2013/9/21 Francis Tyers <[email protected]>
> El dv 20 de 09 de 2013 a les 18:41 +0800, en/na Gang Chen va escriure:
> >
> >
> >
> > 2013/9/20 Mikel L. Forcada <[email protected]>
> > El 20/09/13 12:18, Francis Tyers escribió:
> >
> > Did he manage to do the "training with no TSX file"
> > part ?
> > Some of Gang's [CCed] experiments do not use any "forbid"
> > rules. Am I right Gang?
> >
> > Yes, Mikel, the new tagger could work without rules. However, it is
> > done by modifying the code (for the experiments), and there is no such
> > a option for whether or not to use the rules in the TSX file in the
> > command line.
> >
> >
> > I suppose what Fran meant was whether the new tagger supports
> > "retraining"? If so, it does :)
> >
> > The resources that the LSW tagger needs are the same as those for the
> > original HMM tagger. The LSW tagger supports "unsupervised training"
> > and "unsupervised retraining", by using the following commands. So,
> > for "retraining" the TSX file is not used.
> > apertium-tagger -w -t=n DIC CRP TSX TAGGER_DATA
> > apertium-tagger -w -r=n CRP TAGGER_DATA
>
> No, basically I'm asking if it can work without specifying the set of
> coarse tags.
>
> It can't work without the "tagset" section in the TSX file. As Mikel said,
the DTD requires the "tagset" section.
<!ELEMENT tagger
(tagset,forbid?,enforce-rules?,preferences?,discard-on-ambiguity?)>
Fran
>
>
>
------------------------------------------------------------------------------
LIMITED TIME SALE - Full Year of Microsoft Training For Just $49.99!
1,500+ hours of tutorials including VisualStudio 2012, Windows 8, SharePoint
2013, SQL 2012, MVC 4, more. BEST VALUE: New Multi-Library Power Pack includes
Mobile, Cloud, Java, and UX Design. Lowest price ever! Ends 9/22/13.
http://pubads.g.doubleclick.net/gampad/clk?id=64545871&iu=/4140/ostg.clktrk
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
