uima-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Silvestre Losada <silvestre.los...@gmail.com>
Subject Re: Ruta unmark tokens from given possition.
Date Thu, 08 Jan 2015 16:21:28 GMT
Yes I want to filter the number of annotations passed to my actions and
also to another analysis engines that I'm executing from ruta script. I can
add a configuration param to every action and to every analysis engine, but
is to minimize the configuration. I'm adding a configuration parameter to
limit the processed annotations in every action and analysis engine.

Thanks.

On 8 January 2015 at 14:06, Peter Klügl <pkluegl@uni-wuerzburg.de> wrote:

> Hi,
>
> Am 08.01.2015 um 09:07 schrieb Silvestre Losada:
> > Peter thanks for your answer,
> >
> > I tried your Rule and is not working for me, I also tried some variants
>
> Hmm sorry, I haven't tested the rule. I will take a look at it sometime
> maybe. It's probably not so important if your rule does its job.
>
> > without results. What I want to remove RutaBasic annotations to pass less
> > information to an action that I created processing only  annotations that
> > are below a threshold, if I remove at the end then my acction will
> process
> > all information.
>
> I don't know if I understood your task completely. Maybe you want to
> filter the annotations in your actions, not processing RutaBasic?
>
> Best,
>
> Peter
>
>
> > Best,
> >
> > Silvestre.
> >
> >
> > On 7 January 2015 at 17:52, Peter Klügl <pkluegl@uni-wuerzburg.de>
> wrote:
> >
> >> Hi,
> >>
> >> Am 07.01.2015 um 17:36 schrieb Silvestre Losada:
> >>> I'm answering to my self,
> >>>
> >>> (ANY{->UNMARK(ANY)}){CONTEXTCOUNT(Document,20,2000)};
> >> nice rule :-D
> >> but it's probably a bit slow
> >>
> >> My first guess was something with a min/max quantifier since these
> >> counting conditions like CONTEXTCOUNT and POSITION are slow.
> >>
> >> Something like
> >>
> >> Document{-> MARKFIRST(FirstToken)};
> >> FirstToken ANY[19, 19] ANY[1800,1800]{-> UNMARK(ANY)};
> >>
> >>
> >>> Removes all annotations generated by default seeder that are in
> postions
> >> 20
> >>> to 2000. This is working, however it does not work for RutaBasic
> >>> annotations using the same expression
> >>>
> >>> (RutaBasic{->UNMARK(RutaBasic)}){CONTEXTCOUNT(Document,20,2000)};
> >> RutaBasic annotations may not be removed by rules. These annotations
> >> build the complete disjunct partitioning of the document and store
> >> important information. The rule need them for working properly. If you
> >> do not want them in your CAS, you can remove them after applying a ruta
> >> analysis engine. There's a configuration parameter "removeBasics". When
> >> activated, the RutaBasic and the seeding annotations are removed as the
> >> last action of the process() method.
> >>
> >>
> >> Best,
> >>
> >> Peter
> >>
> >>
> >>> I dont know the explanation.
> >>>
> >>> Regards
> >>>
> >>> On 7 January 2015 at 16:46, Silvestre Losada <
> silvestre.losada@gmail.com
> >>>
> >>> wrote:
> >>>
> >>>> HI,
> >>>>
> >>>> I'm creating a ruta script, and I want to remove all tokens that are
> >> after
> >>>> position X, in other words I only want to keep first X tokens. I was
> >>>> playing with ruta conditions, actions and I dont know how to make it.
> do
> >>>> you think is possible to do that?
> >>>>
> >>>> Kind regards.
> >>>>
> >>
> >>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message