Skip to content

/ scikit-learn

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

Update fbeta_score to accept beta=0 #13231

Merged
merged 7 commits into from Jul 19, 2019
Merged

Update fbeta_score to accept beta=0 #13231

merged 7 commits into from Jul 19, 2019

Conversation

@corona10
Copy link
Contributor

corona10 commented Feb 23, 2019

Reference Issues/PRs

Fixes: #13224

What does this implement/fix? Explain your changes.

Update fbeta_score API to accept beta=0

Any other comments?

I am the first-time contributor for this project.
Please let me know if I make mistakes!
Thanks
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Feb 23, 2019

@qinhanmin2014
Hi, Can you take a look please?

IMHO, Current failure does not relate with this PR
>>> from sklearn.feature_extraction import image 343 >>> # Use the array data from the first image in this dataset: 344

Thanks a lot!
@corona10 corona10 force-pushed the corona10:gh-13224 branch from 8f6fd33 to 97513c7 Feb 24, 2019
@qinhanmin2014
qinhanmin2014 reviewed Feb 24, 2019
View changes
Copy link
Member

qinhanmin2014 left a comment

please update the doc accordingly, e.g., beta : float >= 0
ignore the test failure if it's irrelevant
@qinhanmin2014
qinhanmin2014 reviewed Feb 24, 2019
View changes
sklearn/metrics/classification.py Outdated Show resolved Hide resolved
@corona10 corona10 force-pushed the corona10:gh-13224 branch from 85db8ef to 32224ae Feb 24, 2019
@qinhanmin2014
qinhanmin2014 reviewed Feb 24, 2019
View changes
sklearn/metrics/classification.py Outdated Show resolved Hide resolved
@qinhanmin2014

This comment has been minimized.

Sign in to view
Copy link
Member

qinhanmin2014 commented Feb 24, 2019

Please add an entry to the change log at doc/whats_new/v*.rst. Like the other entries there, please reference this pull request with :issue: and credit yourself (and other contributors if applicable) with :user:.
@qinhanmin2014
qinhanmin2014 reviewed Feb 24, 2019
View changes
doc/whats_new/v0.21.rst Outdated Show resolved Hide resolved
doc/whats_new/v0.21.rst Outdated Show resolved Hide resolved
@corona10 corona10 force-pushed the corona10:gh-13224 branch from 4d58598 to 6c0789c Feb 24, 2019
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Feb 24, 2019

@qinhanmin2014 Thanks for the kind review!
@qinhanmin2014
qinhanmin2014 approved these changes Feb 25, 2019
View changes
Copy link
Member

qinhanmin2014 left a comment

LGTM, thanks @corona10
sklearn/metrics/tests/test_classification.py Outdated
@@ -205,10 +205,12 @@ def test_precision_recall_f_binary_single_class():
assert_equal(1., precision_score([1, 1], [1, 1]))
assert_equal(1., recall_score([1, 1], [1, 1]))
assert_equal(1., f1_score([1, 1], [1, 1]))
assert_equal(1., fbeta_score([1, 1], [1, 1], 0))

This comment has been minimized.

Sign in to view
Copy link
@qinhanmin2014

qinhanmin2014 Feb 25, 2019

Member

you need to update the comment above : Test precision, recall and F1 score -> Test precision, recall and fscore or something you like.

This comment has been minimized.

Sign in to view
Copy link
@qinhanmin2014

qinhanmin2014 Feb 28, 2019

Member

you need to update the comment above : Test precision, recall and F1 score -> Test precision, recall and fscore or something you like

Where's this change? You don't need to squash your commits.
Maybe also add assert_equal(XXX, fbeta_score([1, 1], [1, 1], float('inf'))) here.

This comment has been minimized.

Sign in to view
Copy link
@corona10

corona10 Feb 28, 2019

Author Contributor

@qinhanmin2014 Looks like there's some accident during I worked on another machine.

This comment has been minimized.

Sign in to view
Copy link
@qinhanmin2014

qinhanmin2014 Feb 28, 2019

Member

Maybe also add assert_equal(XXX, fbeta_score([1, 1], [1, 1], float('inf'))) here.

