ls_vccore.m

最新的模式识别分类工具箱,希望对朋友们有用!

% Learns classifier and classifies test set
%  using the least-squares algorithm
% Inputs:
% Usage
%      [trainError, testError, estTrainLabels, estTestLabels] = ...
%           LVQ1_VC(trainFeatures, trainLabels,Nmu ,testFeatures, testLabels)
% where
%
% Inputs:
% 	trainFeatures	- the training set vectors, one vector per column
%	trainLabels    - the labels of the above
%       Nmu             - Number of centroids
%       testFeatures   - test set, one column per vector
%       testLabels     - labels  for test set%
% Outputs
%	trainError     - the error rate on the training set (one entry per
%	                   class + total error)
%	testError      - the error rate on the test set (one entry per class
%	                  + total error)
%       estTrainLabels - the labels produced by the algorithm for the
%                          training samples
%       estTestLabels - the labels produced by the algorithm for the
%                          test samples

function [trainError, testError,estTrainLabels, estTestLabels] = ...
    LS_VC(trainFeatures, trainLabels, weights, testFeatures, testLabels)

[Nclasses, classes]  = find_classes([trainLabels(:);testLabels(:)]); 

if (Nclasses > 2)
  fprintf('The current version of the classifier deals with 2 classes only');
  trainError = ones(size(classes));
  testError  = ones(size(classes));
  return;
end

[Dim, Nf]       = size(trainFeatures);
Dim             = Dim + 1;
trainFeatures(Dim,:) = ones(1,Nf);
[Dim1,Nf1]      = size(testFeatures);
testFeatures(Dim1+1,:) = ones(1,Nf1);

%Weighted LS or not?
switch length(weights),
case Nf + 1,
    %Ada boost form
    weights = weights(1:Nf);
case Nf,
    %Do nothing
otherwise
    weights = ones(1, Nf);
end

maxLabel = max(trainLabels);
minLabel = min(trainLabels);


if ( maxLabel ~= 1 | minLabel ~= 0)
  
else
  train_one  = find(trainLabels == 1);
  train_zero = find(trainLabels == 0);
  
  %Preprocess the labels

  mod_trainLabels = 2*trainLabels - 1; 

  w = inv((trainFeatures .* (ones(Dim,1)*weights)) * trainFeatures') * ...
      (trainFeatures .* (ones(Dim,1)*weights)) * mod_trainLabels'; 
  
  estTrainLabels = w'*trainFeatures;
  estTrainLabels = estTrainLabels > 0;
  trainError = computeError(classes, trainLabels, estTrainLabels);

  estTestLabels = w'*testFeatures;
  estTestLabels = estTestLabels > 0;
  testError = computeError(classes, testLabels, estTestLabels);
end