Thanks,

you said "train the language model on truecased data" which means I
have to train a truecase model first, pipe my training data to
truecase.perl. Correct?


-Andreas

On Wed, Mar 16, 2011 at 12:18 PM, Barry Haddow <[email protected]> wrote:
> Hi Andreas
>
> In general, the data used to train the language model should be cased the same
> way as the target data you give to train-model.perl. So, yes, you should
> train the language model on truecased data, if you're using truecasing. With
> a very large language model the casing may not have much impact,
>
> best regards  - Barry
>
> On Wednesday 16 March 2011 11:12, Andreas Kull wrote:
>> Hi,
>>
>> a simple question about the truecaser: Do I have to train the language
>> model in mixed case like I do with train-model.perl? So I don't use
>> lowercase.perl at all in the training process, correct?
>>
>>
>> Regards,
>> Andreas
>> _______________________________________________
>> Moses-support mailing list
>> [email protected]
>> http://mailman.mit.edu/mailman/listinfo/moses-support
>
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>

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

Reply via email to