This comment has been minimized.

Sign in to view
Copy link
@corona10

corona10 Feb 28, 2019

Author Contributor

@qinhanmin2014
aa90c28
I've added the test for that case.
Due to https://bugs.python.org/issue28579, I use numpy.isnan method for this test.

This comment has been minimized.

Sign in to view
Copy link
@qinhanmin2014

qinhanmin2014 Feb 28, 2019

Member

assert_true is deprecated, use bare assert instead

This comment has been minimized.

Sign in to view
Copy link
@corona10

corona10 Mar 1, 2019

Author Contributor

@qinhanmin2014 done
@corona10 corona10 force-pushed the corona10:gh-13224 branch from b854efb to ff0d281 Feb 25, 2019
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Feb 25, 2019

@qinhanmin2014
Done! Thanks!
@jnothman
jnothman reviewed Feb 28, 2019
View changes
Copy link
Member

jnothman left a comment

If we're supporting beta=0, we should support beta=inf. Can we please add a test for that?
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Feb 28, 2019

@jnothman Sure!
@corona10 corona10 force-pushed the corona10:gh-13224 branch from ff0d281 to 0491d54 Feb 28, 2019
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Feb 28, 2019

@qinhanmin2014 PTAL
@corona10 corona10 changed the title Update fbeta_score to accept beta=0 [WIP] Update fbeta_score to accept beta=0 Feb 28, 2019
@corona10 corona10 changed the title [WIP] Update fbeta_score to accept beta=0 Update fbeta_score to accept beta=0 Feb 28, 2019
@qinhanmin2014

This comment has been minimized.

Sign in to view
Copy link
Member

qinhanmin2014 commented Feb 28, 2019

So actually we do not support beta=inf currently (See the new test). Do you think we need to add support to that @jnothman ? I'm +0 so you can make the decision here :)
@jnothman

This comment has been minimized.

Sign in to view
Copy link
Member

jnothman commented Feb 28, 2019

I'm +0 for beta=0. But if we support beta=0 we should support beta=inf!
@corona10 corona10 mentioned this pull request Mar 1, 2019
Closed
@corona10 corona10 force-pushed the corona10:gh-13224 branch from 4a898a8 to 1b3be89 Mar 1, 2019
@jnothman

This comment has been minimized.

Sign in to view
Copy link
Member

jnothman commented May 1, 2019

Note that if you keep working on this, the what's new entry should move to v0.22.rst as v0.21 has been closed.
@rth

This comment has been minimized.

Sign in to view
Copy link
Member

rth commented Jul 18, 2019

@corona10 Would you be interested in continuing this PR?

You would need to resolve merge conflicts and handle the case of beta=inf. I can review this one that is done.
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Jul 19, 2019

@rth Thanks! I will take a look
corona10 added 4 commits Feb 23, 2019
@corona10
Update fbeta_score to accept beta=0
39d03bc
@corona10
metrics: update doc and docstring
fab3a66
@corona10
metric: Update doc
5c5387f
@corona10
test: update test_classification
254be90
@corona10
metrics.tests: use assert_equal
Loading status checks…
6b2e92a
@corona10 corona10 force-pushed the corona10:gh-13224 branch from cbdef5d to 6b2e92a Jul 19, 2019
@corona10 corona10 changed the title Update fbeta_score to accept beta=0 [WIP] Update fbeta_score to accept beta=0 Jul 19, 2019
@corona10 corona10 force-pushed the corona10:gh-13224 branch 2 times, most recently from 92f153e to 4663d09 Jul 19, 2019
@corona10
metrics: Update fbeta_score to sccept beta=inf
Loading status checks…
d99b40a
@corona10 corona10 force-pushed the corona10:gh-13224 branch from 4663d09 to d99b40a Jul 19, 2019
@corona10 corona10 changed the title [WIP] Update fbeta_score to accept beta=0 Update fbeta_score to accept beta=0 Jul 19, 2019
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Jul 19, 2019

@rth cc @jnothman
Please take a look
@rth
rth reviewed Jul 19, 2019
View changes
Copy link
Member

rth left a comment

Thanks! A last few comments. Also please update in the beta docstring, that it needs to be positive.

