I managed to figure out the issue Apparently I needed to add to the extract options -T2S as i had source-syntax
Regards, Ayah On Fri, Sep 16, 2016 at 1:00 PM Ayah El Maghraby <ayah.elmaghr...@gmail.com> wrote: > I noticed it works when I remove the -ghkm option. Is there a specific > problem with GHKM extraction in tree-to-string models? > > On Fri, Sep 16, 2016 at 12:54 PM, Ayah El Maghraby < > ayah.elmaghr...@gmail.com> wrote: > >> training2.out contains output of stdout from step 4 to last step >> err.log contains output of stderr in from etract_parallel files & >> score_parallel >> >> On Fri, Sep 16, 2016 at 12:34 PM, Ayah El Maghraby < >> ayah.elmaghr...@gmail.com> wrote: >> >>> training2.out contains output of stdout from step 4 to last step >>> err.log contains output of stderr in from etract_parallel files & >>> score_parallel >>> training.out contains all the training output >>> >>> >>> On Fri, Sep 16, 2016 at 9:12 AM, Hieu Hoang <hieuho...@gmail.com> wrote: >>> >>>> can you redirect the standard out AND standard error to separate files >>>> and send them to me. >>>> >>>> Hieu Hoang >>>> http://www.hoang.co.uk/hieu >>>> >>>> On 16 September 2016 at 08:31, Ayah El Maghraby < >>>> ayah.elmaghr...@gmail.com> wrote: >>>> >>>>> I redirected the stderr output to another file I got this: >>>>> Segmentation fault (core dumped) >>>>> Segmentation fault (core dumped) >>>>> Segmentation fault (core dumped) >>>>> Segmentation fault (core dumped) >>>>> Score v2.1 -- scoring methods for extracted rules >>>>> processing hierarchical rules >>>>> adjusting phrase translation probabilities with Good Turing discounting >>>>> Loading lexical translation table from >>>>> /home/ayah/Documents/working/train/model/lex.f2e............................................................... >>>>> >>>>> Score v2.1 -- scoring methods for extracted rules >>>>> using inverse mode >>>>> processing hierarchical rules >>>>> Loading lexical translation table from >>>>> /home/ayah/Documents/working/train/model/lex.e2f............................................................... >>>>> >>>>> >>>>> gzip: >>>>> /home/ayah/Documents/working/train/model/tmp.16751/phrase-table.half.0000000.gz: >>>>> not in gzip format >>>>> >>>>> On Fri, Sep 16, 2016 at 4:14 AM, Hieu Hoang <hieuho...@gmail.com> >>>>> wrote: >>>>> >>>>>> what does the error say? >>>>>> >>>>>> look in tne training2.out file >>>>>> >>>>>> On 15/09/2016 15:56, Ayah El Maghraby wrote: >>>>>> >>>>>> Hello, >>>>>> >>>>>> I am getting a segmentation fault (core dumped) in step number 5 >>>>>> during training extract phrases parallel. >>>>>> I am using mosesdecoder latest version from github, I also tried >>>>>> Release 3.0 >>>>>> I am working on ubuntu 14.04 I also tried 16.04 >>>>>> I tried building moses with default options & i tried building after >>>>>> specifying boost >>>>>> I am using this command for training : >>>>>> nohup nice /home/ayah/mosesdecoder/scripts/training/train-model.perl >>>>>> --first-step 4 -alignment-file >>>>>> /home/ayah/Documents/working/train/model/aligned --source-syntax >>>>>> -root-dir >>>>>> train -corpus /home/ayah/Documents/trainData/100%/mosesTree -f en -e ar >>>>>> -hierarchical -glue-grammar -max-phrase-length 5 -ghkm >>>>>> --score-options="--GoodTuring" -lm 0:3:/home/ayah/Documents/lm/ >>>>>> truecase.en-ar.blm.ar:8 -external-bin-dir >>>>>> /home/ayah/mosesdecoder/tools/mgizapp -mgiza -mgiza-cpus 2 -snt2cooc >>>>>> snt2cooc.pl -sort-batch-size 253 -sort-compress gzip --giza-option >>>>>> m1=5,m2=2,mh=0,m3=3,m4=3 > training2.out >>>>>> >>>>>> Any ideas? I have plenty of memory >>>>>> >>>>>> Regards, >>>>>> Ayah >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Moses-support mailing >>>>>> listMoses-support@mit.eduhttp://mailman.mit.edu/mailman/listinfo/moses-support >>>>>> >>>>>> >>>>>> >>>>> >>>> >>> >> >
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
