from sklearn.cluster import AffinityPropagation
import numpy as np
X = np.array([[1, 2], [1, 4], [1, 0], [4, 2], [4, 4], [4, 0]])
clustering = AffinityPropagation(affinity = 'euclidean', random_state=5).fit(X)
labels = clustering.labels_ # label to each element
centers = clustering.cluster_centers_ # center of each cluster
# if you need a distance different from euclidean
# calculate your custom, pairwise distance among vectors
# and store them into a matrix M.
# Note: cluster_centers are no longer available
clustering = AffinityPropagation(affinity='precomputed', random_state=5).fit(M)