Hi..
I am building a statistical machine translator from English to
Malayalam. I have completed steps upto tuning. But while running the
decoder with an English sentence as input, the translated output contains
only UNKs.
Defined parameters (per moses.ini or switch):
config: /home/user/tools/model/moses.ini
distortion-limit: 6
input-factors: 0
lmodel-file: 1 0 3 /home/user/tools/corpus/training/final.lm.ml.gz
mapping: 0 T 0
ttable-file: 0 0 0 5 /home/user/tools/model/phrase-table.gz
ttable-limit: 20
weight-d: 0.6
weight-l: 0.5000
weight-t: 0.20 0.20 0.20 0.20 0.20
weight-w: -1
/home/user/tools/mosesdecoder/bin
ScoreProducer: Distortion start: 0 end: 1
ScoreProducer: WordPenalty start: 1 end: 2
ScoreProducer: !UnknownWordPenalty start: 2 end: 3
Loading lexical distortion models...have 0 models
Start loading LanguageModel /home/user/tools/corpus/training/final.lm.ml.gz
: [0.014] seconds
In LanguageModelIRST::Load: nGramOrder = 3
Language Model Type of /home/user/tools/corpus/training/final.lm.ml.gz is 1
Language Model Type is 1
iARPA
loadtxt_ram()
1-grams: reading 2582 entries
done level 1
2-grams: reading 4388 entries
done level 2
3-grams: reading 733 entries
done level 3
done
OOV code is 2581
OOV code is 2581
IRST: m_unknownId=2581
ScoreProducer: LM start: 3 end: 4
Finished loading LanguageModels : [0.104] seconds
Start loading PhraseTable /home/user/tools/model/phrase-table.gz : [0.104]
seconds
filePath: /home/user/tools/model/phrase-table.gz
ScoreProducer: PhraseModel start: 4 end: 9
Finished loading phrase tables : [0.106] seconds
Start loading phrase table from /home/user/tools/model/phrase-table.gz :
[0.106] seconds
Reading /home/user/tools/model/phrase-table.gz
----5---10---15---20---25---30---35---40---45---50---55---60---65---70---75---80---85---90---95--100
****************************************************************************************************
Finished loading phrase tables : [0.433] seconds
IO from STDOUT/STDIN
Created input-output object : [0.433] seconds
How are you.
Translating line 0 in thread id 3050224448
Translating: How are you.
Line 0: Collecting options took 0.000 seconds
Line 0: Search took 0.000 seconds
How are you.
BEST TRANSLATION: How|UNK|UNK|UNK are|UNK|UNK|UNK you.|UNK|UNK|UNK [111]
[total=-322.812]
core=(0.000,-3.000,-300.000,-51.624,0.000,0.000,0.000,0.000,0.000)
Line 0: Translation took 0.026 seconds total
reset caches
Please help me to get rid of this problem.
Thanking you in advance.
--
Nithya
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support