Hi,

thanks for pointing out the error - I fixed the web page.

-phi

On Fri, Mar 19, 2010 at 10:00 AM, Sara Stymne <[email protected]> wrote:
> Hi Maria!
>
> The model msd-bidirectional-e will not work, since you can only
> condition lexical reordering models on either the foreign phrase (f) or
> both the foreign and the target (fe). Only on the target (e) is not
> supported.
>
> There is actually an error on the Moses webpage
> (http://www.statmt.org/moses/?n=FactoredTraining.BuildReorderingModel),
> where it says that this e-model is acceptable.
>
> If you want to train unidirectional models, you should not use
> -unidirectional in your command for the training script, it is chosen by
> default. So change your second model to: msd-fe, at training time.
>
> At decoding time: if you're using a monotonicity model, make sure that
> the number of weights specified in the moses.ini file, actually are the
> same as in the reordering table.
>
> /Sara
>
>
> 2010-03-18 21:57, maria sol ferrer skrev:
>> Hey, thanks for your help.
>>
>> Does anyone else know what might be wrong with changing the reordering
>> model? has anyone done it successfully? I really cant get it working and
>> there's no help on the web.
>>
>> Thanks a lot.
>>
>> 2010/3/19 Hieu Hoang <[email protected] <mailto:[email protected]>>
>>
>>     i can't see any obvious problems.
>>
>>     i don't know much about the lexicalized reordering. It seems like
>>     the number of scores is not what is expected by the decoder for the
>>     particular setting specified in the ini file. I think in training,
>>     you have to specify what type of reordering model you want, and
>>     specify the SAME type when decoding.
>>
>>     sorry i can't be of more help
>>
>>
>>     On 17 March 2010 21:35, maria sol ferrer <[email protected]
>>     <mailto:[email protected]>> wrote:
>>
>>         Hey, thanks for your answer, here is the ini file and the output
>>         from when i try to translate...
>>
>>         2010/3/18 Hieu Hoang <[email protected]
>>         <mailto:[email protected]>>
>>
>>             can you send me the ini file, i'll try & see if there's
>>             anything obviously wrong
>>
>>
>>             On 17/03/2010 05:09, maria sol ferrer wrote:
>>>             Hi, I was running some tests on different reordering types
>>>             and when I tried with
>>>
>>>             msd-bidirectional-e, msd-unidirectional-fe and
>>>             monotonicity-bidirectional-fe
>>>
>>>             the first 2 gave the error that reordering doesnt exist
>>>             during training, the third trained fine but when trying to
>>>             translate, it didnt, the only message there was was:
>>>             Reordering types NOT individualy weighted! (By the way,
>>>             the default reordering works fine)
>>>
>>>             I searched the web and all I found was another question on
>>>             the mailing list like this, about the monotonicity
>>>             reordering (all of them) not working, but it was never
>>>             answered.
>>>
>>>             I'm running an english-spanish decoder and the reordering
>>>             is not so good, it just pretty much translates in the
>>>             english order, which is not always correct in spanish.
>>>
>>>             I'd apreciate your help, thanks.
>>>
>>>
>>>             _______________________________________________
>>>             Moses-support mailing list
>>>             [email protected]  <mailto:[email protected]>
>>>             http://mailman.mit.edu/mailman/listinfo/moses-support
>>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Moses-support mailing list
>> [email protected]
>> http://mailman.mit.edu/mailman/listinfo/moses-support
>
> _______________________________________________
> Moses-support mailing list
> [email protected]
> http://mailman.mit.edu/mailman/listinfo/moses-support
>

_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to