import numpy as np
from sklearn import datasets
from sklearn.metrics import pairwise_distances
from sklearn.preprocessing import LabelEncoder
def silhouette_samples(X, labels, metric='euclidean', **kwds):
le = LabelEncoder()
labels = le.fit_transform(labels)
unique_labels = le.classes_
distances = pairwise_distances(X, metric=metric, **kwds)
intra_clust_dists = np.ones(distances.shape[0], dtype=distances.dtype)
inter_clust_dists = np.inf * intra_clust_dists
for curr_label in unique_labels:
mask = curr_label == labels
current_distances = distances[mask]
n_samples_curr_lab = np.sum(mask) - 1
if n_samples_curr_lab != 0:
intra_clust_dists[mask] = np.sum(current_distances[:, mask], axis=1) / n_samples_curr_lab
for other_label in unique_labels:
if other_label != curr_label:
other_mask = other_label == labels
other_distances = np.mean(current_distances[:, other_mask], axis=1)
inter_clust_dists[mask] = np.minimum(inter_clust_dists[mask], other_distances)
sil_samples = inter_clust_dists - intra_clust_dists
sil_samples /= np.maximum(intra_clust_dists, inter_clust_dists)
return sil_samples
dataset = datasets.load_iris()
data = dataset.data
target = dataset.target
result = silhouette_samples(data, target)
print(result)
