Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread J C Read
I'm working on a laptop as well, with 512MB memory. I wouldn't advise trying to run giza on a whole corpus. I tried it and 3 days later gave up waiting for it to finish (paqe files = Slw). My feelings about giza is that it needs some kind of dynamic way of figuring out how much actual physical

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread Hubert Crépy
J C Read a écrit : According to wikipedia http://en.wikipedia.org/wiki/SIGSEGV signal 11 indicates an invalid memory reference. Yes, definitely, what we also call a "coredump" under AIX. I eventually figured out that this was because of the data I was using. That's often the case, an unfo

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread J C Read
According to wikipedia http://en.wikipedia.org/wiki/SIGSEGV signal 11 indicates an invalid memory reference. I kept getting a signal 6 (abort). I eventually figured out that this was because of the data I was using. Things to check: Is the data sentence aligned? Has the data been cleaned with t

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread J C Read
This bug has already been noted and an updated file has been emailed to the developers. I don't know whether it has been merged yet. Quoting Hubert Crépy <[EMAIL PROTECTED]>: > In addition to my own failings and setup oddities, I found what I think > to be a small bug in the build setup: > I had

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread Hubert Crépy
Okay, I've been able to set it all up, compute the LM, and have started the training. To "clean the pipes", I'm doing a run with little data (started from 100K sentences only, which were trimmed to about 74K sentences when removing long ones). The training fails in step 2.1b, GIZA:

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread Hubert Crépy
In addition to my own failings and setup oddities, I found what I think to be a small bug in the build setup: I had to edit released-filesWIN32 to list: training/mbr/mbr.exe It was missing the trailing .exe and rsync was failing. -- Hubert Crépy

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread Hubert Crépy
Okay, I've been able to set it all up, compute the LM, and have started the training. To "clean the pipes", I'm doing a run with little data (started from 100K sentences only, which were trimmed to about 74K sentences when removing long ones). The training fails in step 2.1b, GIZA:

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread Hubert Crépy
J C Read <[EMAIL PROTECTED]> writes: > Did you run ./regenerate-makefiles.sh ? I had, but had missed a couple of error messages (due to incomplete Cygwin installation, missing aclocal, automake and autoconf) and just relied on the comforting message: "you should be able to run ./configure..." T

Re: [Moses-support] Training scripts/executables under Win32 ?

2008-02-22 Thread J C Read
The reason you are getting the message 'No such file or directory' is because the configure file has not been created yet. This means you have not yet successfully completed the regenerate makefiles process. Did you run ./regenerate-makefiles.sh ? Quoting Hubert Crépy <[EMAIL PROTECTED]>: > J C

[Moses-support] Training scripts/executables under Win32 ?

2008-02-20 Thread Hubert Crépy
Hello all, newbie warning !!! I am reading somewhat conflicting information about the feasibility of running a MOSES training under Windows (XP Pro, if that matters). Am getting the impression that this is a fluctuating situation and at the moment not for the faint-hearted... Unfortunately, I

[Moses-support] Training scripts/executables under Win32 ?

2008-02-20 Thread Hubert Crépy
Hello all, newbie warning !!! I am reading somewhat conflicting information about the feasibility of running a MOSES training under Windows (XP Pro, if that matters). Am getting the impression that this is a fluctuating situation and at the moment not for the faint-hearted... Unfortunately, I