Support for users of CCG software closed 7-27-20

[Lev-Arie Ratinov <arie.ratinov AT gmail.com>]

Hi Iman. Thanks for the interest. It's been out for probably 5 year,
and it's the first complain I hear about the speed. Several
suggestions:
1) Make sure that you're sending short fragments of text. The ting
tries to analyze text with non-local features. So if you send it 6MB
of text at once, it'll blow up your heap space.
2) Make sure your text doesn't contain funny characters. The tokenizer
may fail. Sentence splitter relies on spaces. If your text starts with
a space, sentence splitter may fail.

Play with inputs and see what happens. The system should work well out
of the box.
Peace&Love


On Wed, Apr 3, 2013 at 12:34 AM, Iman Mirrezaei 
<smirre2 AT uic.edu>
 wrote:
> Hi,
> I used  online demo of Illinois NER and its running time is very good. But
> when I integrate Illinois NER in to my java code. it is very slow. Could you
> guide me how to improve its running performance? maybe I should change some
> thing in config file.
> Thanks
> Iman
>
> _______________________________________________
> illinois-ml-nlp-users mailing list
> illinois-ml-nlp-users AT cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/illinois-ml-nlp-users
>