Hi,
I'm training a translation model on Ubuntu with mgiza, nplm but somehow it
failed at the lexical reordering. Could you help me out what's going on
here? Thanks!
The system description:
-------
Distributor ID: Ubuntu
Description: Ubuntu 14.04.4 LTS
Release: 14.04
Codename: trusty
-------
Here is the command I used:
-------
"nohup nice /home/ubuntu/m/mosesdecoder/scripts/training/train-model.perl
--root-dir ./training --parallel --model-dir ./training/model --mgiza
--mgiza-cpus 8 --corpus /home/ubuntu/m/data/train-clean --external-bin-dir
/home/ubuntu/m/mosesdecoder/tools --f ja --e ko --alignment
grow-diag-final-and --reordering msd-bidirectional-fe --score-options
"--GoodTuring" --lm 0:5:/home/ubuntu/m/nn/kolm.nnlm:8 --cores 16
--sort-buffer-size 10G >& training.out &"
-------
And, the error log snippet from the "training.out" file generated:
-------
H333444 Training Finished at: Tue Aug 9 07:49:18 2016
Entire Viterbi H333444 Training took: 1814 seconds
==========================================================
Entire Training took: 3950 seconds
Program Finished at: Tue Aug 9 07:49:18 2016
==========================================================
Merging A3.final.part* tables
Executing: /home/ubuntu/m/mosesdecoder/tools/merge_alignment.py
/home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final.part*>/home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final
Combined 8 files, totally 780579 sents
Executing: rm -f /home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final.gz
Executing: gzip /home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final
Waiting for second GIZA process...
(3) generate word alignment @ Tue Aug 9 07:49:46 UTC 2016
Combining forward and inverted alignment from files:
/home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final.{bz2,gz}
/home/ubuntu/m/training/giza.ko-ja/ko-ja.A3.final.{bz2,gz}
Executing: mkdir -p /home/ubuntu/m/training/model
Executing: /home/ubuntu/m/mosesdecoder/scripts/training/giza2bal.pl -d
"gzip -cd /home/ubuntu/m/training/giza.ko-ja/ko-ja.A3.final.gz" -i "gzip
-cd /home/ubuntu/m/training/giza.ja-ko/ja-ko.A3.final.gz"
|/home/ubuntu/m/mosesdecoder/scripts/../bin/symal -alignment="grow"
-diagonal="yes" -final="yes" -both="yes" >
/home/ubuntu/m/training/model/aligned.grow-diag-final-and
symal: computing grow alignment: diagonal (1) final (1)both-uncovered (1)
skip=<0> counts=<780579>
(4) generate lexical translation table 0-0 @ Tue Aug 9 07:51:13 UTC 2016
(/home/ubuntu/m/data/train-clean.ja,/home/ubuntu/m/data/train-clean.ko,/home/ubuntu/m/training/model/lex)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Saved: /home/ubuntu/m/training/model/lex.f2e and
/home/ubuntu/m/training/model/lex.e2f
FILE: /home/ubuntu/m/data/train-clean.ko
FILE: /home/ubuntu/m/data/train-clean.ja
FILE: /home/ubuntu/m/training/model/aligned.grow-diag-final-and
(5) extract phrases @ Tue Aug 9 07:52:16 UTC 2016
/home/ubuntu/m/mosesdecoder/scripts/generic/extract-parallel.perl 16 split
"sort -S 10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/extract
/home/ubuntu/m/data/train-clean.ko /home/ubuntu/m/data/train-clean.ja
/home/ubuntu/m/training/model/aligned.grow-diag-final-and
/home/ubuntu/m/training/model/extract 7 orientation --model wbe-msd
--GZOutput
Executing:
/home/ubuntu/m/mosesdecoder/scripts/generic/extract-parallel.perl 16 split
"sort -S 10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/extract
/home/ubuntu/m/data/train-clean.ko /home/ubuntu/m/data/train-clean.ja
/home/ubuntu/m/training/model/aligned.grow-diag-final-and
/home/ubuntu/m/training/model/extract 7 orientation --model wbe-msd
--GZOutput
MAX 7 1 0
Started Tue Aug 9 07:52:16 2016
using gzip
isBSDSplit=0
Executing: mkdir -p /home/ubuntu/m/training/model/tmp.9146; ls -l
/home/ubuntu/m/training/model/tmp.9146
total=780579 line-per-split=48787
split -d -l 48787 -a 7 /home/ubuntu/m/data/train-clean.ko
/home/ubuntu/m/training/model/tmp.9146/target.split -d -l 48787 -a 7
/home/ubuntu/m/data/train-clean.ja
/home/ubuntu/m/training/model/tmp.9146/source.split -d -l 48787 -a 7
/home/ubuntu/m/training/model/aligned.grow-diag-final-and
/home/ubuntu/m/training/model/tmp.9146/align.merging extract / extract.inv