Otherwise LGTM.
sklearn/metrics/classification.py Outdated Show resolved Hide resolved
sklearn/metrics/tests/test_classification.py Show resolved Hide resolved
doc/whats_new/v0.21.rst Outdated Show resolved Hide resolved
@corona10
metrics: update unit test code, docstring and whats_news
Loading status checks…
fd958d3
@corona10

This comment has been minimized.

Sign in to view
Copy link
Contributor Author

corona10 commented Jul 19, 2019

@rth
I've updated the PR, please take a look
@corona10
corona10 reviewed Jul 19, 2019
View changes
sklearn/metrics/classification.py
@@ -1067,7 +1067,7 @@ def fbeta_score(y_true, y_pred, beta, labels=None, pos_label=1,
The `beta` parameter determines the weight of recall in the combined
score. ``beta < 1`` lends more weight to precision, while ``beta > 1``
favors recall (``beta -> 0`` considers only precision, ``beta -> inf``
favors recall (``beta -> 0`` considers only precision, ``beta -> +inf``

This comment has been minimized.

Sign in to view
Copy link
@corona10

corona10 Jul 19, 2019

Author Contributor

@rth Is this the docstring you mean?
if not please let me know what docstring should be updated.
Thanks!

This comment has been minimized.

Sign in to view
Copy link
@rth

rth Jul 19, 2019

Member

I mean the,

beta : float
        Weight of precision in harmonic mean.

but nevermind, this description is enough.
@rth
rth approved these changes Jul 19, 2019
View changes
Copy link
Member

rth left a comment

Thanks @corona10 !

Merging with a +2, despite Joel's +0. I don't think this is too controversial.
@rth rth merged commit b4c1c4e into scikit-learn:master Jul 19, 2019
17 checks passed
17 checks passed
LGTM analysis: C/C++ No code changes detected
Details
LGTM analysis: JavaScript No code changes detected
Details
LGTM analysis: Python No new or fixed alerts
Details
ci/circleci: deploy Your tests passed on CircleCI!
Details
ci/circleci: doc Your tests passed on CircleCI!
Details
ci/circleci: doc-min-dependencies Your tests passed on CircleCI!
Details
ci/circleci: lint Your tests passed on CircleCI!
Details
codecov/patch 100% of diff hit (target 96.87%)
Details
codecov/project 96.87% (+<.01%) compared to 9310d5e
Details
scikit-learn.scikit-learn Build #20190719.29 succeeded
Details
scikit-learn.scikit-learn (Linux py35_conda_openblas) Linux py35_conda_openblas succeeded
Details
scikit-learn.scikit-learn (Linux py35_ubuntu_atlas) Linux py35_ubuntu_atlas succeeded
Details
scikit-learn.scikit-learn (Linux pylatest_conda_mkl_pandas) Linux pylatest_conda_mkl_pandas succeeded
Details
scikit-learn.scikit-learn (Linux32 py35_ubuntu_atlas_32bit) Linux32 py35_ubuntu_atlas_32bit succeeded
Details
scikit-learn.scikit-learn (Windows py35_pip_openblas_32bit) Windows py35_pip_openblas_32bit succeeded
Details
scikit-learn.scikit-learn (Windows py37_conda_mkl) Windows py37_conda_mkl succeeded
Details
scikit-learn.scikit-learn (macOS pylatest_conda_mkl) macOS pylatest_conda_mkl succeeded
Details
@corona10 corona10 deleted the corona10:gh-13224 branch Jul 19, 2019
@amueller amueller mentioned this pull request Aug 6, 2019
Closed
@Wurch Wurch mentioned this pull request Aug 6, 2019
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jnothman jnothman

@qinhanmin2014 qinhanmin2014

@rth rth

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
4 participants
@corona10 @qinhanmin2014 @jnothman @rth
You can’t perform that action at this time.