In what cases is libsvm supposed to returned [nan] as the predicted values of nonlinear regression (with e-svr using linear kernel)?
Is there a faq available ?
My inputs are not nan, but the std.dev of some of my feature columns are nan. Although when i remove this features nothing seems to change...
From the faq (http://www.csie.ntu.edu.tw/~cjlin/libsvm/faq.html)
Q: Why the code gives NaN (not a number) results?
This rarely happens, but few users reported the problem. It seems that their computers for training libsvm have the VPN client running. The VPN software has some bugs and causes this problem. Please try to close or disconnect the VPN client.