scikit-learn-extra
scikit-learn-extra copied to clipboard
scikit-learn-contrib
scikit-learn contrib estimators
In "Uniform Approximation of Functions with Random Bases" by A. Rahimi and Benjamin Recht [[1]](https://authors.library.caltech.edu/75528/1/04797607.pdf) RBF approximation (`RBFSampler` in `sklearn`) as well as 2 other approx kernels is described. Random...
Fastfood is somewhat slower than RBFSampler (while in theory it should be faster). Google Colab timings (sklearn 0.24.2, sklearn_extra 0.2.0, numpy 1.19.5)  Laptop timings (Ubuntu 20.04, Intel 8300H, 32GB...
Thanks in advance to all reviewers who will spend their time reading this pull request. ## What? I have implemented Voronoi Isolation Forest. It is an Anomaly Detection algorithm based...
This adds TF-IGM (Inverse Gravity Moment), a supervised feature weighting scheme for text classification that measures class distinguishing power for each term. This PR implements the following paper: ["Turning from...
I mentioned it on scikit-learn, and someone suggested mentioning it here on extra library as it seems a more appropriate place to add this new feature. I think many people...
Hi I have used sklearn_extra for clustering my data based on cosine similarity. The data is 100-dimentional vectors. After clustering, I reduce dimensionality to visualize the clustering. I am a...
KMedoids currently pre-computes a full distance matrix with `pairwise_distances` resulting in large memory usage making it unsuitable for datasets with more than 20-50k samples. To improve the situation somewhat, following...
Currently some examples takes quite a while to run, which is problematic as we are running them with sphinx-gallery (documentation CI job) in each commit. In particular, - examples/plot_robust_classification_diabete.py :...
I think we should formalize a bit way commit rights are given for `scikit-learn-extra`. Currently following groups can modify code in this repo, - @scikit-learn-contrib/scikit-learn-extra team (currently containing @glemaitre and...
From @rth in #12: > > > A few more comments @zdog234 , otherwise (after a light review) LGTM. > > We adopted [black](https://github.com/psf/black) for code style recently. Please run...
