Without a start-to-finish log file, it's impossible to debug. Based on
your description, however, I'd say you probably (a) ran out of free hard
drive space for temp files somewhere during step 2, or your very large
corpus has invalid character. Both will cause empty GIZA++ alignment
files and empty phrase tables.
Have you ever create a smaller model to validate your corpus preparation?
On 11/2/2016 10:32 PM, moses-support-requ...@mit.edu wrote:
Date: Wed, 2 Nov 2016 17:32:48 +0200
From: Hasan Sait ARSLAN<hasan.sait.ars...@gmail.com>
Subject: Re: [Moses-support] error in mosesdecoder tuning
To: Barry Haddow<bhad...@staffmail.ed.ac.uk>
Cc: moses-support<moses-support@mit.edu>
Message-ID:
<CAJj7HD3zd_r1ZjbYj0ikNNYauxe1KFA2MV4Ft1PfQ5vKtK=b...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Hi Barry,
I have rerun training starting from step 5, and I kept log file. I have
checked log file, and it doesn't show any bug about creating phrase-table,
but phrase-table is still not being created. I am attaching log file, could
you please check it, and show me where is the problem?
Thanks,?
log.gz
<https://drive.google.com/file/d/0BxvJK3H5ZKsnYzJiZmhjUWI0Qlk/view?usp=drive_web>
?
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
