Moses-support Digest, Vol 105, Issue 3

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-support Digest, Vol 105, Issue 1 (Julian Myerscough)
2. exception in running moses using binary models (Barun Halder)
3. Re: exception in running moses using binary models (Hieu Hoang)
4. low Accuracy (fatma elzahraa Eltaher)


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

Message: 1
Date: Wed, 1 Jul 2015 17:23:33 +0100
From: Julian Myerscough <julian@LfBtranslations.co.uk>
Subject: Re: [Moses-support] Moses-support Digest, Vol 105, Issue 1
To: moses-support@mit.edu
Message-ID: <55941405.5010809@LfBtranslations.co.uk>
Content-Type: text/plain; charset=utf-8; format=flowed

Fatma Elzahraa Eltaher - hello. 4633 words seems very small. Usually for
Moses the number of segments need to be in the 1000's and I have seen on
the moses website a 130,000 sentence corpus referred to as 'small'.
Maybe you can do with a lot fewer sentences if the text is limited
domain and controlled language.

For a corpus of 1000's of sentences you want about 1% or more for
tuning. eg 200,000 sentences of which 2500 for tuning.

Julian

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

Julian Myerscough
Quality Assurance Manager - Languages for Business Ltd


On 01/07/2015 09:54, moses-support-request@mit.edu wrote:
> fatma elzahraa Eltaher


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

Message: 2
Date: Thu, 2 Jul 2015 05:39:20 +0000
From: Barun Halder <barun.halder@citrix.com>
Subject: [Moses-support] exception in running moses using binary
models
To: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<959DC74EF82F84449A2F50FAA4510FAF2249488D@SINPEX01CL01.citrite.net>
Content-Type: text/plain; charset="us-ascii"

Hi,
I ran Moses with normal phrase-table and reordering models.
Now, I did binarise the phrase-table and lexicalised reordering models and also did the necessary changes in moses.ini.
[Note:- In moses.ini , I didn't change PhraseDictionaryMemory to PhraseDictionaryCompact as it was showing "Feature name PhraseDictionaryCompact is not registered". ]

But now when I am trying to run moses it is giving me exception in vector range. Taken the screenshot where I have aligned the expected output with my console output.
Shared link for the .ini file and screenshot- https://citrix.sharefile.com/d-sc96631238a8483a9
What could be the reason?

Thanks in advance,
Barun

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150702/e3624fcb/attachment-0001.htm

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

Message: 3
Date: Thu, 2 Jul 2015 11:34:55 +0400
From: Hieu Hoang <hieuhoang@gmail.com>
Subject: Re: [Moses-support] exception in running moses using binary
models
To: Barun Halder <barun.halder@citrix.com>, "moses-support@mit.edu"
<moses-support@mit.edu>
Message-ID: <5594E99F.1080600@gmail.com>
Content-Type: text/plain; charset="windows-1252"

you should binarize the reordering-table file with
processLexicalTable
DON'T give it the suffix .minlexr. Then delete your existing file
reordering-table.minlexr

Alternatively, recompile moses with --with-cmph=... flag and use the
compact phrase-table and compact reordering model.

If you are new to moses, please work through the tutorial
http://www.statmt.org/moses/?n=Moses.Baseline


On 02/07/2015 09:39, Barun Halder wrote:
>
> Hi,
>
> I ran Moses with normal phrase-table and reordering models.
>
> Now, I did binarise the phrase-table and lexicalised reordering
> modelsand also did the necessary changes in moses.ini.
>
> [Note:- In moses.ini , I didn?t change
> |PhraseDictionaryMemory|to|PhraseDictionaryCompact |as it was showing
> ?Feature name PhraseDictionaryCompact is not registered?. ]
>
> But now when I am trying to run moses it is giving me *exception in
> vector range*. Takenthe screenshot where I have aligned the expected
> output with my console output.
>
> Shared link for the .ini file and screenshot-
> https://citrix.sharefile.com/d-sc96631238a8483a9
>
> What could be the reason?
>
> Thanks in advance,
>
> *Barun*
>
>
>
> _______________________________________________
> Moses-support mailing list
> Moses-support@mit.edu
> http://mailman.mit.edu/mailman/listinfo/moses-support

--
Hieu Hoang
Researcher
New York University, Abu Dhabi
http://www.hoang.co.uk/hieu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150702/46a2b777/attachment-0001.htm

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

Message: 4
Date: Thu, 2 Jul 2015 10:42:01 +0200
From: fatma elzahraa Eltaher <fatmaeltaher@gmail.com>
Subject: [Moses-support] low Accuracy
To: moses-support@mit.edu
Message-ID:
<CAOW1BbTzcY0y86sqmAQ7wiZMjZUgYwPMAGKTcb=aF3MT++iSxQ@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Dear All,

I trained my model with 4633 words ,tested it with 342 word and only 99
word was right . How can I increase the accuracy of my model?

note: I use 1000 word for tuning.


thank you,



Fatma El-Zahraa El -Taher

Teaching Assistant at Computer & System department

Faculty of Engineering, Azhar University

Email : fatmaeltaher@gmail.com
mobile: +201141600434
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150702/f5e2f707/attachment.htm

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

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


End of Moses-support Digest, Vol 105, Issue 3
*********************************************

Tweet