Revision history [back]

I am using CVSVM for regression and it worked fine when I tune the parameters manually. But the problem occured when I used it on a large dataset (about 10000 instances and 10 features), it kept giving me the Assertion failed (sv_count != 0) in do_train error, although I got a good result when I tried the same code with a smaller dataset (100 instances and 5 features). Could someone please help me figure out what the problem could be? Thank you.

I am using CVSVM for regression and it worked fine when I tune the parameters manually. But the problem occured when I used it on a large dataset (about 10000 instances and 10 features), it kept giving me the Assertion "Assertion failed (sv_count != 0) in do_train do_train" error, although I got a good result when I tried the same code with a smaller dataset (100 instances and 5 features). Could someone please help me figure out what the problem could be? Thank you.

I am using CVSVM for regression and it worked fine when I tune the parameters manually. But when I use the problem occured grid search method to optimize the parameters: 1.it kept giving me the "Assertion failed (sv_count != 0) in do_train" error when I used it on a large dataset (about 10000 instances and 10 features), it kept giving me the "Assertion failed (sv_count != 0) in do_train" error, although features). 2. I got a good result when I tried the same code with a smaller dataset (100 instances and 5 features). features). Could someone please help me figure out what the problem could be? Thank you.