Moses-support Digest, Vol 114, Issue 59

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. tuning not working properly in factored model (Carlos Escolano)
2. Re: kenlm multithreading (koormoosh)
3. Re: tuning not working properly in factored model (Ondrej Bojar)


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

Message: 1
Date: Wed, 27 Apr 2016 21:48:50 +0200
From: Carlos Escolano <carlos.e.p93@gmail.com>
Subject: [Moses-support] tuning not working properly in factored model
To: moses-support@mit.edu
Cc: martaruizcostajussa@gmail.com
Message-ID:
<CACvrzNT3fy0+5p=MdUh=e=ByGUotmynuMH34oSD4-x_1iotQow@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi,

I trained a chinese to spanish unfacored model and all worked perfectly.
But when I try to train a factored model for the same task I have some
trouble while tuning. The factors I'm using are only words for chinese and
words, lemmas and POS tags for spanish.

Training seems to finish correctly and the phrase tables shows all the
factors but when tuning t it only does 2 runs and prints a message saying
that weights have not change in the last run. Leaving the original weights.
Also when translating, the BLEU obtained is worse than the obtained with
the not factored model.


These are my calls for training and tuning the model:

$SCRIPTS_ROOTDIR/training/train-model.perl \
-external-bin-dir $GIZA_DIR/mgiza-bin -mgiza \
--corpus $WORKING_DIR/train/train \
--alignment grow-diag-final-and \
--score-options '--GoodTuring' \
--root-dir $WORKING_DIR/baseline/ \
--f zh --e es \
--lm 0:5:$WORKING_DIR/baseline/lm/words.lm.es:0 \
--translation-factors 0-0,1,2 \
--reordering msd-bidirectional-fe \
--reordering-factors 0-0 \

$MOSES_SCRIPTS/training/mert-moses.pl \
$WORKING_DIR/dev/dev.zh \
$WORKING_DIR/dev/dev.es \
$MOSES_DIR/moses-cmd/bin/gcc-4.8.5/release/link-static/threading-multi/moses
\
$WORKING_DIR/baseline/model/moses.ini \
--nbest 100 \
--working-dir $WORKING_DIR/baseline/tuning/ \
--decoder-flags "-drop-unknown -mbr -threads 24 -mp -v 0" \
--rootdir $MOSES_SCRIPTS \
--mertdir $MOSES_DIR/bin/ \
-threads 24 \
--filtercmd '/veu4/usuaris24/xtrans/mosesdecoder/scripts/training/
filter-model-given-input.pl'

/veu4/usuaris24/smt/softlic/mosesdecoder/scripts//ems/support/reuse-weights.perl
\
$WORKING_DIR/baseline/tuning/moses.ini <
$WORKING_DIR/baseline/model/moses.ini >
$WORKING_DIR/baseline/tuning/moses.weight-reused.ini


Best regards,

Carlos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20160427/401e0081/attachment-0001.html

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

Message: 2
Date: Thu, 28 Apr 2016 11:34:18 +1000
From: koormoosh <koormoosh@gmail.com>
Subject: Re: [Moses-support] kenlm multithreading
To: moses-support@mit.edu
Message-ID:
<CAN3_CDijnrBdaMF4aQAsJSPrG79VNRo0=BtC-EkFQoVJHFWPpw@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi, Thanks - Is this happening else where, i.e. in the query step?

On Thu, Apr 28, 2016 at 2:01 AM, <moses-support-request@mit.edu> wrote:

