Hi Hieu Hoang,
I've read two file and it looks ok.
glue-grammar:
<s> [X] ||| <s> [S] ||| ||| 1
[X][S] </s> [X] ||| [X][S] </s> [S] ||| 0-0 ||| 1
[X][S] [X][X] [X] ||| [X][S] [X][X] [S] ||| 0-0 1-1 ||| 2.718
rule-table.gz (I show ten head lines here)
! " , [X] ||| ! " [X] ||| ||| 0.0289855 0.0363779 1 0.80712 2.718 |||
34.5 1
! " , [X][X] , [X] ||| ! " [X][X] , [X] ||| 3-2 ||| 0.5 0.0292336 0.5
0.698392 2.718 ||| 0.1 0.1
! " , [X][X] , [X] ||| ! " [X][X] that , [X] ||| 3-2 ||| 0.5 0.0292336
0.5 0.0092224 2.718 ||| 0.1 0.1
! " , [X][X] . [X] ||| ! " [X][X] . [X] ||| 3-2 ||| 0.281171 0.0325239 1
0.799249 2.718 ||| 0.222285 0.0625
! " , [X][X] [X] ||| ! " [X][X] [X] ||| 3-2 ||| 0.197964 0.0363779
0.881617 0.80712 2.718 ||| 3.13491 0.703931
! " , [X][X] [X] ||| ! " [X][X] that [X] ||| 3-2 ||| 0.5 0.0363779
0.118383 0.0106582 2.718 ||| 0.166667 0.703931
! " , [X][X] anh [X] ||| ! " [X][X] he [X] ||| 3-2 ||| 0.5 0.00231865 1
0.0862683 2.718 ||| 0.0588236 0.0294118
! " , [X][X] cùng [X] ||| ! " [X][X] and [X] ||| 3-2 ||| 0.285714
4.182e-05 1 0.0218595 2.718 ||| 0.0729166 0.0208333
! " , [X][X] gia_?ình [X] ||| ! " [X][X] his family [X] ||| 3-2 ||| 0.5
0.0145356 1 0.0114879 2.718 ||| 0.032258 0.016129
! " , [X][X] ga(.ng_ho?i [X] ||| ! " [X][X] demanded [X] ||| 3-2 ||| 0.5
0.00363779 1 0.161424 2.718 ||| 0.181818 0.0909091
On 04/15/2010 05:03 PM, Hieu Hoang wrote:
hi vuong
it looks ok. Please double check that the phrase table files exists
/common_dir/thuvh/50001/model/rule-table.gz
/common_dir/thuvh/50001/model/glue-grammar
and that they have data in them. Can you send me a few lines of each
file to check
On 15/04/2010 04:01, Thu, Vuong Hoai wrote:
Dir Hieu Hoang.
I send you my ini file.
On 04/14/2010 11:05 PM, [email protected] wrote:
Message: 2
Date: Wed, 14 Apr 2010 18:07:30 +0700
From: "Thu, Vuong Hoai"<[email protected]>
Subject: [Moses-support] About tunning step
To: [email protected]
Message-ID:<[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi,
I use moses-chart (reversion 2951). I apply this to translate from
vietnamese to english. I think easy way is set f = vn, but when I
ran, I
got some error like this:
2<word> <word> <probabilistic>
line<line number> in<root-dir>/model/lex.f2e has wrong number of
tokens, skipping:
some error message like that and next to step tunning, each sentence in
developement set have no best tranlation sentences.
If I use more data, will the problem be solved?
Thank you!
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
--
Thu.
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
