gmm.fit(X) means = gmm.means_ covs = gmm.covars_ n_clusters_ = len(means) ax = liac.plot.gca() for i in xrange(n_clusters_): pi = idx == i plot.scatter(X[pi, 0], X[pi, 1], color=liac.random.make_color(i + 10)) e = liac.plot.Gaussian(means[i], covs[i], 5, color=liac.random.make_color(i), alpha=0.75) ax.add_artist(e) x, y = means[i] liac.plot.plot(x, y, 'x', markersize=14, markeredgewidth=2, color='k') liac.plot.plot(x, y, 'x', markersize=12, markeredgewidth=2, color=liac.random.make_color(i)) # for i, center in enumerate(centers): # X, idx = liac.random.make_gaussian(n_samples=100, centers=center) # plot.scatter(X[:,0], X[:,1], color=liac.random.make_color(i)) plot.show()
import sys import os sys.path.append(os.path.join(os.path.dirname(__file__), '..', '..')) import numpy as np import liac from liac import plot data = liac.dataset.load('iris') classes = data['class'].unique() n_classes = len(classes) fig, axes = plot.subplots(nrows=1, ncols=3) for i, label in enumerate(classes): subdata = data.iloc[data['class']==label, 0:4] ax = axes[i] subdata.plot(ax=ax, marker='o', linestyle='None') ax.set_title(label) plot.show()