> 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. Call for Participation: IEEE DIPDMWC2016 Moscow, Russia
> (Sandra Evans)
> 2. kenlm multithreading (koormoosh)
> 3. Re: kenlm multithreading (Kenneth Heafield)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 27 Apr 2016 18:03:34 +0800
> From: Sandra Evans <sandra.sdiwc@gmail.com>
> Subject: [Moses-support] Call for Participation: IEEE DIPDMWC2016
> Moscow, Russia
> To: moses-support@mit.edu
> Message-ID:
> <
> CAEjUHKVizroxu4v0xy4R0uWfHM_qfZrFSi_rNcL0dkNuJ4TVmQ@mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> 2016 Third International Conference on Digital Information Processing, Data
> Mining, and Wireless Communications (DIPDMWC)
>
> National Research Nuclear University MEPhI (Moscow Engineering Physics
> Institute), Moscow, Russia
>
> July 06-08, 2016
>
> http://goo.gl/6HTGyi
>
> ===================================================================================================================
>
> The conference welcomes papers on the following (but not limited to)
> research topics:
>
> - Adaptive Signal Processing
> - Parallel Programming & Processing
> - Artificial Intelligence
> - Expert Systems
> - Image Processing
> - Information Security and Cryptography
> - Modulation, Coding, and Channel Analysis
> - Multimedia Signal Processing
> - Bioinformatics & Biomedical Imaging
> - Biomedical Signal Processing
> - Natural Language Processing
> - Neural Networks and Genetic Algorithms
> - Computer-Aided Surgery
> - Data Compression and Watermarking
> - Data Mining Techniques
> - Ethics of Data Mining
> - Risk Management and Analysis
> - Data Classification and Clustering
> - Abnormally and Outlier Detection
> - Feature Extraction and Data Reduction
> - Multi-Task Learning
> - Optimization Techniques
> - Data Cleaning and Processing
> - Text and Web Mining
> - Bluetooth and Personal Area Networks
> - Wireless System Architecture
> - Mobile Management in Wireless Networks
> - Mobile Database Access and Design
> - IP Multimedia Sub-Systems
> - Key Management Protocols
> - Mobile/ Wireless Network Modeling and Simulation
> - Mobile / Wireless Network Planning
> - Wireless Network Standard and Protocols
> - Digital Right Management and Multimedia Protection
> and many more...
>
>
> ===================================================================================================================
> Important Dates
>
> Submission Dates Open from now until June 6, 2016
> Notification of Acceptance 2-4 weeks from the submission date
> Camera Ready Submission June 26, 2016
> Registration Deadline June 26, 2016
> Conference Dates July 6-8, 2016
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> http://mailman.mit.edu/mailman/private/moses-support/attachments/20160427/6bb03766/attachment-0001.html
>
> ------------------------------
>
> Message: 2
> Date: Thu, 28 Apr 2016 00:25:28 +1000
> From: koormoosh <koormoosh@gmail.com>
> Subject: [Moses-support] kenlm multithreading
> To: moses-support@mit.edu
> Message-ID:
> <
> CAN3_CDg_n4vS1_SUOjPNaYSQr06WFC09iX4szbNdf1yHpxKRfg@mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> Hello,
>
> Out of curiosity, does anyone know if multithreading is enabled by default
> during the execution of lmplz of kenlm? I noticed the lmplz cpu usage goes
> up and down around 300% (sometimes). I didn't pass any parameter to enable
> multithreading in lmplz and couldn't find any parameter on the website so I
> assume it's done internally and by default. Does anyone know when it
> decides to do multithreading?
>
> regards,
> -k
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> http://mailman.mit.edu/mailman/private/moses-support/attachments/20160427/334b0cd7/attachment-0001.html
>
> ------------------------------
>
> Message: 3
> Date: Wed, 27 Apr 2016 15:54:43 +0100
> From: Kenneth Heafield <moses@kheafield.com>
> Subject: Re: [Moses-support] kenlm multithreading
> To: moses-support@mit.edu
> Message-ID: <5720D2B3.6010408@kheafield.com>
> Content-Type: text/plain; charset=windows-1252
>
> Yes, it uses threads when it wants to. There is no option to turn
> threads off (and no code path that would do so). One has limited
> control using block size and counts. Ideally it would be more parallel.
>
> Kenneth
>
> On 04/27/2016 03:25 PM, koormoosh wrote:
> > Hello,
> >
> > Out of curiosity, does anyone know if multithreading is enabled by
> > default during the execution of lmplz of kenlm? I noticed the lmplz cpu
> > usage goes up and down around 300% (sometimes). I didn't pass any
> > parameter to enable multithreading in lmplz and couldn't find any
> > parameter on the website so I assume it's done internally and by
> > default. Does anyone know when it decides to do multithreading?
> >
> > regards,
> > -k
> >
> >
> > _______________________________________________
> > 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 114, Issue 58
> **********************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20160427/5bdcb788/attachment-0001.html

