Hello,

I try to do the Training the Translation System, but i have this problem :
 I got this message  :

This a part oh the whole message

Sent No: 43910 , No. Occurrences: 1
0 4001 351
17 442 5 336 30 31 1336 27 22 357 249 41 103 31 29 1336 27 8 31 1336 27 165
31 29 1336 27 66 1980 3
Corpus fits in memory, corpus has: 43916 sentence pairs.
 Train total # sentence pairs (weighted): 43916
Size of source portion of the training corpus: 525622 tokens
Size of the target portion of the training corpus: 507610 tokens
In source portion of the training corpus, only 25517 unique tokens appeared
In target portion of the training corpus, only 13809 unique tokens appeared
lambda for PP calculation in IBM-1,IBM-2,HMM:= 507610/(569538-43916)==
0.965732
There are 1853136 1853136 entries in table
==========================================================
Model1 Training Started at: Thu Jul 18 17:29:50 2013

-----------
Model1: Iteration 1
Model1: (1) TRAIN CROSS-ENTROPY 15.1022 PERPLEXITY 35174.2
Model1: (1) VITERBI TRAIN CROSS-ENTROPY inf PERPLEXITY inf
Model 1 Iteration: 1 took: 1 seconds
-----------
Model1: Iteration 2
Model1: (2) TRAIN CROSS-ENTROPY 7.9786 PERPLEXITY 252.231
Model1: (2) VITERBI TRAIN CROSS-ENTROPY 9.86202 PERPLEXITY 930.601
Model 1 Iteration: 2 took: 2 seconds
-----------
Model1: Iteration 3
Model1: (3) TRAIN CROSS-ENTROPY 7.59011 PERPLEXITY 192.686
Model1: (3) VITERBI TRAIN CROSS-ENTROPY 9.32305 PERPLEXITY 640.5
Model 1 Iteration: 3 took: 1 seconds
-----------
Model1: Iteration 4
Model1: (4) TRAIN CROSS-ENTROPY 7.45277 PERPLEXITY 175.189
Model1: (4) VITERBI TRAIN CROSS-ENTROPY 9.02775 PERPLEXITY 521.942
Model 1 Iteration: 4 took: 1 seconds
-----------
Model1: Iteration 5
Model1: (5) TRAIN CROSS-ENTROPY 7.39085 PERPLEXITY 167.829
Model1: (5) VITERBI TRAIN CROSS-ENTROPY 8.85007 PERPLEXITY 461.464
Model 1 Iteration: 5 took: 1 seconds
Entire Model1 Training took: 6 seconds
NOTE: I am doing iterations with the HMM model!
Read classes: #words: 0  #classes: 1
ERROR: no word index for '#085a00;'
ERROR: no word index for '#999999;'
ERROR: no word index for '#ffffff;'

...


Could someone please tell me if there is something obvious that I'm missing?
Regards,

Cyrine

-- 
*Cyrine NASRI
Ph.D. Student in Computer Science*
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to