Skip to content

FIX make sure to accept "minority" as a valid strategy in over-samplers#964

Merged
glemaitre merged 3 commits into
scikit-learn-contrib:masterfrom
Prakhyath07:master
Dec 28, 2022
Merged

FIX make sure to accept "minority" as a valid strategy in over-samplers#964
glemaitre merged 3 commits into
scikit-learn-contrib:masterfrom
Prakhyath07:master

Conversation

@Prakhyath07
Copy link
Copy Markdown
Contributor

@Prakhyath07 Prakhyath07 commented Dec 23, 2022

Reference Issue

What does this implement/fix? Explain your changes.

while using smapling strategy ="minority" we were getting error. i found issue in base.py of oversampler where in _parameter constraint majority was used in stroptions instead of minority

Any other comments?

@Prakhyath07
update oversampler base.py
c4ffc26 
updated majority to minority  in str options:  
_parameter_constraints: dict = {
        "sampling_strategy": [
            Interval(numbers.Real, 0, 1, closed="right"),
            StrOptions({"auto", "minority", "not minority", "not majority", "all"}),
            Mapping,
            callable,
        ],
        "random_state": ["random_state"],
    }
@Prakhyath07
Merge pull request #1 from Prakhyath07/Prakhyath07-patch-1
4bfefba 
update oversampler base.py
@codecov
Copy link
Copy Markdown

codecov Bot commented Dec 23, 2022 •
edited
Loading

Codecov Report

Base: 96.50% // Head: 94.25% // Decreases project coverage by -2.24% ⚠️

Coverage data is based on head (14c4a8b) compared to base (7cead9c).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files 
@@            Coverage Diff             @@
##           master     #964      +/-   ##
==========================================
- Coverage   96.50%   94.25%   -2.25%     
==========================================
  Files         103      103              
  Lines        7264     7280      +16     
  Branches     1068     1071       +3     
==========================================
- Hits         7010     6862     -148     
- Misses        147      312     +165     
+ Partials      107      106       -1
Impacted Files Coverage Δ
imblearn/over_sampling/base.py 100.00% <ø> (ø)
...rn/over_sampling/tests/test_random_over_sampler.py 100.00% <100.00%> (ø)
...otype_selection/tests/test_random_under_sampler.py 100.00% <100.00%> (ø)
...ing/_prototype_selection/tests/test_tomek_links.py 100.00% <100.00%> (ø)
imblearn/keras/tests/test_generator.py 9.37% <0.00%> (-90.63%) ⬇️
imblearn/tensorflow/_generator.py 27.58% <0.00%> (-68.97%) ⬇️
imblearn/tensorflow/tests/test_generator.py 10.75% <0.00%> (-54.84%) ⬇️
imblearn/keras/_generator.py 45.20% <0.00%> (-46.58%) ⬇️
imblearn/tests/test_docstring_parameters.py 87.32% <0.00%> (-0.71%) ⬇️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@glemaitre glemaitre changed the title updated sampling strategy string of oversampler base.py from majority to minority FIX make sure to accept "minority" as a valid strategy in over-samplers Dec 28, 2022
@glemaitre
Copy link
Copy Markdown
Member

glemaitre commented Dec 28, 2022

I added some non-regression tests and an entry in the changelog.
I will probably try to make a release soon because it is a blocker.

@glemaitre glemaitre merged commit 79107e8 into scikit-learn-contrib:master Dec 28, 2022
@glemaitre
Copy link
Copy Markdown
Member

glemaitre commented Dec 28, 2022

Thanks @Prakhyath07 I will fix the CI builds that are failing. There are not related.

glemaitre added a commit that referenced this pull request Dec 28, 2022
@Prakhyath07 @glemaitre
FIX make sure to accept "minority" as a valid strategy in over-sample…
0cc02fd 
…rs (#964)

Co-authored-by: Guillaume Lemaitre <g.lemaitre58@gmail.com>
@Prakhyath07
Copy link
Copy Markdown
Contributor Author

Prakhyath07 commented Dec 28, 2022

Reference Issue

What does this implement/fix? Explain your changes.

while using smapling strategy ="minority" we were getting error. i found issue in base.py of oversampler where in _parameter constraint majority was used in stroptions instead of minority

Any other comments?

From my side i didn't find any other issue
Thank you so much

@dront78
Copy link
Copy Markdown

dront78 commented Mar 18, 2023

where is 0.10.1?

@glemaitre
Copy link
Copy Markdown
Member

glemaitre commented Mar 20, 2023

On PyPI and conda-forge, e.g. https://pypi.org/project/imbalanced-learn/

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@Prakhyath07 @glemaitre @dront78