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

Message: 3
Date: Thu, 28 Apr 2016 08:46:16 +0200
From: Ondrej Bojar <bojar@ufal.mff.cuni.cz>
Subject: Re: [Moses-support] tuning not working properly in factored
model
To: Carlos Escolano <carlos.e.p93@gmail.com>, moses-support@mit.edu
Cc: martaruizcostajussa@gmail.com
Message-ID: <0bac042f-805a-4436-8726-f81691dc174e@email.android.com>
Content-Type: text/plain; charset=UTF-8

Dear Carlos,

My frequent mistake in this respect is the match of factor representation in run.X.best.out and the reference sentences.

Technically, both is possible: evaluating only the first factor (form) or all factors of each token. BLEU does not care. Mismatch will cause terribly low scores.

O.


On April 27, 2016 9:48:50 PM CEST, Carlos Escolano <carlos.e.p93@gmail.com> wrote:
>Hi,
>
>I trained a chinese to spanish unfacored model and all worked
>perfectly.
>But when I try to train a factored model for the same task I have some
>trouble while tuning. The factors I'm using are only words for chinese
>and
>words, lemmas and POS tags for spanish.
>
>Training seems to finish correctly and the phrase tables shows all the
>factors but when tuning t it only does 2 runs and prints a message
>saying
>that weights have not change in the last run. Leaving the original
>weights.
>Also when translating, the BLEU obtained is worse than the obtained
>with
>the not factored model.
>
>
>These are my calls for training and tuning the model:
>
>$SCRIPTS_ROOTDIR/training/train-model.perl \
> -external-bin-dir $GIZA_DIR/mgiza-bin -mgiza \
> --corpus $WORKING_DIR/train/train \
> --alignment grow-diag-final-and \
> --score-options '--GoodTuring' \
> --root-dir $WORKING_DIR/baseline/ \
> --f zh --e es \
> --lm 0:5:$WORKING_DIR/baseline/lm/words.lm.es:0 \
> --translation-factors 0-0,1,2 \
> --reordering msd-bidirectional-fe \
> --reordering-factors 0-0 \
>
>$MOSES_SCRIPTS/training/mert-moses.pl \
> $WORKING_DIR/dev/dev.zh \
> $WORKING_DIR/dev/dev.es \
>$MOSES_DIR/moses-cmd/bin/gcc-4.8.5/release/link-static/threading-multi/moses
>\
> $WORKING_DIR/baseline/model/moses.ini \
> --nbest 100 \
>--working-dir $WORKING_DIR/baseline/tuning/ \
>--decoder-flags "-drop-unknown -mbr -threads 24 -mp -v 0" \
> --rootdir $MOSES_SCRIPTS \
>--mertdir $MOSES_DIR/bin/ \
>-threads 24 \
> --filtercmd '/veu4/usuaris24/xtrans/mosesdecoder/scripts/training/
>filter-model-given-input.pl'
>
>/veu4/usuaris24/smt/softlic/mosesdecoder/scripts//ems/support/reuse-weights.perl
>\
> $WORKING_DIR/baseline/tuning/moses.ini <
>$WORKING_DIR/baseline/model/moses.ini >
>$WORKING_DIR/baseline/tuning/moses.weight-reused.ini
>
>
>Best regards,
>
>Carlos
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Moses-support mailing list
>Moses-support@mit.edu
>http://mailman.mit.edu/mailman/listinfo/moses-support

--
Ondrej Bojar (mailto:obo@cuni.cz / bojar@ufal.mff.cuni.cz)
http://www.cuni.cz/~obo



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

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


End of Moses-support Digest, Vol 114, Issue 59
**********************************************

Tweet