python - Moses v1.0 多语言ini文件

Question

我当时使用的是 mosesserver 0.91，一切正常，但现在有了 1.0 版，一切都和以前不一样了。

这是我的情况:

我想要从阿拉伯语到英语和从英语到阿拉伯语的多语言翻译。我拥有的所有数据和配置文件都适用于 0.91 版本的 mosesserver。这是我的配置文件:

-------------------------------------------------
#########################
### MOSES CONFIG FILE ###
#########################

# D - decoding path, R - reordering model, L - language model
[translation-systems]
ar-en D 0 R 0 L 0
en-ar D 1 R 1 L 1

# input factors
[input-factors]
0

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

# translation tables: table type (hierarchical(0), textual (0), binary (1)), source-factors, target-factors, number of scores, file 
# OLD FORMAT is still handled for back-compatibility
# OLD FORMAT translation tables: source-factors, target-factors, number of scores, file 
# OLD FORMAT a binary table type (1) is assumed 
[ttable-file]
1 0 0 5 /mnt/models/ar-en/phrase-table/phrase-table
1 0 0 5 /mnt/models/en-ar/phrase-table/phrase-table

# no generation models, no generation-file section

# language models: type(srilm/irstlm), factors, order, file
[lmodel-file]
1 0 5 /mnt/models/ar-en/language-model/en.qblm.mm
1 0 5 /mnt/models/en-ar/language-model/ar.lm.d1.blm.mm


# limit on how many phrase translations e for each phrase f are loaded
# 0 = all elements loaded
[ttable-limit]
20

# distortion (reordering) files
[distortion-file]
0-0 wbe-msd-bidirectional-fe-allff 6 /mnt/models/ar-en/reordering-table/reordering-table.wbe-msd-bidirectional-fe.gz
0-0 wbe-msd-bidirectional-fe-allff 6 /mnt/models/en-ar/reordering-model/reordering-table.wbe-msd-bidirectional-fe.gz

# distortion (reordering) weight
[weight-d]
0.3
0.3

# lexicalised distortion weights
[weight-lr]
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3

# language model weights
[weight-l]
0.5000
0.5000

# translation model weights
[weight-t]
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2

# no generation models, no weight-generation section

# word penalty
[weight-w]
-1
-1

[distortion-limit]
12
---------------------------------------------------------

所以请有人帮我重写这个配置文件，让它可以在 1.0 版本中工作。我需要一些 python 示例翻译代码。我在 python 和 earler 中使用 xmlrpc 我发送了 http 请求:

import xmlrpclib

client = xmlrpclib.ServerProxy('http://localhost:8080')
client.translate({'text': 'some text', 'system': 'en-ar'})

但现在似乎没有更多的“系统”参数，摩西始终使用默认设置。

Answer 1

我修好了。这是新的 ini 文件:

[input-factors]
 0

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

 [distortion-limit]
 12

 # feature functions
 [feature]
 Distortion
 UnknownWordPenalty
 WordPenalty
 PhraseDictionaryBinary name=TranslationModel0 num-features=5 path=/mnt/models/ar-en/phrase-table/phrase-table input-factor=0 output-factor=0 table-limit=20
 LexicalReordering num-features=6 name=LexicalReordering0 type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0 path=/mnt/models/ar-en/reordering-table/reordering-table.wbe-msd-bidirectional-fe.gz 
 IRSTLM name=LM0 order=5 factor=0 path=/mnt/models/ar-en/language-model/en.qblm.mm
 PhraseDictionaryBinary name=TranslationModel1 num-features=5 path=/mnt/models/en-ar/phrase-table/phrase-table input-factor=0 output-factor=0 table-limit=20
 LexicalReordering num-features=6 name=LexicalReordering1 type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0 path=/mnt/models/en-ar/reordering-model/reordering-table.wbe-msd-bidirectional-fe.gz 
 IRSTLM name=LM1 order=5 factor=0 path=/mnt/models/en-ar/language-model/ar.lm.d1.blm.mm

 # core weights - not used
 [weight]
 Distortion0= 0
 WordPenalty0= 0
 TranslationModel0= 0 0 0 0 0
 LexicalReordering0= 0 0 0 0 0 0
 LM0= 0
 TranslationModel1= 0 0 0 0 0
 LexicalReordering1= 0 0 0 0 0 0
 LM1= 0

 [alternate-weight-setting]
 id=ar-en ignore-ff=LM1,LexicalReordering1 ignore-decoding-path=1
 Distortion0= 0.3
 WordPenalty0= -1
 TranslationModel0= 0.2 0.2 0.2 0.2 0.2
 LexicalReordering0= 0.3 0.3 0.3 0.3 0.3 0.3
 LM0= 0.5
 id=en-ar ignore-ff=LM0,LexicalReordering0 ignore-decoding-path=0
 Distortion0= 0.3
 WordPenalty0= -1
 TranslationModel1= 0.2 0.2 0.2 0.2 0.2
 LexicalReordering1= 0.3 0.3 0.3 0.3 0.3 0.3
 LM1= 0.5

客户端的示例代码是:

import xmlrpclib

client = xmlrpclib.ServerProxy('http://localhost:8080')
client.translate({'text': '<seg weight-setting="%s">%s</seg>' % ('en-ar', some_text)})

其中“en-ar”是翻译系统，就我而言，它是从英语翻译成阿拉伯语(我还有其他翻译系统“ar-en”(从阿拉伯语到英语)，您可以从 ini 文件中看到)。