Hi,
Now the filtering works correctly but the decoder still dies.
The command I'm running now, is:
nohup nice
/usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.en
/usr/local/share/moses/mosesdecoder/dist/bin/moses
train/model/moses.ini --mertdir
/usr/local/share/moses/mosesdecoder/dist/bin/ --filtercmd
'/usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-input.pl
-Binarizer "/usr/local/share/moses/mosesdecoder/dist/bin/CreateOnDisk 1 1 5
100 2" --Hierarchical' &>
mert.out
I'm attaching again the mert.out file although it seems to be similar
to the previous one:
nohup: ignoring input
main::create_extractor_script() called too early to check prototype at
/usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
line 681.
Using SCRIPTS_ROOTDIR: /usr/local/share/moses/mosesdecoder/scripts
Use of uninitialized value $extractorargs in concatenation (.) or
string at
/usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
line 370.
filtering the phrase tables... Wed May 16 20:40:51 EEST 2012
exec:
/usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-input.pl
-Binarizer "/usr/local/share/moses/mosesdecoder/dist/bin/CreateOnDisk 1 1 5
100 2" --Hierarchical ./filtered
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/moses.ini
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et
Executing:
/usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-input.pl
-Binarizer "/usr/local/share/moses/mosesdecoder/dist/bin/CreateOnDisk 1 1 5
100 2" --Hierarchical ./filtered
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/moses.ini
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et >
filterphrases.out 2>
filterphrases.err
Asking moses for feature names and values from filtered/moses.ini
Executing: /usr/local/share/moses/mosesdecoder/dist/bin/moses -config
filtered/moses.ini -inputtype 0 -show-weights > ./features.list
Defined parameters (per moses.ini or switch):
config: filtered/moses.ini
cube-pruning-pop-limit: 1000
input-factors: 0
inputtype: 0
lmodel-file: 8 0 3
/home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en
mapping: 0 T 0 1 T 1
max-chart-span: 20 1000
non-terminals: X
search-algorithm: 3
show-weights:
ttable-file: 2 0 0 5
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
6 0 0 1
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
ttable-limit: 20
weight-l: 0.5000
weight-t: 0.20 0.20 0.20 0.20 0.20 1.0
weight-w: -1
Loading lexical distortion models...have 0 models
Start loading LanguageModel
/home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en : [0.000]
seconds
Finished loading LanguageModels : [0.000] seconds
Using uniform ttable-limit of 20 for all translation tables.
Start loading PhraseTable
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
: [0.000]
seconds
filePath:
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
Start loading PhraseTable
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
: [0.000]
seconds
filePath:
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
Finished loading phrase tables : [0.000] seconds
Start loading phrase table from
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
: [0.000]
seconds
Start loading new format pt model : [0.000] seconds
Finished loading phrase tables : [0.000] seconds
MERT starting values and ranges for random generation:
lm = 0.500 ( 0.00 .. 1.00)
w = -1.000 ( 0.00 .. 1.00)
tm = 0.200 ( 0.00 .. 1.00)
tm = 0.200 ( 0.00 .. 1.00)
tm = 0.200 ( 0.00 .. 1.00)
tm = 0.200 ( 0.00 .. 1.00)
tm = 0.200 ( 0.00 .. 1.00)
tm = 1.000 ( 0.00 .. 1.00)
run 1 start at Wed May 16 20:47:37 EEST 2012
Parsing --decoder-flags: ||
Saving new config to: ./run1.moses.ini
Saved: ./run1.moses.ini
Normalizing lambdas: 0.500000 -1.000000 0.200000 0.200000 0.200000
0.200000 0.200000 1.000000
DECODER_CFG = -w -0.285714 -lm 0.142857 -tm 0.057143 0.057143 0.057143
0.057143 0.057143 0.285714
Executing: /usr/local/share/moses/mosesdecoder/dist/bin/moses
-config filtered/moses.ini -inputtype 0 -w -0.285714 -lm 0.142857 -tm
0.057143 0.057143 0.057143 0.057143 0.057143 0.285714 -n-best-list
run1.best100.out 100 -input-file
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et >
run1.out
(1) run decoder to produce n-best lists
params =
decoder_config = -w -0.285714 -lm 0.142857 -tm 0.057143 0.057143
0.057143 0.057143 0.057143 0.285714
Defined parameters (per moses.ini or switch):
config: filtered/moses.ini
cube-pruning-pop-limit: 1000
input-factors: 0
input-file:
/home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et
inputtype: 0
lmodel-file: 8 0 3
/home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en
mapping: 0 T 0 1 T 1
max-chart-span: 20 1000
n-best-list: run1.best100.out 100
non-terminals: X
search-algorithm: 3
ttable-file: 2 0 0 5
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
6 0 0 1
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
ttable-limit: 20
weight-l: 0.142857
weight-t: 0.057143 0.057143 0.057143 0.057143 0.057143 0.285714
weight-w: -0.285714
Loading lexical distortion models...have 0 models
Start loading LanguageModel
/home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en : [0.000]
seconds
Finished loading LanguageModels : [0.000] seconds
Using uniform ttable-limit of 20 for all translation tables.
