- <[email protected]>
  i have downloaded all the tools of moses successfully and build the
 language model too. but training of the model with the europarl data takes
 more than 48 hrs on my 8 GB machine. so i tried to train the model with
 partial data, but i got the following error
>
 ERROR: source word 22848 is not in the vocabulary list
> Exit code: 255
> ERROR: Giza did not produce the output file
> working-dir/giza.de-en/de-en.A3.final. Is your corpus clean
> (reasonably-sized sentences)?
>
 even if the data is incomplete, the training should complete so that i knw
> tht moses has been build properly and the model is being trained.output
> should come, even if it is incorrect.
>
> is there any other way to test it out with some smaller set of data so
that
> it can be trained in 2 or 3 hrs on my 8 GB machine??
>
> i need to build english-hindi model so i jst want to test it with eng-
> german data before ,whether it works properly or not.
>>
>> i would be really greatful to u if u pls give me the solution to ths
>> problem.waiting for a positive reply from u.
>>
>> with kind regards
>> akjoshi
>>
>
>
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to