contractions in MOR

Leonid Spektor spektor at andrew.cmu.edu
Thu Jun 5 16:57:23 UTC 2014


Ying,

	The errors were due to bad grammar. We are experimenting with a different way of disambiguating results of MOR command. For test purposes POST was modified to work in two different modes. The new test mode uses much shorter morphemic forms than normal mode. But, for POST to work correctly both POST training and POST analyses must be done using the same mode. The test mode is triggered by presence of "connl.cut" file in root grammar folder. Even if you do not get any error messages after change all special single quote characters to simple single quote characters, the results of POST command might not be correct, because POST analyses will be using new test mode and the POST database file "post.db" was created using old mode. The test mode is not ready for public use yet, so the best thing to do is to get new eng grammar from childes web site or at the very least to rename the "connl.cut" file to some other name to disable POST's test mode.

Leonid.

On Jun 5, 2014, at 10:38, Ying Lu <ying.lu at utexas.edu> wrote:

> Dear Leonid,
> 
> Thank you very much for pointing out the special single quotes characters! In fact, the errors were due to the use of characters instead of the bad version of eng grammar:-) I will be more careful about typing the characters in the future!
> 
> Best!
> 
> Ying 
> 
> 
> On Wed, Jun 4, 2014 at 3:36 PM, Leonid Spektor <spektor at andrew.cmu.edu> wrote:
> Ying,
> 
> 	I also noticed that in your files you use special single quotes characters. For example in word "couldn't". Because of that MOR doesn't know how to interpret this word and you get result "?|couldn eq|eq2 ?|t". The single quote character should be " ' ". The special single quotes character has Unicode number 2019 and the normal one that MOR expects has Unicode number 27. Essentially you need to replace ' character with ' character in your data files.
> 
> Leonid.
> 
> On Jun 4, 2014, at 14:24, Leonid Spektor <spektor at andrew.cmu.edu> wrote:
> 
>> Ying,
>> 
>> 	You are using bad version of eng grammar. I would recommend to you to get the latest version of "eng" grammar and CLAN from childes server. If you really want to continue using your version of grammar, then please locate file called "connl.cut" in "eng" grammar folder and rename it to anything else, like "connl-hold.cut" for example.
>> 
>> Leonid.
>> 
>> On Jun 4, 2014, at 10:07, Ying <yl5834 at gmail.com> wrote:
>> 
>>> 
>>> Dear Chibolts,
>>> 
>>> I was trying to run MOR and POST on a CHAT file (attached). I used the codes below.
>>> 
>>> mor +t*CHI MEV001_E_retell.cha +1
>>> post +t*CHI MEV001_E_retell.cha +1
>>> 
>>> There was a problem with contractions such as couldn't. I got error messages like this:
>>> From file <MEV001_E_retell.cha>
>>> *** ERROR 1: In file "MEV001_E_retell.cha"
>>>   in item:    ?|couldn
>>>   Can't find conversion for: ?|couldn
>>> *** ERROR 1: In file "MEV001_E_retell.cha"
>>>   in item:    ?|t
>>>   Can't find conversion for: ?|t
>>> 
>>> What shall I do to avoid such error?
>>> 
>>> Thanks!
>>> Ying
>>> 
>>> -- 
>>> 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/c34eb3fe-8757-4961-a9f9-6056dc808382%40googlegroups.com.
>>> For more options, visit https://groups.google.com/d/optout.
>>> <MEV001_E_retell.cha>
>> 
> 
> 
> -- 
> 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/3350BBCB-0ADB-4333-A168-1E91CC738544%40andrew.cmu.edu.
> 
> For more options, visit https://groups.google.com/d/optout.
> 
> 
> -- 
> 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/CACApr0EsUDPd3ErYRbpsNjWuXUEqYwSJ7WWzhLdyHHrHLzdBqg%40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.

-- 
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/BFD06260-148C-4DDF-833B-D900A8932B04%40andrew.cmu.edu.
For more options, visit https://groups.google.com/d/optout.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20140605/ab352469/attachment.html>


More information about the Chibolts mailing list