Re: [Moses-support] Issue with tuning - decoder died
doubt it. Look at the script. The problem chars are []|<>&'" On 23/10/2015 13:01, Ya-chi Lin wrote: Hi Hieu, Thank for pointing this out. Could you please let me know if Chinese square brackets 「」are also a problem? Best, Ya-Chi *寄件者:* Hieu Hoang *寄件日期:* 2015年10月23日 下午 02:32 *收件者:* Ya-chi Lin; moses-support@mit.edu *主旨:* Re: [Moses-support] Issue with tuning - decoder died you must escape the [ and ] characters. If you had used the moses tokenizer, it will do this for you. If you use a different tokenizer, then run your tokenized output through scripts/tokenizer/escape-special-chars.perl On 23/10/2015 01:22, Ya-chi Lin wrote: Hi ALL, I'm building an English to Traditional Chinese system but have encountered a problem while performing tuning. Below is the content of mert.out file. Can anyone please tell me what went wrong here? Any help will be greatly appreciated. Using SCRIPTS_ROOTDIR: /opt/mosesdecoder/scripts Assuming the tables are already filtered, reusing filtered/moses.ini Using cached features list: ./features.list MERT starting values and ranges for random generation: LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) Distortion0 = 0.300 ( 0.00 .. 1.00) LM0 = 0.500 ( 0.00 .. 1.00) WordPenalty0 = -1.000 ( 0.00 .. 1.00) PhrasePenalty0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: Distortion0=0.30 featlist: LM0=0.50 featlist: WordPenalty0=-1.00 featlist: PhrasePenalty0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 run 1 start at Fri Oct 23 03:07:31 EEST 2015 Parsing --decoder-flags: || Saving new config to: ./run1.moses.ini Saved: ./run1.moses.ini Normalizing lambdas: 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.50 -1.00 0.20 0.20 0.20 0.20 0.20 DECODER_CFG = -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' Executing: /opt/mosesdecoder/bin/moses -config filtered/moses.ini -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' -n-best-list run1.best100.out 100 distinct -input-file /home/yachilin/working/tune.true.en > run1.out Executing: /opt/mosesdecoder/bin/moses -config filtered/moses.ini -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' -n-best-list run1.best100.out 100 distinct -input-file /home/yachilin/working/tune.true.en > run1.out 1-10.20.2 0.2 0.2 0.20.3 0.3 0.3 0.3 0.3 0.30.30.5(1) run decoder to produce n-best lists params = decoder_config = -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' Defined parameters (per moses.ini or switch): config: filtered/moses.ini distortion-limit: 6 feature: UnknownWordPenalty WordPenalty PhrasePenalty PhraseDictionaryMemory name=TranslationModel0 num-features=4 path=/home/yachilin/working/mert-work/filtered/phrase-table.0-0.1.1.gz input-factor=0 output-factor=0 LexicalReordering name=LexicalReordering0 num-features=6 type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0 path=/home/yachilin/working/mert-work/filtered/reordering-table.wbe-msd-bidirectional-fe Distortion KENLM lazyken=0 name=LM0 factor=0 path=/home/yachilin/lm/train.blm.tch order=3 input-factors: 0 input-file: /home/yachilin/working/tune.true.en mapping: 0 T 0 n-best-list: run1.best100.out 100 distinct weight: UnknownWordPenalty0= 1 WordPenalty0= -1 PhrasePenalty0= 0.2 TranslationModel0= 0.2 0.2 0.2 0.2 LexicalReordering0= 0.3 0.3 0.3
Re: [Moses-support] Issue with tuning - decoder died
you must escape the [ and ] characters. If you had used the moses tokenizer, it will do this for you. If you use a different tokenizer, then run your tokenized output through scripts/tokenizer/escape-special-chars.perl On 23/10/2015 01:22, Ya-chi Lin wrote: Hi ALL, I'm building an English to Traditional Chinese system but have encountered a problem while performing tuning. Below is the content of mert.out file. Can anyone please tell me what went wrong here? Any help will be greatly appreciated. Using SCRIPTS_ROOTDIR: /opt/mosesdecoder/scripts Assuming the tables are already filtered, reusing filtered/moses.ini Using cached features list: ./features.list MERT starting values and ranges for random generation: LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) LexicalReordering0 = 0.300 ( 0.00 .. 1.00) Distortion0 = 0.300 ( 0.00 .. 1.00) LM0 = 0.500 ( 0.00 .. 1.00) WordPenalty0 = -1.000 ( 0.00 .. 1.00) PhrasePenalty0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) TranslationModel0 = 0.200 ( 0.00 .. 1.00) featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: LexicalReordering0=0.30 featlist: Distortion0=0.30 featlist: LM0=0.50 featlist: WordPenalty0=-1.00 featlist: PhrasePenalty0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 featlist: TranslationModel0=0.20 run 1 start at Fri Oct 23 03:07:31 EEST 2015 Parsing --decoder-flags: || Saving new config to: ./run1.moses.ini Saved: ./run1.moses.ini Normalizing lambdas: 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.50 -1.00 0.20 0.20 0.20 0.20 0.20 DECODER_CFG = -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' Executing: /opt/mosesdecoder/bin/moses -config filtered/moses.ini -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' -n-best-list run1.best100.out 100 distinct -input-file /home/yachilin/working/tune.true.en > run1.out Executing: /opt/mosesdecoder/bin/moses -config filtered/moses.ini -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' -n-best-list run1.best100.out 100 distinct -input-file /home/yachilin/working/tune.true.en > run1.out 1-10.20.2 0.2 0.2 0.20.3 0.3 0.3 0.3 0.3 0.30.30.5(1) run decoder to produce n-best lists params = decoder_config = -weight-overwrite 'LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391' Defined parameters (per moses.ini or switch): config: filtered/moses.ini distortion-limit: 6 feature: UnknownWordPenalty WordPenalty PhrasePenalty PhraseDictionaryMemory name=TranslationModel0 num-features=4 path=/home/yachilin/working/mert-work/filtered/phrase-table.0-0.1.1.gz input-factor=0 output-factor=0 LexicalReordering name=LexicalReordering0 num-features=6 type=wbe-msd-bidirectional-fe-allff input-factor=0 output-factor=0 path=/home/yachilin/working/mert-work/filtered/reordering-table.wbe-msd-bidirectional-fe Distortion KENLM lazyken=0 name=LM0 factor=0 path=/home/yachilin/lm/train.blm.tch order=3 input-factors: 0 input-file: /home/yachilin/working/tune.true.en mapping: 0 T 0 n-best-list: run1.best100.out 100 distinct weight: UnknownWordPenalty0= 1 WordPenalty0= -1 PhrasePenalty0= 0.2 TranslationModel0= 0.2 0.2 0.2 0.2 LexicalReordering0= 0.3 0.3 0.3 0.3 0.3 0.3 Distortion0= 0.3 LM0= 0.5 weight-overwrite: LM0= 0.108696 LexicalReordering0= 0.065217 0.065217 0.065217 0.065217 0.065217 0.065217 TranslationModel0= 0.043478 0.043478 0.043478 0.043478 PhrasePenalty0= 0.043478 Distortion0= 0.065217 WordPenalty0= -0.217391 line=UnknownWordPenalty FeatureFunction: UnknownWordPenalty0 start: 0 end: 0 line=WordPenalty FeatureFunction: WordPenalty0 start: 1 end: 1 line=PhrasePenalty FeatureFunction: PhrasePena