Prev: segmentation of image
Next: Adding constrains to a nonnegative least-squares solution
From: Vladimir Bajan on 3 Jul 2010 06:54
Hello, I've trying to use svmtrain and svmclassify to classify Landsat images (imgaes on 7 spectral bands). When I first try to run svmtrain I get this:

Error in ==> ciorna_SVM at 4
SVM=svmtrain(train,classes_train);

After that I try to convert the paramaters to double and I het this error:

??? Error using ==> svmtrain at 470
Unable to solve the optimization problem:
Maximum number of iterations exceeded; increase options.MaxIter.
To continue solving the problem with the current solution as the
starting point, set x0 = x before calling quadprog.

Can u help me please ? I am kind of new to Matlab so please give detailed instructions

Thank you
 | 
Pages: 1
Prev: segmentation of image
Next: Adding constrains to a nonnegative least-squares solution