kobkrit / same-size-k-means Goto Github PK

This is a k-means variation that produces clusters of the same size utilizing the scikit-learn Kmeans methods and associated utilities.

The same-size k-Means logic is the same as found in the Elki Same-size k-Means Variation tutorial.

https://elki-project.github.io/tutorial/same-size_k_means

Please note that this implementation only works in scikit-learn 17.X as later versions having breaking changes to this implementation. Also sparse matrices are not yet supported.

Use just like you would utilize the scikit-learn Kmeans class

from clustering.equal_groups import EqualGroupsKMeans

import numpy as np

X = np.array([[1, 2], [1, 4], [1, 0], [4, 2], [4, 4], [4, 0]])

clf = EqualGroupsKMeans(n_clusters=2)

clf.fit(X)

clf.labels_

The performance of this implementation is very slow. It is relatively quick if the number observations is less than 500.

Optimizations are readily accepted via pull-requests.

• More test coverage
• Add support for sparse matrices
• Package for pypi
• Potentially speed up with cython
• scikit-learn 18.X implementation