search

analyticalmindsltd / smote_variants

A collection of 85 minority oversampling techniques (SMOTE) for imbalanced learning with multi-class oversampling and model selection features
http://smote-variants.readthedocs.io
MIT License
623 stars 138 forks source link

Update _runtimes.py #71

Closed szghlm closed 1 year ago

szghlm commented 1 year ago

Extended with the runtime of SMOTEWB

gykovacs commented 1 year ago

I know it's painful, but could you please add unit tests to cover these 3 lines?

image

szghlm commented 1 year ago

Yes, it's a bit painful, but at the same time a perfectly reasonable request. Finding special cases also always helps to find possible bugs, so I'll do that as soon as I have time.

gykovacs commented 1 year ago

Great, thank you! tests/oversampling/algorithm_level - this is where I collected all those test cases that had to be added to cover algorithm level special cases. Also, you might now, but for completeness, coverage statistics can be extracted easily by installing the pytest-cov package and issueing the pytest --cov --cov-report term-missing tests command in the root. Also, pylint helped me a lot to improve code quality all over the package. Let me know if I can help in any way!

gykovacs commented 1 year ago

And warm greetings among the prestigious list of package contributors! :)