8, 2019

Choose a reason for hiding this comment

Uh oh!

leonardbinet Sep 8, 2019

Choose a reason for hiding this comment

Uh oh!

jnothman commented Sep 8, 2019

Uh oh!

jnothman commented Sep 8, 2019 via email

Uh oh!

Uh oh!

Uh oh!

Uh oh!

leonardbinet commented Sep 9, 2019

Uh oh!

jnothman commented Sep 9, 2019

Uh oh!

jnothman commented Sep 10, 2019

Uh oh!

jnothman left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

jnothman commented Sep 10, 2019

Uh oh!

jnothman commented Sep 10, 2019

Uh oh!

NicolasHug left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

NicolasHug Oct 16, 2019

Choose a reason for hiding this comment

Uh oh!

jnothman Oct 17, 2019

Choose a reason for hiding this comment

Uh oh!

jnothman commented Oct 23, 2019

Uh oh!

leonardbinet commented Oct 24, 2019

Uh oh!

jnothman commented Oct 25, 2019

Uh oh!

jnothman commented Oct 25, 2019

Uh oh!

Uh oh!