edit: nevermind I solved it

I am training an svm on a dataset with 5 classes using libsvm. I have a training set and a test set. I am using the easy.py script. The accuracy is a lot worse than the cv rate (accuracy 52%, cv rate 81%). What can be the reason for that difference?

Unlike this post, I have 200 features and 2000 examples

More info about libsvm https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf easy.py is a script that automatically finds the best set of parameters doing grid search and computes the test accuracy

  • $\begingroup$ What does CV rate measure? Is it CV accuracy? $\endgroup$ – Sycorax Dec 10 '15 at 22:15
  • $\begingroup$ cross validation rate. easy.py does grid search to find the best set of parameters and makes a prediction using these parameters at test time. Yes it is cv accuracy $\endgroup$ – user18121212 Dec 10 '15 at 22:16
  • $\begingroup$ So what's accuracy, then? $\endgroup$ – Sycorax Dec 10 '15 at 22:19
  • 1
    $\begingroup$ Could you edit your question to explain this in more detail? In particular, to remove the reference to easy.py and replace it with a description of what you're doing and how you're measuring these quantities. Anyone can have a file on their computer called easy.py and it's probably not the one you're using! $\endgroup$ – Sycorax Dec 10 '15 at 22:23
  • $\begingroup$ Maybe, but CV is not a community that does debugging. If you're interested in the statistical properties of some algorithm, that's on-topic here. If you think there's a bug or a problem with your code, that's not, but perhaps will be on SO.SE or Programmers.SE. The help centers will describe this in more detail. $\endgroup$ – Sycorax Dec 10 '15 at 22:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.