Moses-support Digest, Vol 95, Issue 51

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: Blue output meaning (Philipp Koehn)
2. Re: Blue output meaning (Ihab Ramadan)
3. Binarising Phrase table (Tin Pavelic)


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

Message: 1
Date: Tue, 30 Sep 2014 07:01:33 -0400
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: Re: [Moses-support] Blue output meaning
To: i.ramadan@saudisoft.com
Cc: "moses-support@mit.edu" <moses-support@mit.edu>
Message-ID:
<CAAFADDAMNMCDCnQ5dC9LvAkBUAVuDghPiFxO_aqpatDZgzD9Gg@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

this means that you have a BLEU score of 20.07.

The rest are additional statistics:
- the 1/2/3/4-gram precision
- the brevity penalty resulting from the length ratio of output by reference
- the number of words in the output (hyp) and reference.

You are using a very small test set of maybe only 10 sentences,
so this BLEU score is not all that meaningful as a measure of
the quality of the MT system.

-phi

On Tue, Sep 30, 2014 at 4:33 AM, Ihab Ramadan <i.ramadan@saudisoft.com>
wrote:

> Dears,
>
> After running multi-blue.perl script on my output from Moses I got this
> line
>
> ./multi-blue.perl ?lc reference < hypothesis
>
> BLUE = 20.07, 46.1/23.6/14.8/10.1 (BP=1.000, ratio=1.032 , hyp_len=228,
> ref_len=221)
>
> What this output means?
>
>
>
> Best Regards
>
> *Ihab Ramadan*| Senior Developer| Saudisoft - Egypt | *Tel * +2 02 330
> 320 37 Ext- 0 | Mob+201007570826 | Fax+20233032036 | *Follow us on *[image:
> linked]* | **[image: ZA102637861]** | **[image: ZA102637858]*
>
>
>
> _______________________________________________
> 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/20140930/47c23c25/attachment-0001.htm

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

Message: 2
Date: Tue, 30 Sep 2014 13:47:11 +0200
From: "Ihab Ramadan" <i.ramadan@saudisoft.com>
Subject: Re: [Moses-support] Blue output meaning
To: "'Philipp Koehn'" <pkoehn@inf.ed.ac.uk>
Cc: moses-support@mit.edu
Message-ID: <002901cfdca4$46db05e0$d49111a0$@saudisoft.com>
Content-Type: text/plain; charset="utf-8"

You are right I am already using file with about 10 sentences

Thank you philipp



From: phkoehn@gmail.com [mailto:phkoehn@gmail.com] On Behalf Of Philipp Koehn
Sent: Tuesday, September 30, 2014 1:02 PM
To: i.ramadan@saudisoft.com
Cc: moses-support@mit.edu
Subject: Re: [Moses-support] Blue output meaning



Hi,



this means that you have a BLEU score of 20.07.



The rest are additional statistics:

- the 1/2/3/4-gram precision

- the brevity penalty resulting from the length ratio of output by reference

- the number of words in the output (hyp) and reference.



You are using a very small test set of maybe only 10 sentences,

so this BLEU score is not all that meaningful as a measure of

the quality of the MT system.



-phi



On Tue, Sep 30, 2014 at 4:33 AM, Ihab Ramadan <i.ramadan@saudisoft.com> wrote:

Dears,

After running multi-blue.perl script on my output from Moses I got this line

./multi-blue.perl ?lc reference < hypothesis

BLUE = 20.07, 46.1/23.6/14.8/10.1 (BP=1.000, ratio=1.032 , hyp_len=228, ref_len=221)

What this output means?



Best Regards

Ihab Ramadan| Senior Developer| Saudisoft - Egypt | Tel +2 02 330 320 37 Ext- 0 | Mob+201007570826 | Fax+20233032036 | Follow us on | |




_______________________________________________
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/20140930/fa026385/attachment-0001.htm

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

Message: 3
Date: Mon, 29 Sep 2014 02:16:08 +0200
From: Tin Pavelic <tpavelic@ffzg.hr>
Subject: [Moses-support] Binarising Phrase table
To: moses-support@mit.edu
Message-ID: <5428A4C8.5070405@ffzg.hr>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Sorry to bother,

The first problem happened when i tried to tune the translation. I
followed all the steps, tokenized, and truecasted the data and ran the
mort-moses.pl. The result was phrase table with size of 0. The original
training data is about 1.2 Mil lines, and the tuning data was around
600k line. Is size the problem?

After i waited (quite some time) for the tuning process to finish and i
noticed it failed i went straight on to translation with only the
training part. so using moses.ini file from the "train/model" directory.
The loading as mentioned on the site was quite slow so i decided to
binarize it.

the output error is:

terminate called after throwing an instance of 'util::Exception'
what(): moses/PDTAimp.cpp:169 in void Moses::PDTAimp::Create(const
std::vector<long unsigned int>&, const std::vector<long unsigned int>&,
const string&, const std::vector<float>&) threw util::Exception'.
bin ttable does not exist
Aborted


Please help and thank you all in advance,
--
__________________________

Tin Pavelic
__________________________


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

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


End of Moses-support Digest, Vol 95, Issue 51
*********************************************

Tweet