Regularization for Stanford?

[larsmans]

This is more of a question than an issue, but I was browsing through your code after reading the paper and I can't find any place where regularization is tuned for the Stanford tagger. Where do I find this? Or does Stanford NER do that automatically?

(I couldn't find it out from the javadocs how this is set, so I might have been looking in the wrong place.)

[MvanErp]

Dear Lars,

I'm not an expert on the inner workings of the Stanford NER tool, so
perhaps your questions is best posed on their mailinglist at
https://mailman.stanford.edu/mailman/listinfo/java-nlp-user

As far as I know, some properties such as QNsize do something with L2
regularisation, but I don't really know the specifics.

Good luck!

grtz,
Marieke

On 28 November 2013 14:39, Lars Buitinck notifications@github.com wrote:

This is more of a question than an issue, but I was browsing through your
code after reading the paper and I can't find any place where
regularization is tuned for the Stanford tagger. Where do I find this? Or
does Stanford NER do that automatically?

(I couldn't find it out from the javadocshttp://nlp.stanford.edu/nlp/javadoc/javanlp/edu/stanford/nlp/sequences/SeqClassifierFlags.htmlhow this is set, so I might have been looking in the wrong place.)

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/1
.

Catch a wave and you're sittin' on top of the world
(Beach Boys)