A simple implementation of k-means, trying to make it similar to the sklearn interface. The test data is from: http://cs.joensuu.fi/sipu/datasets/
First you must get the necessary packages:
pip install -r requirement.txt
The solution consists of the Kmeans
class and you can use it like this:
kmeans = KMeans(n_clusters=16).fit(data)
also, you can see the example.py
file as a guide.