Hi Souhir, please take a close look at the log file.
It contains all the commands that were run in sequence. I would suggest to check if any of the intermediate files were generated as expected, and if not, execute single commands on the command line by hand. This should help you to at least pinpoint the problem. Regards, Philipp On Mon, Jul 9, 2012 at 1:48 AM, Souhir Gahbiche <[email protected]> wrote: > 2012/7/6, Christophe Servan <[email protected]>: >> Hi all, >> it's ok for me, it can be done. >> >> Christophe >> >> Le 06/07/2012 12:20, Hieu Hoang a écrit : >>> if christoph doesn't object, I'd like to move mert-moses-multi.pl into >>> the contrib directory. We could look at it at a later date and see if we >>> can merge it into the main mert script. >>> >>> On 05/07/2012 20:49, Marcin Junczys-Dowmunt wrote: >>>> As I learnt the hard way, the "multi" in mert-moses-multi.pl is not >>>> related to multithread, but to a different scoring method :) Maybe this >>>> should be somehow emphasized. >>>> Best, >>>> Marcin >>>> >>>> W dniu 05.07.2012 19:50, Tom Hoar pisze: >>>>> mert-moses.pl and mert-moses-multi.pl both have the command >>>>> argument >>>>> >>>>> --threads=NUMBER ... Use multi-threaded mert (must be >>>>> compiled >>>>> in) >>>>> >>>>> Can someone explain the caveat "must be compiled in"? >>>>> >>>>> Also, since both mert-moses.pl and mert-moses-multi.pl have the >>>>> --threads argument, what is the difference between the two and is >>>>> there >>>>> a situation when one is preferred over the other? >>>>> >>>>> >>>>> >>>>> >>>>> On Thu, 05 Jul 2012 17:48:03 +0100, Hieu Hoang >>>>> <[email protected]> wrote: >>>>>> hi souhir >>>>>> >>>>>> there was a bug in the score program a year ago that caused it to run >>>>>> slow. If you haven't done so, you should update to the latest source >>>>>> code. The latest code also lets you parallelize step 6 so that it >>>>>> will >>>>>> run faster on multi-core machines >>>>>> >>>>>> On 05/07/2012 15:37, Souhir Gahbiche wrote: >>>>>>> Hi all, >>>>>>> >>>>>>> I'm scoring a huge phrase-table (6.5M phrase pairs). Moses is >>>>>>> running >>>>>>> since 4 days and stills in step 6.2. >>>>>>> The program seems running very slowly and not running constantly. >>>>>>> I remember, I trained a huge phrase table bigger than this but the >>>>>>> training was faster. >>>>>>> Is it normal that training stills all this time? >>>>>>> >>>>>>> Regards >>>>>>> -S >>>>>>> _______________________________________________ >>>>>>> Moses-support mailing list >>>>>>> [email protected] >>>>>>> http://mailman.mit.edu/mailman/listinfo/moses-support >>>>>>> >>>>>> _______________________________________________ >>>>>> Moses-support mailing list >>>>>> [email protected] >>>>>> http://mailman.mit.edu/mailman/listinfo/moses-support > Hi all, > Thanks a lot for your answers. > I tried the last version of moses, it was much faster, but the problem > is that I got an empty phrase-table, and the reordering model wasn't > empty. > I joined the log file, it seems ok? > S > > >>>>> _______________________________________________ >>>>> Moses-support mailing list >>>>> [email protected] >>>>> http://mailman.mit.edu/mailman/listinfo/moses-support >>>>> >>> >>> _______________________________________________ >>> Moses-support mailing list >>> [email protected] >>> http://mailman.mit.edu/mailman/listinfo/moses-support >> >> >> _______________________________________________ >> Moses-support mailing list >> [email protected] >> http://mailman.mit.edu/mailman/listinfo/moses-support >> > > _______________________________________________ > Moses-support mailing list > [email protected] > http://mailman.mit.edu/mailman/listinfo/moses-support > _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
