Dear Eleni,
You can refer to the introduction in the source codes of iSenWeb in the
GITHUB.COM of mosesdecoder. More details are introduced in the link:
https://github.com/moses-smt/mosesdecoder/tree/master/contrib
--
You can refer to the release file in Moses package:
https://github.com/moses-smt/mosesdecoder/tree/master/contrib
In this folder, the application named iSenWeb (written by University of
Macau)is a web-based SMT system, you can read it and have a try.
If you have any problem, you can email to me
Hi Xinghua,
You can refer to the introduction about the Web translation in the Moses
manual. In this introduction, there is a demon.pl script, which will active the
moses server with different port numbers. I think if you want to translate
different language pairs you can set the different port
Dear Hongbin,
There is no "extract.o.gz" in your result, so there is maybe something wrong
with your training process. This is my genereated files in the model file:
tianliang@tianliang-VirtualBox:~/research/mosesdecoder/scripts/training/model$
ls -l
total 174336
-rw-rw-r-- 1 tianliang tianliang
Hi Henry,
please make sure the format of your command input in your terminal. I noted
that in your command there was no space in the "-tttble" parameter.
In the reference manual, -ttable int int string -- translation table file, use
'-' for stdin,
so, there is one space between the two "-", ple
Hi Barry,
It is a significative task to update the existing Moses manual. I think more
people can help during the task with some open source projects.
Recently, I have been developing my own webpage. In my point of view, maybe
collectiong the suggestions from many users (maybe focus on the profes
1801
发件人： 亮 田
收件人： Philipp Koehn
抄送： "moses-support@mit.edu"
发送日期： 2012年4月15日, 星期日, 下午 5:09
主题: 回复： [Moses-support] Question 2 in reordering table file
Dear Fong Puichi,
As Philipp had said, your first file is a word alignment file ( the file name
shoul be
Dear Fong Puichi,
As Philipp had said, your first file is a word alignment file ( the file name
shoul be named as grow-diag-final-and), which is not a reordering model file.
If you want to know more about it, you can run the nine training steps one by
one with the parameter --first-step --last
Hi,
It seems that you are running Moses on Windows with the cygwin. I think the
error comes from the path. You wanted to give the absolute path, but the path
may be not right. Try:
export SCRIPTS_ROOTDIR=/home/user/moses/scripts
$ train-model.perl --scripts-root-dir $SCRIPTS_ROOTDIR --corpus
Hi Tom,
Thank you very much for your kindly reply. I will refer to your suggestions
and if I successfully build the system I will share the building process with
others.
Liang
发件人： Tom Hoar
收件人： 亮 田
抄送： moses-support@mit.edu
发送日期： 2012年3月3日, 星期六, 下午 4
Hi everyone,
I want to try to build a phrase-based SMT system using Moses these days. But I
can find few documents or introductions in the Internet.
Does anyone have an experience to develop a web-based SMT system in your
Windows or Linux system/ server?
There is already an introduction in Mose
Dear sir,
Recently ,I use the rule-based machine translation to get the translation
results. But the result is not very fluent. Now I want to use the Moses decoder
to optimize the result,Can you kindly enough to give me some suggestion?
For example:
one result is
12 matches
Mail list logo
