MATLAB: Help in using svmclassify

support vector machinesvmsvmclassifysvmtrain

Hi,

the output of svm depends on the sign of (w*x+b),

how do i get the output of the exact value of w*x+b before it outputs the sign and then identify the classes?

Best Answer

See this thread

Related Solutions

MATLAB: Compare matrices built with repelem and repmat

The transformation

c = repmat(a, [1,1,A,A]);
d = c(:,1,:,:);

is an indirection to reproduce b.

Z = 2;
W = 3;  
A = 4;  % Smaller to check the output manually
a = rand(Z, W);
b = repelem(a(:,1),A,A);
c = reshape(repmat(a(:,1).', [A, A]), Z*A, A);
isequal(b, c)

But if you really want

c = repmat(a, [1,1,A,A]);
d = c(:,1,:,:);

Then:

f = reshape(permute(d, [3,1,4,2]), [Z*A,A]);
isequal(b, f)    % 1: equal

MATLAB: How to do a ROC analysis using Matlab build-in SVM (Not LibSVM)

Well, there is an answer here http://www.mathworks.com/matlabcentral/answers/64475-does-anybody-have-expertise-with-matlab-svm-classifier, with a reference to another thread. I am collecting all the pieces in one place.

Assume your class labels are -1 and +1, assume that you have trained with 'autoscale' set to true by default, let svm be the struct for the trained SVM model, and let Xnew be the new data for which you need to compute the soft scores.

shift = svm.ScaleData.shift;
scale = svm.ScaleData.scaleFactor;
Xnew = bsxfun(@plus,Xnew,shift);
Xnew = bsxfun(@times,Xnew,scale);
sv = svm.SupportVectors;
alphaHat = svm.Alpha;
bias = svm.Bias;
kfun = svm.KernelFunction;
kfunargs = svm.KernelFunctionArgs;
f = kfun(sv,Xnew,kfunargs{:})'*alphaHat(:) + bias;
f = -f; % flip the sign to get the score for the +1 class

Then call perfcurve(true_labels,f,1).

Best Answer

Related Solutions

MATLAB: Compare matrices built with repelem and repmat

MATLAB: How to do a ROC analysis using Matlab build-in SVM (Not LibSVM)

Related Question