Moses-support Digest, Vol 94, Issue 29

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: moses_chart and Generation Models (Philipp Koehn)
2. Re: moses_chart and Generation Models (Hieu Hoang)


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

Message: 1
Date: Sun, 24 Aug 2014 09:17:01 -0400
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] moses_chart and Generation Models
To: Marcin Junczys-Dowmunt <junczys@amu.edu.pl>
Cc: moses-support <moses-support@mit.edu>
Message-ID:
<CAAFADDCESCjz4ZYMio+ohkdH6jSL4nY1GaS55J56-A=KOMVeDA@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

as far as I know the chart decoder can handle multiple factors
(I never tried it, but there has been interest at various points in
time, and I am not sure if they triggered enough motivation to
make it to work fully - the architecture of the code supports them),
but it definitely cannot handle decoding steps

The whole setup of decoding steps could use some major re-thinking
since the current implementation easily blows up when too many
translations rules can be generated - it's a mix of exhaustive
computation and panic pruning.

-phi



On Sat, Aug 23, 2014 at 9:43 AM, Marcin Junczys-Dowmunt <junczys@amu.edu.pl>
wrote:

> Hi list,
> are Generation Models supposed to work with hierarchical models? When
> using moses_chart it seem that the generation model is being ignored.
> Thanks,
> Marcin
> _______________________________________________
> 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/20140824/2f885141/attachment-0001.htm

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

Message: 2
Date: Sun, 24 Aug 2014 16:36:09 +0100
From: Hieu Hoang <Hieu.Hoang@ed.ac.uk>
Subject: Re: [Moses-support] moses_chart and Generation Models
To: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Cc: moses-support <moses-support@mit.edu>
Message-ID:
<CAEKMkbhnU1G994YAGz5zvdsUA3cKzLA=CMDv0ZXUYNpr8kv1Lw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

moses_chart can take input and phrase-tables that have multiple factors
(I've tried already) but there's no support for generation steps in
moses_chart.


On 24 August 2014 14:17, Philipp Koehn <pkoehn@inf.ed.ac.uk> wrote:

> Hi,
>
> as far as I know the chart decoder can handle multiple factors
> (I never tried it, but there has been interest at various points in
> time, and I am not sure if they triggered enough motivation to
> make it to work fully - the architecture of the code supports them),
> but it definitely cannot handle decoding steps
>
> The whole setup of decoding steps could use some major re-thinking
> since the current implementation easily blows up when too many
> translations rules can be generated - it's a mix of exhaustive
> computation and panic pruning.
>
> -phi
>
>
>
> On Sat, Aug 23, 2014 at 9:43 AM, Marcin Junczys-Dowmunt <
> junczys@amu.edu.pl> wrote:
>
>> Hi list,
>> are Generation Models supposed to work with hierarchical models? When
>> using moses_chart it seem that the generation model is being ignored.
>> Thanks,
>> Marcin
>> _______________________________________________
>> Moses-support mailing list
>> Moses-support@mit.edu
>> http://mailman.mit.edu/mailman/listinfo/moses-support
>>
>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
>


--
Hieu Hoang
Research Associate
University of Edinburgh
http://www.hoang.co.uk/hieu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140824/2f455478/attachment-0001.htm

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

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


End of Moses-support Digest, Vol 94, Issue 29
*********************************************

Tweet