[pet] [Matrix-dev] AW: PET error: No morphology registered

[Michael Wayne Goodman]

Yes, it is a rather unsatisfying solution, but thanks for the
workaround, Ravi and Bart.

On Fri, Nov 19, 2010 at 12:53 AM, Bart Cramer <bart.cramer at gmail.com> wrote:
> I was trying this 'minimal grammar' as well a few years back, and I
> recall (but correct me if I'm wrong) that I had the same error message
> when no there were no morphologcal rules at all. The solution was to
> make one rule that is extremely unlikely to match (I had %prefix xyx
> xyx).
>
> HTH,
>
> Bart.
>
> On Fri, Nov 19, 2010 at 2:35 AM, Michael Wayne Goodman
> <goodmami at u.washington.edu> wrote:
>> Hi, I am also running into this problem. Did you figure it out, Ravi,
>> or do any of the Pet people know, how to resolve the "No morphology
>> registered" error message?
>>
>> Thanks,
>>
>> On Mon, Aug 2, 2010 at 8:15 AM, Emily M. Bender
>> <ebender at u.washington.edu> wrote:
>>> My guess is that there's some parameter that should be
>>> in the .set file to indicate something about morphological processing,
>>> and that our .set files (as generated by the customization system)
>>> don't specify this parameter.  But, I don't know off hand what
>>> the parameter would be.
>>>
>>> Emily
>>>
>>>
>>> 2010/8/2 Coote, Ravi <ravi.coote at fkie.fraunhofer.de>:
>>>> Dear Michael,
>>>>
>>>> sorry for the delay of answering.
>>>>
>>>> For simplicity, i created for now the sample grammar "minimal grammar" with the matrix customization webpage (see choices file attached).
>>>>
>>>> I flopped the grammar which works fine now.
>>>>
>>>> When analyzing the artificial sample sentence "n1 iv" cheap results in "No morphology registered" (see below) :
>>>>> cheap \(minimal\ grammar\)-pet.grm
>>>> reading `pet/(minimal grammar)-pet.set'... including `pet/global.set'...
>>>> loading `(minimal grammar)-pet.grm' ((minimal grammar) (2010-08-02_14:44:09_UTC)) [ disabling unification quickcheck ] [ disabling subsumption quickcheck ]
>>>> 1333 types in 0.04 s
>>>> n1 iv
>>>> No morphology registered
>>>>
>>>> ##########################
>>>>
>>>> Any ideas what the problem could be ?
>>>>
>>>> I am using cheap version 0.99.14svn_cm!
>>>>
>>>> Thank you a lot!
>>>>
>>>> Kind Regards,
>>>>
>>>> Ravi Coote
>>>>
>>>>
>>>> -----Ursprüngliche Nachricht-----
>>>> Von: goodman.m.w at gmail.com [mailto:goodman.m.w at gmail.com] Im Auftrag von Michael Wayne Goodman
>>>> Gesendet: Dienstag, 20. Juli 2010 17:15
>>>> An: Coote, Ravi
>>>> Cc: pet at delph-in.net; matrix-dev at u.washington.edu
>>>> Betreff: Re: [pet] PET error: No morphology registered
>>>>
>>>> Hi,
>>>>
>>>> I recently updated the morphology components of the grammar matrix, so it is likely my changes have introduced some bugs. I tried creating a grammar as you described, then flopping it and parsing it with cheap, but I could not reproduce the error. Could you post your choices file (assuming you used the customization system) so I could take a look?
>>>>
>>>> I've also CC'd the matrix-dev mailinglist.
>>>>
>>>> 2010/7/20 Coote, Ravi <ravi.coote at fkie.fraunhofer.de>:
>>>>> Dear list,
>>>>>
>>>>> I created a simple grammar-file based on the lingo grammar matrix. The grammar does only contain full-form words. There are no morphems configured. Running cheap results in the error message 'No morphology registered':
>>>>>
>>>>>> cheap arabic-pet.grm
>>>>> reading `pet/arabic-pet.set'... including `pet/global.set'...
>>>>> loading `arabic-pet.grm' (arabic (2010-07-20_14:11:46_UTC)) [
>>>>> disabling unification quickcheck ] [ disabling subsumption quickcheck
>>>>> ]
>>>>> 1334 types in 0.03 s
>>>>> عاد سفير
>>>>> No morphology registered
>>>>> سفير عاد
>>>>> No morphology registered
>>>>>
>>>>> Does anybody know if morphology can be turned off in PET? Options -no-online-morph and -no-fullform-morph don't help out, unfortuneately.
>>>>>
>>>>> Thank you for any hints!
>>>>>
>>>>> Regards,
>>>>>
>>>>> Ravi Coote
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> -Michael Wayne Goodman
>>>>
>>>> _______________________________________________
>>>> Matrix-dev mailing list
>>>> Matrix-dev at u.washington.edu
>>>> http://mailman2.u.washington.edu/mailman/listinfo/matrix-dev
>>>>
>>>>
>>> _______________________________________________
>>> Matrix-dev mailing list
>>> Matrix-dev at u.washington.edu
>>> http://mailman2.u.washington.edu/mailman/listinfo/matrix-dev
>>>
>>
>>
>>
>> --
>> -Michael Wayne Goodman
>>
>>
>



-- 
-Michael Wayne Goodman