-3

I have thousands of samples for training and test and I want to use SVM with RBF kernel to classify them. The problem is the fact that the Libsvm's implementation of RBF kernel is very slow when using 10k or more data. The main focus of the slow performance is the grid search.

I read about the Liblinear and Lasvm. But liblinear is not what I want because Svms with linear kernel usually have smaller accuracy than RBF kernel accuracy.

I was searching for Lasvm and I can't find useful informations about it. The project site is very poor about information of it. I want to know if Lasvm can use the RBF kernel or it has a specific kind of kernel, if I should scale the test and treining data and if I can do the grid search for my kernel parameters with cross validation.

mad
  • 2,677
  • 8
  • 35
  • 78
  • This question appears to be off-topic because it is about statistics rather than programming. – joran Aug 01 '13 at 02:06
  • 2015 news: you may want to check the relatively new `--ksvm` option to vowpal wabbit. It implements LASVM as a reduction in `vw` (online kernel SVM) a short tutorial is in https://github.com/JohnLangford/vowpal_wabbit/wiki/ksvm.pdf you may use `--kernel rbf`. Don't forget regularization via `--l2` – arielf Dec 13 '15 at 23:59

1 Answers1

3

LaSVM has an RBF kernel implementation too. Based on my experience on large data (>100.000 instances in >1.000 dimensions), it is no faster than LIBSVM though. If you really want to use a nonlinear kernel for huge data, you could try EnsembleSVM.

If your data is truly huge and you are not familiar with ensemble learning, LIBLINEAR is the way to go. If you have a high number of input dimensions, the linear kernel is usually not much worse than RBF while being orders of magnitude faster.

Marc Claesen
  • 16,778
  • 6
  • 27
  • 62
  • Thank you for your answer. My data is not so big as you cite (my data are 16 k for training and 40 k for testing). Is EnsembleSvm faster than Lasvm? Can you help me how to make the parameter selection with Lasvm? thanks again. – mad Jul 24 '13 at 14:02
  • Depending on how you use it, EnsembleSVM *can* be significantly faster than LaSVM. – Marc Claesen Jul 24 '13 at 14:20