xml gra tiers

[Brian MacWhinney]

Paul,
    The name %xgra means that it is not a real %gra tier that aligns properly with the %mor line.  
It needed to make that change because I recreated all the %mor tiers in the whole database two weeks ago,
as I noted in an earlier message.  In that message, I also noted that we will update the %gra soon.  
    In the XML, the term "flavor" indicates that a tier is user-defined and non-standard.  Once I run MEGRASP
over the tiers, I will convert them back to %gra.  If there is some urgency in doing this for a particular data set,
I can do this now, but I wanted to finish up a bit of retraining of MEGRASP to improve accuracy.

--Brian MacW

On Feb 26, 2014, at 3:51 PM, Paul Feitzinger <paul.feitzinger at gmail.com> wrote:

> I noticed the most recent XML files available online (at least in
> Valian, Brown and MacWhinney) have their gra tiers isolated in one <a>
> node per utterance
> 
>  <a type="extension" flavor="gra">1|0|ROOT 2|1|PUNCT</a>
> 
> while chatter still produces XML files with per-word <gra> nodes.
> 
> Will the xml files return to the <gra> node format or is this an
> intentional change?
> 

-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
To post to this group, send email to chibolts at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/0477CCE3-31F5-4F36-83FC-0EAC38EA4E90%40cmu.edu.
For more options, visit https://groups.google.com/groups/opt_out.