scikit-learn：聚类算法K-Means

class sklearn.cluster.KMeans (n_clusters=8, 
								init=’k-means++’, 
								n_init=10, 
								max_iter=300, 
								tol=0.0001,
								precompute_distances=’auto’, 
								verbose=0, 
								random_state=None, 
								copy_x=True, 
								n_jobs=None, 
								algorithm=’auto’
							)

from sklearn.metrics import silhouette_score
from sklearn.metrics import silhouette_samples
X
y_pred
silhouette_score(X,y_pred)
silhouette_score(X,cluster_.labels_)
silhouette_samples(X,y_pred)

from sklearn.metrics import calinski_harabaz_score
X
y_pred
calinski_harabaz_score(X, y_pred)

虽然calinski-Harabaz指数没有界，在凸型的数据上的聚类也会表现虚高。但是比起轮廓系数，它有一个巨大的优
点，就是计算非常快速。

X y
plus = KMeans(n_clusters = 10).fit(X)
plus.n_iter_
random = KMeans(n_clusters = 10,init="random",random_state=420).fit(X)
random.n_iter_

random = KMeans(n_clusters = 10,init="random",max_iter=10,random_state=420).fit(X)
y_pred_max10 = random.labels_
silhouette_score(X,y_pred_max10)
random = KMeans(n_clusters = 10,init="random",max_iter=20,random_state=420).fit(X)
y_pred_max20 = random.labels_
silhouette_score(X,y_pred_max20)