Thank you for taking a look so quickly. :) Is this something that mert should be capable of gracefully skipping over? Or is the appropriate behavior really to crash?
On Wed, Dec 1, 2010 at 3:50 PM, Barry Haddow <bhad...@inf.ed.ac.uk> wrote: > Hi Lane > > You have a -inf in your run2.features.dat, line 27768. This is what is > causing > mert to crash. Apologies for the lack of a meaningful error message, > > cheers > Barry > > On Wednesday 01 Dec 2010 17:18:21 Lane Schwartz wrote: > > I've got a MERT run (decoding took quite a while) that died during the > run2 > > optimization. I pinned down where it dies. If anyone could help figure > out > > what's going on here, I would really appreciate it. :) For reference, > > I've put the .dat and .opt files that mert is using at > > http://www.cs.umn.edu/~lane/mert.features.zip > > > > $ moses-trunk/mert/mert -d 15 --scconfig case:true -n 20 --ffile > > run1.features.dat,run2.features.dat --scfile > > run1.scores.dat,run2.scores.dat --ifile run2.init.opt > > > > > > Seeding random numbers with system clock > > > > Scorer config string: case:true > > > > name: case value: true > > > > Using scorer regularisation strategy: none > > > > Using scorer regularisation window: 0 > > > > Using case preservation: 1 > > > > Using reference length strategy: closest > > > > Data::score_type BLEU > > > > Data::Scorer type from Scorer: BLEU > > > > BleuScorer: 9 > > > > ScoreData: number_of_scores: 9 > > > > Loading Data from: run1.scores.dat and run1.features.dat > > > > loading feature data from run1.features.dat > > > > loading score data from run1.scores.dat > > > > Loading Data from: run2.scores.dat and run2.features.dat > > > > loading feature data from run2.features.dat > > > > loading score data from run2.scores.dat > > > > Data loaded : [1] seconds > > > > mert: Optimizer.cpp:161: statscore_t Optimizer::LineOptimize(const > Point&, > > const Point&, Point&) const: Assertion > > `abs(leftmost->first-gradient.rbegin()->first)<0.0001' failed. > > > > Aborted > > > > > > Cheers, > > Lane > > > > -- When a place gets crowded enough to require ID's, social collapse is not far away. It is time to go elsewhere. The best thing about space travel is that it made it possible to go elsewhere. -- R.A. Heinlein, "Time Enough For Love"
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
