Moses-support Digest, Vol 110, Issue 52

Send Moses-support mailing list submissions to
moses-support@mit.edu

To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.mit.edu/mailman/listinfo/moses-support
or, via email, send a message with subject or body 'help' to
moses-support-request@mit.edu

You can reach the person managing the list at
moses-support-owner@mit.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Moses-support digest..."


Today's Topics:

1. Re: tuning question (Philipp Koehn)
2. Re: tuning question (Read, James C)


----------------------------------------------------------------------

Message: 1
Date: Wed, 30 Dec 2015 12:04:02 -0500
From: Philipp Koehn <phi@jhu.edu>
Subject: Re: [Moses-support] tuning question
To: "Read, James C" <jcread@essex.ac.uk>
Cc: Moses Support <moses-support@mit.edu>
Message-ID:
<CAAFADDDFNJ1LVeLpw2=gjSzrX3YiHd7Y+Zj+Z=e4oixAUW8KjQ@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

since the phrase table can be quite huge, it has been standard practice to
filter and binarize the filtered phrase table for tuning and testing.

It is not clear, if this is actually still the best practice, since in the
last years RAM in machines has increased and translation model size
compared to language model size is much smaller.

-phi


On Tue, Dec 29, 2015 at 9:24 AM, Read, James C <jcread@essex.ac.uk> wrote:

> The tutorial here http://www.statmt.org/moses/?n=moses.baseline seems to
> suggest that the tuning phase should be done without first binarising the
> phrase table and relinking the moses.ini file the binarised version. Am I
> reading this wrong or is this just an oversight. I am about to tune a
> number of systems so if binarising the phrase table before hand is likely
> to speed up the tuning process it would be useful to know before I set this
> thing running. Of course, I wouldn't usually complain about having an
> excuse not to do any work for a few days but, well, you know...
>
>
> James
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20151230/349909db/attachment-0001.html

------------------------------

Message: 2
Date: Wed, 30 Dec 2015 17:35:46 +0000
From: "Read, James C" <jcread@essex.ac.uk>
Subject: Re: [Moses-support] tuning question
To: Philipp Koehn <phi@jhu.edu>
Cc: Moses Support <moses-support@mit.edu>
Message-ID:
<AM4PR06MB14744A415D7236695B32904285FD0@AM4PR06MB1474.eurprd06.prod.outlook.com>

Content-Type: text/plain; charset="iso-8859-1"

Thanks for that. Should save me quite some time.


Regarding the TM being much smaller than the LM I wouldn't place bets on that being the situation for much longer (smiles knowingly).


James

________________________________
From: phkoehn@gmail.com <phkoehn@gmail.com> on behalf of Philipp Koehn <phi@jhu.edu>
Sent: Wednesday, December 30, 2015 5:04 PM
To: Read, James C
Cc: Moses Support
Subject: Re: [Moses-support] tuning question

Hi,

since the phrase table can be quite huge, it has been standard practice to filter and binarize the filtered phrase table for tuning and testing.

It is not clear, if this is actually still the best practice, since in the last years RAM in machines has increased and translation model size compared to language model size is much smaller.

-phi


On Tue, Dec 29, 2015 at 9:24 AM, Read, James C <jcread@essex.ac.uk<mailto:jcread@essex.ac.uk>> wrote:

The tutorial here http://www.statmt.org/moses/?n=moses.baseline seems to suggest that the tuning phase should be done without first binarising the phrase table and relinking the moses.ini file the binarised version. Am I reading this wrong or is this just an oversight. I am about to tune a number of systems so if binarising the phrase table before hand is likely to speed up the tuning process it would be useful to know before I set this thing running. Of course, I wouldn't usually complain about having an excuse not to do any work for a few days but, well, you know...


James

_______________________________________________
Moses-support mailing list
Moses-support@mit.edu<mailto:Moses-support@mit.edu>
http://mailman.mit.edu/mailman/listinfo/moses-support


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20151230/6cb6dac4/attachment-0001.html

------------------------------

_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support


End of Moses-support Digest, Vol 110, Issue 52
**********************************************

Tweet