Thanks Hieu.
It's not a problem with the Moses scripts or temp files. My bad in my description. A DoMY CE user built a virtual machine for testing with a very small virtual disk root (only) partition. The root (not temp) partition ran out of space when step 4 tried to write its output. He increased the partition size and everything worked. Tom On Wed, 13 Jun 2012 07:04:29 +0100, Hieu Hoang wrote: Hi tom That's odd. Nothing should be writing to the temp position when you use the training scripts, not even intermediate sort files. For the reason you encountered. Is the source of the problem one of the Moses scripts? Hieu Sent from my flying horse On 13 Jun 2012, at 01:20 AM, Tom Hoar wrote: Found the source of the problem. The new host didn't have enough free space on the temp partition. On Tue, 12 Jun 2012 15:22:41 +0700, Tom Hoar wrote: I have an en-fr corpus that appears normal, i.e. ran clean-corpus-n, removed control characters, etc. Train-model.perl succeeded steps 1, 2, and 3. Step 4, however, failed. The original log was accidentally over-written and I don't have immediate access to the output files. So, before re-running steps 1-3, does anyone have ideas about what kinds of data can cause train-model.perl to fail at step 4? Thanks. Tom _______________________________________________ Moses-support mailing list [email protected] [3] http://mailman.mit.edu/mailman/listinfo/moses-support [4] Links: ------ [1] mailto:[email protected] [2] mailto:[email protected] [3] mailto:[email protected] [4] http://mailman.mit.edu/mailman/listinfo/moses-support
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
