Hi Bui
I think it's the same problem as you had before. The filtering script is
getting rid of all the rules. You're either using the wrong filtering
script, or there's a bug in it. The correct filtering script for chart
decoding is
filter-rule-table.py
Alternatively, don't filter at all but binarize the whole phrase table,
which is explained at the bottom of this page
http://www.statmt.org/moses/?n=Moses.SyntaxTutorial
On 10/05/2010 09:27, Bui Hung wrote:
Hi,
I see that when I run <http://filter-model-given-input.pl>
mert-moses.pl <http://mert-moses.pl> in moses-chart
In the result there are two line as following:
filtering
/home/i8001/s0810207/Mose-chartN/scripts/working-dir/model/rule-table.gz
->
/home/i8001/s0810207/Mose-chartN/scripts/working-dir/tuning/mert/filtered/phrase-table.0-0.1...
0 of 36566112 phrases pairs used (0.00%) - note: max length 10
filtering
/home/i8001/s0810207/Mose-chartN/scripts/working-dir/model/glue-grammar ->
/home/i8001/s0810207/Mose-chartN/scripts/working-dir/tuning/mert/filtered/phrase-table.0-0.2...
But I see that phrase-table.0-0.1 and phrase-table.0-0.2 are empty in
filterd folder
So that in the result of mert-moses is No Best Translation.
Translating: <s> encore une étape cruciale pour les balkans </s>
[0..0]=1 [1..1]=1 [2..2]=1 [3..3]=1 [4..4]=1 [5..5]=1 [6..6]=1
[7..7]=1 [8..8]=1 [0..1]=0 [1..2]=0 [2..3]=0 [3..4]=0 [4..5]=0
[5..6]=0 [6..7]=0 [7..8]=0 [0..2]=0 [1..3]=0 [2..4]=0 [3..5]=0
[4..6]=0 [5..7]=0 [6..8]=0 [0..3]=0 [1..4]=0 [2..5]=0 [3..6]=0
[4..7]=0 [5..8]=0 [0..4]=0 [1..5]=0 [2..6]=0 [3..7]=0 [4..8]=0
[0..5]=0 [1..6]=0 [2..7]=0 [3..8]=0 [0..6]=0 [1..7]=0 [2..8]=0
[0..7]=0 [1..8]=0 [0..8]=0Num of hypo = 9 --- cells:
0 1 2 3 4 5 6 7 8
1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0
0 0 0 0
0 0 0
0 0
0
NO BEST TRANSLATION
Do you think what is my problem, How can I solve it
I hope I will get your support
Thank you so much
Bui Hung Thanh
Japan Advanced Institute of Science and Technology (JAIST)
<http://www.jaist.ac.jp/index-e.html>
School of Information Science <http://www.jaist.ac.jp/is/index.html>
Shimazu and Shirai NLP Lab
1-1, Asahidai, Nomi, Ishikawa 923-1292 Japan
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
