fast non linear clustering on millions of datapoints

kernel-k-means.py

import numpy as np
import matplotlib.pyplot as plt
from sklearn.kernel_approximation import Nystroem
from sklearn.cluster import MiniBatchKMeans

# dot in the middle
X = np.random.randn(100, 2)
# circle around
Y = X  / np.sqrt((X**2).mean(1, keepdims=True)) * 8
Y = Y + np.random.randn(100, 2)

data = np.concatenate([X, Y], 0)
emb = Nystroem(n_components=100, gamma=0.1).fit_transform(data)
out = MiniBatchKMeans(n_clusters=2).fit_predict(emb)
plt.scatter(*data.T, c=out)