Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

train-model.perl failed #230

Open
superyyy1202 opened this issue Apr 29, 2022 · 4 comments
Open

train-model.perl failed #230

superyyy1202 opened this issue Apr 29, 2022 · 4 comments

Comments

@superyyy1202
Copy link

I call train-model.perl, failed to generate the model folder and the inside of the content, only generate the folder, as shown in the figure below train. Out the log error is/home/yyy/working/train/corpus/useful. The VCB
Died with signal 11, with coredump. I have set ulimit -c unlimited to not limit the size of the core file, but still report this error. Could you please help me figure out what the problem is? I've been stuck here for three days.

@superyyy1202
Copy link
Author

/home/yyy/working/train/corpus/zh.vcb
ERROR: Execution of: /home/yyy/mosesdecoder-master/tools/GIZA++ -CoocurrenceFile /home/yyy/working/train/giza.zh-en/zh-en.cooc -c /home/yyy/working/train/corpus/zh-en-int-train.snt -m1 5 -m2 0 -m3 3 -m4 3 -model1dumpfrequency 1 -model4smoothfactor 0.4 -nodumps 1 -nsmooth 4 -o /home/yyy/working/train/giza.zh-en/zh-en -onlyaldumps 1 -p0 0.999 -s /home/yyy/working/train/corpus/en.vcb -t /home/yyy/working/train/corpus/zh.vcb
died with signal 11, with coredump

@hieuhoang
Copy link
Contributor

it's usually an issue with the data. eg. the data isn't encoded in utf8, there's non-printing characters, there's a double space separating words.

@superyyy1202
Copy link
Author

I have checked that the codes of my parallel corpus are ALL UTF-8, but I still reported the following errors. Could you please help me to check again./home/yyy/working/train/corpus/zh-en-int-train.snt > /home/yyy/working/train/giza.zh-en/zh-en.cooc
Segmentation fault (core dumped)
Exit code: 139
ERROR at /home/yyy/mosesdecoder-master/scripts/training/train-model.perl line 1351.

@superyyy1202
Copy link
Author

I found another set of data, which failed to call the kenLM provided by Moses, but could be generated successfully with IRSTLM.The file ./ build_binary can be called successfully.

Reading /home/yyy/corpus/train.seq.mx.clean.zh
----5---10---15---20---25---30---35---40---45---50---55---60---65---70---75---80---85---90---95--100


Segmentation fault (core dumped)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants