I have a problem with the CRF++ Package. CRF++ cannot handle large training parts-of-speech corpora (large tagset and large number of words).

In fact, the number of feature functions automatically generated by CRF++ is very large, which extremely increases the training time. So, what is the best solution for reducing the size of the feature function set?

  • 2
    What are the automatic features like? Are there any configuration parameters for CRF++?
    – Mitch
    Nov 13 '11 at 14:10
  • 1
    This notation (crfpp.sourceforge.net) can help you
    – EMAI
    Nov 13 '11 at 16:46

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.