Hi Moses Team,

I'm running an MT system with 2 phrase tables and 2 reordering tables in
conjunction using the default 4 probabilities in the former and 6
probabilities in the latter.

When I call it with this (and I'm running lots of jobs in parallel so I may
run this command 20 times at once just varying the suffix on the input file
'del')

$MOSESDECODER_ROOT/bin/moses -threads 4 -config
$PWD/seed/$seedSize/iteration$iterations/moses.ini -input-file data/del$var
1> seed/$seedSize/iteration$iterations/del$var.lev 2>
seed/$seedSize/iteration$iterations/del$var.log

it never translates a single line, but it also doesn't die in the same
place every time. Sometimes the error message is:

Loading UnknownWordPenalty0

Loading WordPenalty0

Loading PhrasePenalty0

Loading LexicalReordering0

Loading table into memory...found inconsistent number of probabilities...
found 4 expected 6

and sometimes it loads the lexical reordering table just fine but segFaults
while trying to translate the first line.

I am positive that there are in fact 6 probabilities in every line of the
reordering table so I'm especially confused about that error... any ideas?

Here is my config file:

# input factors
[input-factors]
0

# mapping steps
[mapping]
0 T 0
1 T 1

[distortion-limit]
6

# feature functions
[feature]
UnknownWordPenalty
WordPenalty
PhrasePenalty
PhraseDictionaryMemory name=TranslationModel0 num-features=4
path=/scratch/ae1541/EMNLP/seed/$$$/iteration0/phrase-table
PhraseDictionaryMemory name=TranslationModel1 num-features=4
path=/scratch/ae1541/EMNLP/seed/$$$/iteration***/phrase-table
input-factor=0 output-factor=0
LexicalReordering name=LexicalReordering0 num-features=6
type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0
path=/scratch/ae1541/EMNLP/seed/$$$/iteration0/reordering-table
LexicalReordering name=LexicalReordering1 num-features=6
type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0
path=/scratch/ae1541/EMNLP/seed/$$$/iteration***/reordering-table
Distortion
KENLM name=LM0 factor=0 path=/scratch/ae1541/EMNLP/seed/LM/LM.lev.lm order=5

# dense weights for feature functions
[weight]
UnknownWordPenalty0= 1
WordPenalty0= -1
PhrasePenalty0= 0.2
TranslationModel0= 0.2 0.2 0.2 0.2
TranslationModel1= 0.2 0.2 0.2 0.2
LexicalReordering0= 0.3 0.3 0.3 0.3 0.3 0.3
LexicalReordering1= 0.3 0.3 0.3 0.3 0.3 0.3
Distortion0= 0.3
LM0= 0.5

Thanks!

-- 
Alex Erdmann
PhD Student in Linguistics at The Ohio State University
Visiting Scholar at NYU Abu Dhabi
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to