Start loading PhraseTable
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
: [0.000]
seconds
filePath:
/home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filtered/phrase-table.0-0.1.1.bin
Start loading PhraseTable
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
: [0.000]
seconds
filePath:
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
Finished loading phrase tables : [0.000] seconds
Start loading phrase table from
/home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/glue-grammar
: [0.000]
seconds
Start loading new format pt model : [0.000] seconds
Finished loading phrase tables : [0.000] seconds
Created input-output object : [0.000] seconds
Translating line 0 in thread id 0
ERROR:ERROR: search. Aborting
Aborted
Exit code: 134
The decoder died. CONFIG WAS -w -0.285714 -lm 0.142857 -tm 0.057143
0.057143 0.057143 0.057143 0.057143 0.285714
Any further ideas, what could be wrong?
Kairit
Barry Haddow <[email protected]> kirjutas:
> Hi Kairit
>
> This is the problem:
>
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filte
>> red/phrase-table.0-0.1.1.bin Can't exec "CreateOnDiskPt": No such file or
>> directory at
>
> Either CreateOnDIskPt does not exist, or you've provided an
> incorrect path for
> it.
>
> Not sure why the mert script didn't notice that filtering fails...
>
> cheers - Barry
>
>
> On Tuesday 15 May 2012 13:25:52 [email protected] wrote:
>> Hi,
>>
>> The phrase tables have data and also the paths to the tuning sets are
>> correct. However, filtering process seems to produce errors:
>>
>> The file mert-work/filterphrases.err has the content:
>>
>> Executing: mkdir -p
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filter
>> ed Considering factor 0
>> Done.
>> filtering
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/rule
>> -table.gz ->
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filte
>> red/phrase-table.0-0.1.1... binarizing...CreateOnDiskPt 1 1 5 100 2
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filter
>> ed/phrase-table.0-0.1.1
>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/filte
>> red/phrase-table.0-0.1.1.bin Can't exec "CreateOnDiskPt": No such file or
>> directory at
>> /usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-inp
>> ut.pl line 278, <PIPE> line 3721058.
>>
>> Kairit
>>
>> Hieu Hoang <[email protected]> kirjutas:
>>> hi Kairit
>>>
>>> The phrase table
>>> ---/filtered/phrase-table.0-0.1.1.gz
>>> seems to have loaded in 0 seconds.
>>>
>>> This is the filtered phrase table. Check that the original phrase
>>> table has data, and that the tuning set has data.
>>>
>>> hieu
>>>
>>> (ps. please subscribe to the mailing list to post emails)
>>>
>>> On 09/05/2012 09:23, [email protected] wrote:
>>>> Hi,
>>>>
>>>> I'm trying to tune a hierarchical model and the process is aborted
>>>> with Exit code 134 and the message that the decoder died.
