Hi Hieu,

thx for the answer.

i guess the problem relies in the fact that moses.ini is missing the  
third factor (credits to Nicola). It should be

[input-factors]
0
1
2

fact is, if i use a 2 factor input i get the error:

Collecting options took 0.039 seconds
Assertion failed: ((*contextFactor[count-1])[factorType] != NULL),  
function GetValue, file LanguageModelSRI.cpp, line 154.
Abort trap

if i use a 3 factor input (the right one in my opinion), i get:

[ERROR] Malformed input at
  Expected input to have words composed of 2 factor(s) (form FAC1| 
FAC2|...)
  but instead received input with 3 factor(s).
Abort trap


Maybe the moses.ini is not correctly generated, or (more probably) i  
missed something about parameters and switches in the training (the  
complete command is in my first mail).

however, i built a new model using other parameters and now i can  
decode.
well, didn't finish yet because of memory problems after 50% of  
decoding, but at least the decoding started.

Thus, the problem could be considered as solved.

Thanks again for your support,

Stefan


P.S. the new parameters are:
--translation-factors 0-0,2+1,2-0,2+1-0,2
--generation-factors 1,2-0+1-2
  --decoding-steps t0,g0,t1,g1:t2
--alignment-factors 0,1-0,1
--reordering-factors 0-0



On 11/giu/09, at 12:09, Hieu Hoang wrote:

> hey stefan
>
> the decoding can use either the 1st or second phrase table. however,  
> the 1st table doesn't have factor 2 (lemmas, i believe) so the  
> decoder crashes
>
> Stefan Rigo wrote:
>> here it is
>> thx
>>> The moses.ini is:
>>>
>>> #########################
>>> ### MOSES CONFIG FILE ###
>>> #########################
>>>
>>> # input factors
>>> [input-factors]
>>> 0
>>> 1
>>>
>>> # mapping steps
>>> [mapping]
>>> 0 T 0
>>> 1 T 1
>>>
>>> # translation tables: source-factors, target-factors, number of   
>>> scores, file
>>> [ttable-file]
>>> 0 0 5 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/ 
>>> work/ pos_srilm_generation5/model/phrase-table.0-0.gz
>>> 1 0,2 5 /Volumes/data/Users/stefan/Documents/studi/Master/mt/ 
>>> tools/ work/pos_srilm_generation5/model/phrase-table.1-0,2.gz
>>>
>>> # no generation models, no generation-file section
>>>
>>> # language models: type(srilm/irstlm), factors, order, file
>>> [lmodel-file]
>>> 0 0 3 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/  
>>> langmodels/baseline.3token.srilm
>>> 0 2 3 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/  
>>> langmodels/baseline.3pos.srilm
>>>
>>>
>>> # limit on how many phrase translations e for each phrase f are  
>>> loaded
>>> # 0 = all elements loaded
>>> [ttable-limit]
>>> 20
>>> 0
>>>
>>> # distortion (reordering) files
>>> [distortion-file]
>>> 0-0 msd-bidirectional-fe 6 /Volumes/data/Users/stefan/Documents/ 
>>> studi/ Master/mt/tools/work/pos_srilm_generation5/model/reordering- 
>>> table.0-0.gz
>>> 1-1 msd-bidirectional-fe 6 /Volumes/data/Users/stefan/Documents/ 
>>> studi/ Master/mt/tools/work/pos_srilm_generation5/model/reordering- 
>>> table.1-1.gz
>>>
>>> # distortion (reordering) weight
>>> [weight-d]
>>> 0.3
>>> 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]
>>> 0.2500
>>> 0.2500
>>>
>>>
>>> # 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
>>>
>>> [distortion-limit]
>>> 6
>>
>> On 10/giu/09, at 20:26, Hieu Hoang wrote:
>>
>>>> The moses.ini is:
>>>>
>>>> #########################
>>>> ### MOSES CONFIG FILE ###
>>>> #########################
>>>>
>>>> # input factors
>>>> [input-factors]
>>>> 0
>>>> 1
>>>>
>>>> # mapping steps
>>>> [mapping]
>>>> 0 T 0
>>>> 1 T 1
>>>>
>>>> # translation tables: source-factors, target-factors, number of   
>>>> scores, file
>>>> [ttable-file]
>>>> 0 0 5 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/ 
>>>> work/ pos_srilm_generation5/model/phrase-table.0-0.gz
>>>> 1 0,2 5 /Volumes/data/Users/stefan/Documents/studi/Master/mt/ 
>>>> tools/ work/pos_srilm_generation5/model/phrase-table.1-0,2.gz
>>>>
>>>> # no generation models, no generation-file section
>>>>
>>>> # language models: type(srilm/irstlm), factors, order, file
>>>> [lmodel-file]
>>>> 0 0 3 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/  
>>>> langmodels/baseline.3token.srilm
>>>> 0 2 3 /Volumes/data/Users/stefan/Documents/studi/Master/mt/tools/  
>>>> langmodels/baseline.3pos.srilm
>>>>
>>>>
>>>> # limit on how many phrase translations e for each phrase f are  
>>>> loaded
>>>> # 0 = all elements loaded
>>>> [ttable-limit]
>>>> 20
>>>> 0
>>>>
>>>> # distortion (reordering) files
>>>> [distortion-file]
>>>> 0-0 msd-bidirectional-fe 6 /Volumes/data/Users/stefan/Documents/ 
>>>> studi/ Master/mt/tools/work/pos_srilm_generation5/model/ 
>>>> reordering-table.0-0.gz
>>>> 1-1 msd-bidirectional-fe 6 /Volumes/data/Users/stefan/Documents/ 
>>>> studi/ Master/mt/tools/work/pos_srilm_generation5/model/ 
>>>> reordering-table.1-1.gz
>>>>
>>>> # distortion (reordering) weight
>>>> [weight-d]
>>>> 0.3
>>>> 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]
>>>> 0.2500
>>>> 0.2500
>>>>
>>>>
>>>> # 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
>>>>
>>>> [distortion-limit]
>>>> 6
>>>
>>
>>
>> -------------------------------------
>> I'm on the web:
>> http://www.onexposure.net/?member=8999&stefan-rigo 
>> <http://www.onexposure.net/?member=8999&stefan-rigo 
>> >
>> http://www.grandwazoo.net
>>
>> Skype name: stfz65
>>
>>
>>
>>
>>
>>
>>
>>


-------------------------------------
I'm on the web:
http://www.onexposure.net/?member=8999&stefan-rigo
http://www.grandwazoo.net

Skype name: stfz65








_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to