Evaluation on whole corpus

Use the corpus benchmark tool - details in the user guide
Diana

Pawel Mazur wrote:

> Hi,
>
> I wonder how to do evaluations on whole corpus, apart
from 
> concatenating  all documents into one and apart from
writing own code 
> which goes through all documents and adds
"correct", "missing" etc and 
> calculates f-measure by itself?
>
> regards,
> pawel
>