[pymvpa] SVR help

[ibinliu at sina.com]

Hi all,I used the Epsilon-insensitive SVR to predict subject-specific score (one group to train and one group to predict), and the predicted value was highly correlated with the actual value. However, the actual value was ranging from 50 to 150 but the predicted value was 75-90. Dose it make sense?Thank you very much in advance.Best regards,Bin Liu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20160327/34041113/attachment.html>