Moses-support Digest, Vol 85, Issue 28

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. parser in tree-based model (Yaqin Yang)
2. Re: parser in tree-based model (Philip Williams)


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

Message: 1
Date: Tue, 12 Nov 2013 15:25:32 -0500
From: Yaqin Yang <yaqin276@gmail.com>
Subject: [Moses-support] parser in tree-based model
To: moses-support@mit.edu
Message-ID:
<CADnOx6YL2-9CJVEgGWo0gSawpOxAbAZ-M0=g=O8u6G+AWQ==DQ@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

Dear all,

I try to run the tree-based model on chinese-to-english corpus. It seems
the default parser in the tree-based model is collins parser.

Is it possible to use berkeley parser instead? Does moses have the wrapper
for English berkeley parsing results?

Thanks!
Yaqin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20131112/621b3e0b/attachment-0001.htm

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

Message: 2
Date: Wed, 13 Nov 2013 10:33:49 +0000
From: Philip Williams <philip.williams@mac.com>
Subject: Re: [Moses-support] parser in tree-based model
To: Yaqin Yang <yaqin276@gmail.com>
Cc: moses-support@mit.edu
Message-ID: <13DC0673-DBB4-4C4D-89D0-C55E3958A833@mac.com>
Content-Type: text/plain; charset=iso-8859-1

Hi Yaqin,

although the name suggests otherwise, the parse-de-berkeley.perl wrapper script works fine with the Berkeley parser's English grammar.

I'd recommend passing the -penn option to tokenizer.perl and detokenizer.perl on the English side of your data so that the tokenization more closely matches the Penn Treebank-style tokenization that the Berkeley parser expects.

Phil

On 12 Nov 2013, at 20:25, Yaqin Yang <yaqin276@gmail.com> wrote:

> Dear all,
>
> I try to run the tree-based model on chinese-to-english corpus. It seems the default parser in the tree-based model is collins parser.
>
> Is it possible to use berkeley parser instead? Does moses have the wrapper for English berkeley parsing results?
>
> Thanks!
> Yaqin
> _______________________________________________
> 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


End of Moses-support Digest, Vol 85, Issue 28
*********************************************

Tweet