port sys reload(sys) sys.setdefaultencoding('utf-8') import matplotlib.pyplot as plt import numpy as np culster1 = np.random.uniform(0.5, 1.5, (2, 20)) culster2 = np.random.uniform(1.5, 2.5, (2, 20)) culster3 = np.random.uniform(1.5, 3.5, (2, 20)) culster4 = np.random.uniform(3.5, 4.5, (2, 20)) x1 = np.hstack((culster1,culster2)) x2 = np.hstack((culster2,culster3)) x = np.hstack((x1,x2)).T plt.figure() plt.axis([0, 5, 0, 5]) plt.xlabel('x') plt.ylabel('y') plt.grid(True) plt.plot(x[:,0],x[:,1], 'k.', markersize = 12) from sklearn.cluster import KMeans from scipy.spatial.distance import cdist kmeans = KMeans(n_clusters = 2) kmeans.fit(x) plt.plot(kmeans.cluster_centers_[:,0],kmeans.cluster_centers_[:,1],'ro') K = range(1, 10) meandistortions = [] for k in K: kmeans = KMeans(n_clusters=k) kmeans.fit(x) meandistortions.append(sum(np.min(cdist(x, kmeans.cluster_centers_,'euclidean'), axis=1)) / x.shape[0])#選擇每行最小距離求和 plt.figure() plt.grid(True) plt1 = plt.subplot(2,1,1) plt1.plot(x[:,0], x[:,1], 'k.') plt2 = plt.subplot(2,1,2) plt2.plot(K, meandistortions)
