PCA

import numpy as np, matplotlib.pyplot as plt
from sklearn.datasets import make_blobs
from sklearn.cluster import KMeans
from sklearn.decomposition import PCA
from sklearn.metrics import silhouette_score
X,_=make_blobs(400,n_features=5,centers=4,random_state=3)
for k in [2,3,4,5]: print("k=",k,"silhouette=",round(silhouette_score(X,KMeans(k,n_init=10,random_state=0).fit_predict(X)),3))
lab=KMeans(4,n_init=10,random_state=0).fit_predict(X); Z=PCA(2).fit_transform(X)
plt.scatter(Z[:,0],Z[:,1],c=lab,cmap="viridis",s=15); plt.title("K-Means clusters (PCA view)"); plt.show()