>>>>
>>>> The command I'm running is:
>>>>
>>>> nohup nice
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
>>>> ../../parallel/europarl-v6.et-en.dev.et
>>>> ../../parallel/europarl-v6.et-en.dev.en
>>>> /usr/local/share/moses/mosesdecoder/dist/bin/moses
>>>> train/model/moses.ini --mertdir
>>>> /usr/local/share/moses/mosesdecoder/dist/bin/ --filtercmd
>>>> '/usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given
>>>> -input.pl --Hierarchical'&>
>>>> mert.out
>>>>
>>>> I have also tried this one and it doesn't work either:
>>>>
>>>> nohup nice
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
>>>> ../../parallel/europarl-v6.et-en.dev.et
>>>> ../../parallel/europarl-v6.et-en.dev.en
>>>> /usr/local/share/moses/mosesdecoder/dist/bin/moses
>>>> train/model/moses.ini --mertdir
>>>> /usr/local/share/moses/mosesdecoder/dist/bin/ --filtercmd
>>>> '/usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given
>>>> -input.pl -Binarizer "CreateOnDiskPt 1 1 5 100 2" --Hierarchical'&>
>>>> mert.out
>>>>
>>>> The content of moses.ini file is:
>>>>
>>>> #########################
>>>> ### MOSES CONFIG FILE ###
>>>> #########################
>>>>
>>>> # input factors
>>>> [input-factors]
>>>>
>>>>
>>>> # 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]
>>>> 6 0 0 5
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/r
>>>> ule-table.gz 6 0 0 1
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar
>>>>
>>>> # no generation models, no generation-file section
>>>>
>>>> # language models: type(srilm/irstlm), factors, order, file
>>>> [lmodel-file]
>>>> 8 0 3 /home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en
>>>>
>>>>
>>>> # limit on how many phrase translations e for each phrase f are loaded
>>>> # 0 = all elements loaded
>>>> [ttable-limit]
>>>> 20
>>>>
>>>>
>>>> # language model weights
>>>> [weight-l]
>>>> 0.5000
>>>>
>>>>
>>>> # translation model weights
>>>> [weight-t]
>>>> 0.20
>>>> 0.20
>>>> 0.20
>>>> 0.20
>>>> 0.20
>>>> 1.0
>>>>
>>>> # no generation models, no weight-generation section
>>>>
>>>> # word penalty
>>>> [weight-w]
>>>> -1
>>>>
>>>> [cube-pruning-pop-limit]
>>>> 1000
>>>>
>>>> [non-terminals]
>>>> X
>>>>
>>>> [search-algorithm]
>>>> 3
>>>>
>>>> [inputtype]
>>>> 3
>>>>
>>>> [max-chart-span]
>>>> 20
>>>> 1000
>>>>
>>>> The content of mert.out file is:
>>>>
>>>> nohup: ignoring input
>>>> main::create_extractor_script() called too early to check prototype at
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
>>>> line 681.
>>>> Using SCRIPTS_ROOTDIR: /usr/local/share/moses/mosesdecoder/scripts
>>>> Use of uninitialized value $extractorargs in concatenation (.) or
>>>> string at
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/mert-moses.pl
>>>> line 370.
>>>> filtering the phrase tables... Tue May 8 18:02:26 EEST 2012
>>>> exec:
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-
>>>> input.pl --Hierarchical ./filtered
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/m
>>>> oses.ini
>>>> /home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et
>>>> Executing:
>>>> /usr/local/share/moses/mosesdecoder/scripts/training/filter-model-given-
>>>> input.pl --Hierarchical ./filtered
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/m
>>>> oses.ini
>>>> /home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et>
>>>> filterphrases.out 2>
>>>> filterphrases.err
>>>> Asking moses for feature names and values from filtered/moses.ini
>>>> Executing: /usr/local/share/moses/mosesdecoder/dist/bin/moses -config
>>>> filtered/moses.ini -inputtype 0 -show-weights> ./features.list
>>>> Defined parameters (per moses.ini or switch):
>>>> config: filtered/moses.ini
>>>> cube-pruning-pop-limit: 1000
>>>> input-factors: 0
>>>> inputtype: 0
>>>> lmodel-file: 8 0 3
>>>> /home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en
>>>> mapping: 0 T 0 1 T 1
>>>> max-chart-span: 20 1000
>>>> non-terminals: X
>>>> search-algorithm: 3
>>>> show-weights:
>>>> ttable-file: 6 0 0 5
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz 6 0 0 1
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar ttable-limit: 20
>>>> weight-l: 0.5000
>>>> weight-t: 0.20 0.20 0.20 0.20 0.20 1.0
>>>> weight-w: -1
>>>> Loading lexical distortion models...have 0 models
>>>> Start loading LanguageModel
>>>> /home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en : [0.000]
>>>> seconds
>>>> Finished loading LanguageModels : [0.000] seconds
>>>> Using uniform ttable-limit of 20 for all translation tables.
