lmnn Large Margin Nearest Neighbour 代码阅读

本次代码的作者是：

(C) Laurens van der Maaten, Delft University of Technology

thanks,it is very helpful. I am trying to interprete this code.

matlab toolbox

代码加注释：

function [M, L, Y, C] = lmnn(X, labels)
%LMNN Learns a metric using large-margin nearest neighbor metric learning
%
%   [M, L, Y, C] = lmnn(X, labels)
%
% The function uses large-margin nearest neighbor (LMNN) metric learning to
% learn a metric on the data set specified by the NxD matrix X and the
% corresponding Nx1 vector labels. The metric is returned in M.
%
%

% This file is part of the Matlab Toolbox for Dimensionality Reduction.
% The toolbox can be obtained from http://homepage.tudelft.nl/19j49
% You are free to use, change, or redistribute this code in any way you
% want for non-commercial purposes. However, it is appreciated if you 
% maintain the name of the original author.
%
% (C) Laurens van der Maaten, Delft University of Technology


    % Initialize some variables
    [N, D] = size(X);
    %如果不满足assert的条件就会退出
    assert(length(labels) == N);
    %获取所有种类,lablist里面存储的是所有的label种类
    [lablist, ~, labels] = unique(labels);
    %一共有多少种类
    K