Bug in MeanShift with small number of samples #2356
Description
The code from this SO question, reposted below for reference, sometimes works and sometimes fails. I'm too tired to check if this is really a bug, but if it isn't, the error message could be made friendlier:
import numpy as np
from sklearn.cluster import MeanShift, estimate_bandwidth
from sklearn.datasets.samples_generator import make_blobs
# Generate sample data
centers = [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1]
]
X, _ = make_blobs(n_samples=100, centers=centers, cluster_std=0.6)
# Compute clustering with MeanShift
# The following bandwidth can be automatically detected using
bandwidth = estimate_bandwidth(X, quantile=0.2, n_samples=500)
ms = MeanShift(bandwidth=bandwidth, bin_seeding=True)
ms.fit(X)
labels = ms.labels_
cluster_centers = ms.cluster_centers_
labels_unique = np.unique(labels)
n_clusters_ = len(labels_unique)
print("number of estimated clusters : %d" % n_clusters_)