Search Criteria
Package Details: python-yellowbrick 1.5-1
Package Actions
Git Clone URL: | https://aur.archlinux.org/python-yellowbrick.git (read-only, click to copy) |
---|---|
Package Base: | python-yellowbrick |
Description: | A suite of visual analysis and diagnostic tools for machine learning |
Upstream URL: | https://github.com/DistrictDataLabs/yellowbrick |
Licenses: | Apache |
Submitter: | hexd |
Maintainer: | hexd |
Last Packager: | hexd |
Votes: | 0 |
Popularity: | 0.000000 |
First Submitted: | 2018-08-30 22:03 (UTC) |
Last Updated: | 2023-07-15 20:59 (UTC) |
Dependencies (14)
- python-cycler
- python-matplotlib (python-matplotlib-gitAUR)
- python-numpy (python-numpy-flameAUR, python-numpy-mkl-binAUR, python-numpy-gitAUR, python-numpy1AUR, python-numpy-mklAUR)
- python-scikit-learn (python-scikit-learn-gitAUR)
- python-scipy (python-scipy-mkl-binAUR, python-scipy-gitAUR, python-scipy-mklAUR)
- python-setuptools (make)
- python-pytest (check)
- python-pytest-cov (check)
- python-pytest-flakesAUR (check)
- nltk-data (optional) – Alternative to download the corpora/treebank for NLTK
- python-nltk (optional) – NLTK as the PosTagVisualizer/postag parser
- python-pandas (optional) – Datasets as pandas.DataFrame instead of numpy arrays
- python-spacyAUR (python-spacy-gitAUR) (optional) – SpaCy as the PosTagVisualizer/postag parser
- python-umap-learnAUR (optional) – UMAP Corpus Visualization with UMAPVisualizer/umap
Latest Comments
hexd commented on 2023-07-15 21:03 (UTC)
This one was quite difficult to let tests pass in v1.5-1.
As of now, it seems the library is relying on old APIs of old versions of the requirements (e.g. there are many deprecation warnings from matplotlib), which is okay while it's still compatible with the recent libs, but that's unfortunately not the case everywhere. I wasn't able to make
DispersionPlot
work (I would need to dig further in yellowbrick's code and this would go beyond packaging needs), but for all the rest, it should be working.I applied a small patch to make
yellowbrick.cluster.elbow.distortion_score
work with sparse 2D arrays. It used to work with sparse matrices, but these are no longer accepted by the upstream lib. All other patches are only for testing.