>>>> Start loading PhraseTable
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz : [0.000]
>>>> seconds
>>>> filePath:
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz Start loading PhraseTable
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar : [0.000]
>>>> seconds
>>>> filePath:
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar Finished loading phrase tables : [0.000] seconds
>>>> Start loading phrase table from
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz : [0.000]
>>>> seconds
>>>> Start loading new format pt model : [0.000] seconds
>>>> Finished loading phrase tables : [72.000] seconds
>>>> Start loading phrase table from
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar : [72.000]
>>>> seconds
>>>> Start loading new format pt model : [72.000] seconds
>>>> Finished loading phrase tables : [72.000] seconds
>>>> MERT starting values and ranges for random generation:
>>>> lm = 0.500 ( 0.00 .. 1.00)
>>>> w = -1.000 ( 0.00 .. 1.00)
>>>> tm = 0.200 ( 0.00 .. 1.00)
>>>> tm = 0.200 ( 0.00 .. 1.00)
>>>> tm = 0.200 ( 0.00 .. 1.00)
>>>> tm = 0.200 ( 0.00 .. 1.00)
>>>> tm = 0.200 ( 0.00 .. 1.00)
>>>> tm = 1.000 ( 0.00 .. 1.00)
>>>> run 1 start at Tue May 8 18:05:59 EEST 2012
>>>> Parsing --decoder-flags: ||
>>>> Saving new config to: ./run1.moses.ini
>>>> Saved: ./run1.moses.ini
>>>> Normalizing lambdas: 0.500000 -1.000000 0.200000 0.200000 0.200000
>>>> 0.200000 0.200000 1.000000
>>>> DECODER_CFG = -w -0.285714 -lm 0.142857 -tm 0.057143 0.057143 0.057143
>>>> 0.057143 0.057143 0.285714
>>>> Executing: /usr/local/share/moses/mosesdecoder/dist/bin/moses
>>>> -config filtered/moses.ini -inputtype 0 -w -0.285714 -lm 0.142857 -tm
>>>> 0.057143 0.057143 0.057143 0.057143 0.057143 0.285714 -n-best-list
>>>> run1.best100.out 100 -input-file
>>>> /home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et>
>>>> run1.out
>>>> (1) run decoder to produce n-best lists
>>>> params =
>>>> decoder_config = -w -0.285714 -lm 0.142857 -tm 0.057143 0.057143
>>>> 0.057143 0.057143 0.057143 0.285714
>>>> Defined parameters (per moses.ini or switch):
>>>> config: filtered/moses.ini
>>>> cube-pruning-pop-limit: 1000
>>>> input-factors: 0
>>>> input-file:
>>>> /home/kairit/Documents/korpused/parallel/europarl-v6.et-en.dev.et
>>>> inputtype: 0
>>>> lmodel-file: 8 0 3
>>>> /home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en
>>>> mapping: 0 T 0 1 T 1
>>>> max-chart-span: 20 1000
>>>> n-best-list: run1.best100.out 100
>>>> non-terminals: X
>>>> search-algorithm: 3
>>>> ttable-file: 6 0 0 5
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz 6 0 0 1
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar ttable-limit: 20
>>>> weight-l: 0.142857
>>>> weight-t: 0.057143 0.057143 0.057143 0.057143 0.057143 0.285714
>>>> weight-w: -0.285714
>>>> Loading lexical distortion models...have 0 models
>>>> Start loading LanguageModel
>>>> /home/kairit/Documents/korpused/lm/europarl-v6.et-en.blm.en : [0.000]
>>>> seconds
>>>> Finished loading LanguageModels : [0.000] seconds
>>>> Using uniform ttable-limit of 20 for all translation tables.
>>>> Start loading PhraseTable
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz : [0.000]
>>>> seconds
>>>> filePath:
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz Start loading PhraseTable
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar : [0.000]
>>>> seconds
>>>> filePath:
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar Finished loading phrase tables : [0.000] seconds
>>>> Start loading phrase table from
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/mert-work/fil
>>>> tered/phrase-table.0-0.1.1.gz : [0.000]
>>>> seconds
>>>> Start loading new format pt model : [0.000] seconds
>>>> Finished loading phrase tables : [73.000] seconds
>>>> Start loading phrase table from
>>>> /home/kairit/Documents/korpused/working/hierarchical_et_en/train/model/g
>>>> lue-grammar : [73.000]
>>>> seconds
>>>> Start loading new format pt model : [73.000] seconds
>>>> Finished loading phrase tables : [73.000] seconds
>>>> Created input-output object : [73.000] seconds
>>>> Translating line 0 in thread id 0
>>>> ERROR:ERROR: search. Aborting
>>>>
>>>> Aborted
>>>> Exit code: 134
>>>> The decoder died. CONFIG WAS -w -0.285714 -lm 0.142857 -tm 0.057143
>>>> 0.057143 0.057143 0.057143 0.057143 0.285714
>>>>
>>>>
>>>> Thanks if you can help me to find the problem.
>>>>
>>>> Kairit
>>>>
>>>>
>>>> _______________________________________________
>>>> 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
>>
>
> --
> Barry Haddow
> University of Edinburgh
> +44 (0) 131 651 3173
>
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
