Another word aligner you can use is Chris Dyer's fast aligner that comes
with cdec.

http://cdec-decoder.org/index.php?title=Lexical_translation_/_Word_alignment
There is support for it in the EMS
   http://www.statmt.org/moses/?n=FactoredTraining.EMS#ntoc13

On 31 July 2013 10:33, CHATZITHEODOROU Konstantinos <[email protected]>wrote:

> Hi,
>
> Thank you for your prompt reply. I think that you are right but regarding
> their forum. In my projects I use GIZA and MGIZA but I just wanted to test
> other aligners. If I find something I will share it as soon as possible.
> Thanks again for your help.
>
> Cheers,
>
> Konstantinos
>
>
> 2013/7/30 Hieu Hoang <[email protected]>
>
>> they don't answer posts on their forum, eg
>>    http://code.google.com/p/berkeleyaligner/issues/detail?id=4
>> And the code hasn't been updated in years
>>
>> Personally, I use mgiza and am prepared to answer questions on it
>>    https://sourceforge.net/projects/mgizapp/
>>
>> On 30 July 2013 16:34, CHATZITHEODOROU Konstantinos 
>> <[email protected]>wrote:
>>
>>> Hi,
>>>
>>> Thank you for your reply. I didn't add the STDERR because it is written
>>> only 1 error, 0 warnings. Could you please tell me why Berkeley aligner is
>>> not often used by the MT community?
>>> Thank you in advance.
>>>
>>> Regards,
>>>
>>> Kosnatninos
>>>
>>>
>>> 2013/7/30 Hieu Hoang <[email protected]>
>>>
>>>> it's best if you look in the STDERR file (it might be named something
>>>> like TRAINING_run-berkeley.1.STDERR) and seeing how exactly the parser is
>>>> run.
>>>>
>>>> this is not often used so you pretty much have to do your own
>>>> debugging. If you manage to fix it, please let us know.
>>>>
>>>> On 30 July 2013 10:29, CHATZITHEODOROU Konstantinos <
>>>> [email protected]> wrote:
>>>>
>>>>> Hi All,
>>>>>
>>>>> I get the error "ForwardModels and NumIters lengths differ: (2, 0)"
>>>>> running an EMS using Berkeley aligner. Does anybody know how to fix it?
>>>>> Thanks in advance.
>>>>>
>>>>> my log file.
>>>>>
>>>>> main() {
>>>>>   Execution directory: en
>>>>>   ERROR: java.util.InputMismatchException: ForwardModels and NumIters
>>>>> lengths differ: (2, 0):
>>>>> edu.berkeley.nlp.wordAlignment.Main.sameCheck(Main.java:359)
>>>>>
>>>>> edu.berkeley.nlp.wordAlignment.Main.getNumStagesAndCheckRegimen(Main.java:325)
>>>>> edu.berkeley.nlp.wordAlignment.Main.checkOptions(Main.java:266)
>>>>> edu.berkeley.nlp.wordAlignment.Main.run(Main.java:100)
>>>>> fig.exec.Execution.runWithObjArray(Execution.java:466)
>>>>> fig.exec.Execution.run(Execution.java:419)
>>>>> edu.berkeley.nlp.wordAlignment.Main.main(Main.java:95)
>>>>>   Execution directory: en
>>>>> }
>>>>>
>>>>>
>>>>> Cheers,
>>>>> Konstatninos
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Moses-support mailing list
>>>>> [email protected]
>>>>> http://mailman.mit.edu/mailman/listinfo/moses-support
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Hieu Hoang
>>>> Research Associate
>>>> University of Edinburgh
>>>> http://www.hoang.co.uk/hieu
>>>>
>>>>
>>>
>>>
>>> --
>>> CHATZITHEODOROU Konstantinos
>>>
>>>
>>
>>
>> --
>> Hieu Hoang
>> Research Associate
>> University of Edinburgh
>> http://www.hoang.co.uk/hieu
>>
>>
>
>
> --
> CHATZITHEODOROU Konstantinos
>
>


-- 
Hieu Hoang
Research Associate
University of Edinburgh
http://www.hoang.co.uk/hieu
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to