SMMC 解决流形聚类问题 好用 function [cluster_labels,ppca_label,mse,time_mppca,time_smmc,time_sc,W] = smmc(X,nClusts,ppca_dim,ncentres,knn,power) %%%% spectral multi-manifold clustering (smmc) algorithm %%%% Input % X D by N data matrix D*N的矩阵 % nClusts number of clusters 聚类的数目 % ppca_dim dimension of principal component subspace in PPCA PPCA中主元子空间维数 % ncentres number of centres in the mixture model 混合模型中的中心数 % knn number of nearest neighbors 最近邻数 % power power of affinity 权重 %%%% Output % cluster_labels the label of each point using smmc % ppca_label the mixture label of each point in PPCA % mse average L2 error using the mixture model % time_mppca running time of MPPCA % time_smmc running time of smmc % time_sc running time of spectral clustering on the affinity matrix % W affinity matrix used in spectral clustering % Begin copyright notice % % May,2010 % % Written by Yong Wang (yongwang82@gmail.com) % % This code is provided as is, with no guarantees except that % bugs are almost surely present. % % Comments and bug reports are welcome. % % You are free to modify, extend or distribute this code, as long % as this copyright notice is included whole and unchanged. % X =evalin('base','data'); % X=X'; % ppca_dim=2;