Adaptation of the Bartlett statistical test for equal variances, to a procedure of detection of the optimal number of clusters, in the context of spectral clustering.
bartlett(eigvals, thres=0.95, maxk=20)
A vector of eigenvalues, as returned by the eigendecomposition of the normalized Laplacian (see examples), i.e. in decreasing order.
1 - alpha significance threshold, defaults to 95
upper bound to the returned value.
The computed optimal number of clusters.
1 2 3 4 5 6 7 8 9 10 11 12
Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.