gunzip -c /home/ubuntu/m/training/model/tmp.9146/extract.0000000.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000001.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000002.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000003.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000004.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000005.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000006.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000007.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000008.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000009.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000010.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000011.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000012.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000013.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000014.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000015.gz | LC_ALL=C sort
-S 10G -T /home/ubuntu/m/training/model/tmp.9146 2>> /dev/stderr | gzip
-c > /home/ubuntu/m/training/model/extract.sorted.gz 2>> /dev/stderr
gunzip -c /home/ubuntu/m/training/model/tmp.9146/extract.0000000.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000001.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000002.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000003.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000004.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000005.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000006.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000007.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000008.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000009.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000010.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000011.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000012.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000013.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000014.inv.gz
/home/ubuntu/m/training/model/tmp.9146/extract.0000015.inv.gz | LC_ALL=C
sort -S 10G -T /home/ubuntu/m/training/model/tmp.9146 2>> /dev/stderr |
gzip -c > /home/ubuntu/m/training/model/extract.inv.sorted.gz 2>>
/dev/stderr
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000000.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000001.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000002.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000003.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000004.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000005.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000006.gz: No such
file or directory
gzip: gzip: ubuntu/m/training/model/tmp.9146/extract.0000007.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000008.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000009.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000010.gz: No such
file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000011.gz: No such
file or directory
gzip: gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000012.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000003.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000014.gz: No such
file or directory
gzip: /home//home/ubuntu/m/training/model/tmp.9146/extract.0000015.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000005.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000006.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000007.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000008.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000009.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000010.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000011.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000012.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000013.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000014.inv.gz: No
such file or directory
gzip: /home/ubuntu/m/training/model/tmp.9146/extract.0000015.inv.gz: No
such file or directory
Finished Tue Aug 9 07:52:18 2016
(6) score phrases @ Tue Aug 9 07:52:18 UTC 2016
(6.1) creating table half
/home/ubuntu/m/training/model/phrase-table.half.f2e @ Tue Aug 9 07:52:18
UTC 2016
/home/ubuntu/m/mosesdecoder/scripts/generic/score-parallel.perl 16 "sort -S
10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/extract.sorted.gz
/home/ubuntu/m/training/model/lex.f2e
/home/ubuntu/m/training/model/phrase-table.half.f2e.gz --GoodTuring 0
(6.1) creating table half
/home/ubuntu/m/training/model/phrase-table.half.e2f @ Tue Aug 9 07:52:18
UTC 2016
Executing: /home/ubuntu/m/mosesdecoder/scripts/generic/score-parallel.perl
16 "sort -S 10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/extract.sorted.gz
/home/ubuntu/m/training/model/lex.f2e
/home/ubuntu/m/training/model/phrase-table.half.f2e.gz --GoodTuring 0
/home/ubuntu/m/mosesdecoder/scripts/generic/score-parallel.perl 16 "sort -S
10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/extract.inv.sorted.gz
/home/ubuntu/m/training/model/lex.e2f
