def gaussian_3(): gaussian3 = load_3gaussian() plot(em(gaussian3, 3, 70), gaussian3)
__author__ = 'jiachiliu' from nulearn.dataset import load_2gaussian, load_3gaussian from nulearn.clustering import KMeans import matplotlib.pyplot as plt def plot(clusters): colors = ['r', 'g', 'b'] for k in clusters: split = clusters[k] plt.scatter(split[:, 0], split[:, 1], marker='o', c=colors[k]) plt.show() if __name__ == '__main__': data = load_3gaussian() cf = KMeans() K = 3 cf.fit(data, K=K, max_iter=20) for k in range(K): print 'n%s = %s' % (k, len(cf.clusters[k])) plot(cf.clusters)