mars.learn.cluster.KMeans¶

class
mars.learn.cluster.
KMeans
(n_clusters=8, init='kmeans', n_init=1, max_iter=300, tol=0.0001, verbose=0, random_state=None, copy_x=True, algorithm='auto', oversampling_factor=2, init_iter=5)[source]¶ KMeans clustering.
Read more in the User Guide.
 Parameters
n_clusters (int, default=8) – The number of clusters to form as well as the number of centroids to generate.
init ({'kmeans++', 'kmeans', 'random'} or tensor of shape (n_clusters, n_features), default='kmeans') –
Method for initialization, defaults to ‘kmeans’:
’kmeans++’ : selects initial cluster centers for kmean clustering in a smart way to speed up convergence. See section Notes in k_init for more details.
’kmeans’: scalable kmeans++.
’random’: choose k observations (rows) at random from data for the initial centroids.
If a tensor is passed, it should be of shape (n_clusters, n_features) and gives the initial centers.
n_init (int, default=1) – Number of time the kmeans algorithm will be run with different centroid seeds. The final results will be the best output of n_init consecutive runs in terms of inertia.
max_iter (int, default=300) – Maximum number of iterations of the kmeans algorithm for a single run.
tol (float, default=1e4) – Relative tolerance with regards to inertia to declare convergence.
verbose (int, default=0) – Verbosity mode.
random_state (int, RandomState instance, default=None) – Determines random number generation for centroid initialization. Use an int to make the randomness deterministic. See Glossary.
copy_x (bool, default=True) – When precomputing distances it is more numerically accurate to center the data first. If copy_x is True (default), then the original data is not modified, ensuring X is Ccontiguous. If False, the original data is modified, and put back before the function returns, but small numerical differences may be introduced by subtracting and then adding the data mean, in this case it will also not ensure that data is Ccontiguous which may cause a significant slowdown.
algorithm ({"auto", "full", "elkan"}, default="auto") – Kmeans algorithm to use. The classical EMstyle algorithm is “full”. The “elkan” variation is more efficient by using the triangle inequality, but currently doesn’t support sparse data. “auto” chooses “elkan” for dense data and “full” for sparse data.
oversampling_factor (int, default=2) – Only work for kmeans, used in each iteration in kmeans.
init_iter (int, default=5) – Only work for kmeans, indicates how may iterations required.

cluster_centers_
¶ Coordinates of cluster centers. If the algorithm stops before fully converging (see
tol
andmax_iter
), these will not be consistent withlabels_
. Type
tensor of shape (n_clusters, n_features)

labels_
¶ Labels of each point
 Type
tensor of shape (n_samples,)
See also
MiniBatchKMeans
Alternative online implementation that does incremental updates of the centers positions using minibatches. For large scale learning (say n_samples > 10k) MiniBatchKMeans is probably much faster than the default batch implementation.
Notes
The kmeans problem is solved using either Lloyd’s or Elkan’s algorithm.
The average complexity is given by O(k n T), were n is the number of samples and T is the number of iteration.
The worst case complexity is given by O(n^(k+2/p)) with n = n_samples, p = n_features. (D. Arthur and S. Vassilvitskii, ‘How slow is the kmeans method?’ SoCG2006)
In practice, the kmeans algorithm is very fast (one of the fastest clustering algorithms available), but it falls in local minima. That’s why it can be useful to restart it several times.
If the algorithm stops before fully converging (because of
tol
ormax_iter
),labels_
andcluster_centers_
will not be consistent, i.e. thecluster_centers_
will not be the means of the points in each cluster. Also, the estimator will reassignlabels_
after the last iteration to makelabels_
consistent withpredict
on the training set.Examples
>>> from mars.learn.cluster import KMeans >>> import mars.tensor as mt >>> X = mt.array([[1, 2], [1, 4], [1, 0], ... [10, 2], [10, 4], [10, 0]]) >>> kmeans = KMeans(n_clusters=2, random_state=0, init='kmeans++').fit(X) >>> kmeans.labels_ array([1, 1, 1, 0, 0, 0], dtype=int32) >>> kmeans.predict([[0, 0], [12, 3]]) array([1, 0], dtype=int32) >>> kmeans.cluster_centers_ array([[10., 2.], [ 1., 2.]])

__init__
(n_clusters=8, init='kmeans', n_init=1, max_iter=300, tol=0.0001, verbose=0, random_state=None, copy_x=True, algorithm='auto', oversampling_factor=2, init_iter=5)[source]¶ Initialize self. See help(type(self)) for accurate signature.
Methods
__init__
([n_clusters, init, n_init, …])Initialize self.
fit
(X[, y, sample_weight, session, run_kwargs])Compute kmeans clustering.
fit_predict
(X[, y, sample_weight, session, …])Compute cluster centers and predict cluster index for each sample.
fit_transform
(X[, y, sample_weight, …])Compute clustering and transform X to clusterdistance space.
get_params
([deep])Get parameters for this estimator.
predict
(X[, sample_weight, session, run_kwargs])Predict the closest cluster each sample in X belongs to.
score
(X[, y, sample_weight, session, run_kwargs])Opposite of the value of X on the Kmeans objective.
set_params
(**params)Set the parameters of this estimator.
transform
(X[, session, run_kwargs])Transform X to a clusterdistance space.