/home/ubuntu/m/training/model/phrase-table.half.e2f.gz --Inverse 1
Executing: /home/ubuntu/m/mosesdecoder/scripts/generic/score-parallel.perl
16 "sort -S 10G " /home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/extract.inv.sorted.gz
/home/ubuntu/m/training/model/lex.e2f
/home/ubuntu/m/training/model/phrase-table.half.e2f.gz --Inverse 1
using gzip
Started Tue Aug 9 07:52:18 2016
using gzip
Started Tue Aug 9 07:52:18 2016
/home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/tmp.9229/extract.0.gz
/home/ubuntu/m/training/model/lex.f2e
/home/ubuntu/m/training/model/tmp.9229/phrase-table.half.0000000.gz
--GoodTuring 2>> /dev/stderr
/home/ubuntu/m/mosesdecoder/scripts/../bin/score
/home/ubuntu/m/training/model/tmp.9230/extract.0.gz
/home/ubuntu/m/training/model/lex.e2f
/home/ubuntu/m/training/model/tmp.9230/phrase-table.half.0000000.gz
--Inverse 2>> /dev/stderr
/home/ubuntu/m/training/model/tmp.9229/
run.0.sh/home/ubuntu/m/training/model/tmp.9229/run.1.sh/home/ubuntu/m/training/model/tmp.9229/run.2.sh/home/ubuntu/m/training/model/tmp.9229/run.3.sh/home/ubuntu/m/training/model/tmp.9229/run.4.sh/home/ubuntu/m/training/model/tmp.9229/run.5.sh/home/ubuntu/m/training/model/tmp.9229/run.6.sh/home/ubuntu/m/training/model/tmp.9229/run.7.sh/home/ubuntu/m/training/model/tmp.9229/run.8.sh/home/ubuntu/m/training/model/tmp.9230/run.0.sh/home/ubuntu/m/training/model/tmp.9229/run.9.sh/home/ubuntu/m/training/model/tmp.9230/run.1.sh/home/ubuntu/m/training/model/tmp.9230/run.2.sh/home/ubuntu/m/training/model/tmp.9229/run.11.sh/home/ubuntu/m/training/model/tmp.9229/run.10.sh/home/ubuntu/m/training/model/tmp.9229/run.13.sh/home/ubuntu/m/training/model/tmp.9230/run.4.sh/home/ubuntu/m/training/model/tmp.9229/run.14.sh/home/ubuntu/m/training/model/tmp.9230/run.5.sh/home/ubuntu/m/training/model/tmp.9230/run.3.sh/home/ubuntu/m/training/model/tmp.9229/run.12.sh/home/ubuntu/m/training/model/tmp.9230/run.7.sh/home/ubuntu/m/training/model/tmp.9230/run.8.sh/home/ubuntu/m/training/model/tmp.9230/run.6.sh/home/ubuntu/m/training/model/tmp.9230/run.9.sh/home/ubuntu/m/training/model/tmp.9230/run.11.sh/home/ubuntu/m/training/model/tmp.9230/run.10.sh/home/ubuntu/m/training/model/tmp.9230/run.14.sh/home/ubuntu/m/training/model/tmp.9230/run.13.sh/home/ubuntu/m/training/model/tmp.9230/run.15.shmv
/home/ubuntu/m/training/model/tmp.9229/phrase-table.half.0000000.gz
/home/ubuntu/m/training/model/phrase-table.half.f2e.gzmv: cannot stat
'/home/ubuntu/m/training/model/tmp.9229/phrase-table.half.0000000.gz': No
such file or directory
Exit code: 1
ERROR: Scoring of phrases failed at
/home/ubuntu/m/mosesdecoder/scripts/training/train-model.perl line 1786.
gunzip -c /home/ubuntu/m/training/model/tmp.9230/phrase-table.half.*.gz 2>>
/dev/stderr| LC_ALL=C sort -S 10G -T
/home/ubuntu/m/training/model/tmp.9230 | gzip -c >
/home/ubuntu/m/training/model/phrase-table.half.e2f.gz 2>> /dev/stderr rm
-rf /home/ubuntu/m/training/model/tmp.9230
Finished Tue Aug 9 07:52:18 2016
(6.6) consolidating the two halves @ Tue Aug 9 07:52:18 UTC 2016
Executing: /home/ubuntu/m/mosesdecoder/scripts/../bin/consolidate
/home/ubuntu/m/training/model/phrase-table.half.f2e.gz
/home/ubuntu/m/training/model/phrase-table.half.e2f.gz /dev/stdout
--GoodTuring /home/ubuntu/m/training/model/phrase-table.half.f2e.gz.coc |
gzip -c > /home/ubuntu/m/training/model/phrase-table.gz
/home/ubuntu/m/mosesdecoder/scripts/../bin/consolidate: error while loading
shared libraries: libboost_serialization.so.1.59.0: cannot open shared
object file: No such file or directory
Executing: rm -f /home/ubuntu/m/training/model/phrase-table.half.*
(7) learn reordering model @ Tue Aug 9 07:52:18 UTC 2016
(7.1) [no factors] learn reordering model @ Tue Aug 9 07:52:18 UTC 2016
(7.2) building tables @ Tue Aug 9 07:52:18 UTC 2016
Executing:
/home/ubuntu/m/mosesdecoder/scripts/../bin/lexical-reordering-score
/home/ubuntu/m/training/model/extract.o.sorted.gz 0.5
/home/ubuntu/m/training/model/reordering-table. --model "wbe msd
wbe-msd-bidirectional-fe"
Lexical Reordering Scorer
scores lexical reordering models of several types (hierarchical,
phrase-based and word-based-extraction
terminate called after throwing an instance of 'util::ErrnoException'
what(): util/file.cc:76 in int util::OpenReadOrThrow(const char*) threw
ErrnoException because `-1 == (ret = open(name, 00))'.
No such file or directory while opening
/home/ubuntu/m/training/model/extract.o.sorted.gz
Aborted (core dumped)
Exit code: 134
ERROR: Lexical reordering scoring failed at
/home/ubuntu/m/mosesdecoder/scripts/training/train-model.perl line 1924.
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support