mlxtend icon indicating copy to clipboard operation
mlxtend copied to clipboard

Published 20 hours ago • verified publisher icon rasbt

Update sparse array handling to run unit tests on newer pandas & scipy versions

Open rasbt opened this issue 3 years ago • 1 comments

Mlxtend currently requires the following versions for the unit tests (see workflow):

  • scikit-learn=1.0.2
  • pandas=1.3.5
  • numpy=1.21.4
  • scipy=1.7.3

I noticed unit tests fail with more recent versions. Presumably due to an internal change regarding sparse arrays/data frames in pandas and SciPy. I currently don't have time to fix it, hence the Issue as a reminder (and help and PRs would be super welcome :))

Steps/Code to Reproduce

Install latest packages via

conda install pandas scipy numpy scikit-learn tensorflow pytest -y

Then run unit tests via

PYTHONPATH="." pytest mlxtend/

rasbt avatar Feb 17 '22 03:02 rasbt

Hey @rasbt I have fixed the load_boston dereciation in test_exhaustive_feature_selector.py can you please approve my pull request and I am working on other 3 errors. So, Could you please assign it to me.

namanmistry avatar May 16 '23 06:05 namanmistry