thuvh
unread,Nov 16, 2009, 10:10:06 PM11/16/09Sign in to reply to author
Sign in to forward
You do not have permission to delete messages in this group
Sign in to report message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to JosHUa - JHU Open Source Architecture (Technical List)
I've 2 questions:
In folder example of joshua package, i see two configuration file:
example.config.bloomfilterlm and example.config.javalm. could you
please show me how to train data to get these file or can I use ZMERT
to train data to get these file?
The next question is about decode_example_javalm.sh
When I run script, I get exception and this is log of process:
Nov 17, 2009 9:53:06 AM joshua.decoder.JoshuaConfiguration
readConfigFile
INFO: you use a LM feature function, so make sure you have a LM
grammar
Nov 17, 2009 9:53:06 AM joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA
<init>
INFO: use java lm
Nov 17, 2009 9:53:06 AM joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA
read_lm_grammar_from_file
INFO: Reading grammar from file example/example.trigram.lm.gz
Nov 17, 2009 9:53:06 AM joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA
read_lm_grammar_from_file
INFO: begin to read ngrams with order 1
Nov 17, 2009 9:53:06 AM joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA
read_lm_grammar_from_file
INFO: begin to read ngrams with order 2
Nov 17, 2009 9:53:07 AM joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA
read_lm_grammar_from_file
INFO: begin to read ngrams with order 3
Nov 17, 2009 9:53:07 AM joshua.decoder.JoshuaDecoder
initializeGlueGrammar
INFO: Constructing glue grammar...
Nov 17, 2009 9:53:07 AM
joshua.decoder.ff.tm.hiero.MemoryBasedBatchGrammar print_grammar
INFO: ###########Grammar###########
Nov 17, 2009 9:53:07 AM
joshua.decoder.ff.tm.hiero.MemoryBasedBatchGrammar print_grammar
INFO: ####num_rules: 2; num_bins: 2; num_pruned: 0; sumest_cost:
0.00000
Nov 17, 2009 9:53:07 AM joshua.decoder.JoshuaDecoder
initializeTranslationGrammars
INFO: Using grammar read from file example/example.hiero.tm.gz
Nov 17, 2009 9:53:08 AM
joshua.decoder.ff.tm.hiero.MemoryBasedBatchGrammar print_grammar
INFO: ###########Grammar###########
Nov 17, 2009 9:53:08 AM
joshua.decoder.ff.tm.hiero.MemoryBasedBatchGrammar print_grammar
INFO: ####num_rules: 15939; num_bins: 2029; num_pruned: 0;
sumest_cost: 0.00000
Exception in thread "main" java.lang.ClassCastException:
joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA$LMHash cannot be cast to
java.lang.Double
at joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA.get_valid_prob
(LMGrammarJAVA.java:199)
at
joshua.decoder.ff.lm.buildin_lm.LMGrammarJAVA.ngramLogProbability_helper
(LMGrammarJAVA.java:165)
at joshua.decoder.ff.lm.AbstractLM.ngramLogProbability
(AbstractLM.java:68)
at
joshua.decoder.ff.lm.DefaultNGramLanguageModel.sentenceLogProbability
(DefaultNGramLanguageModel.java:82)
at joshua.decoder.ff.lm.AbstractLM.sentenceLogProbability
(AbstractLM.java:45)
at joshua.decoder.ff.lm.LanguageModelFF.score_chunk
(LanguageModelFF.java:279)
at joshua.decoder.ff.lm.LanguageModelFF.estimate_rule_prob
(LanguageModelFF.java:314)
at joshua.decoder.ff.lm.LanguageModelFF.estimate
(LanguageModelFF.java:125)
at joshua.decoder.ff.tm.MonolingualRule.estimateRuleCost
(MonolingualRule.java:194)
at joshua.decoder.ff.tm.BasicRuleCollection.sortRules
(BasicRuleCollection.java:104)
at joshua.decoder.ff.tm.BasicRuleCollection.sortRules
(BasicRuleCollection.java:119)
at joshua.decoder.ff.tm.AbstractGrammar.sort
(AbstractGrammar.java:118)
at joshua.decoder.ff.tm.AbstractGrammar.sort
(AbstractGrammar.java:153)
at joshua.decoder.ff.tm.AbstractGrammar.sortGrammar
(AbstractGrammar.java:70)
at joshua.decoder.JoshuaDecoder.initialize(JoshuaDecoder.java:
304)
at joshua.decoder.JoshuaDecoder.<init>(JoshuaDecoder.java:108)
at joshua.decoder.JoshuaDecoder.main(JoshuaDecoder.java:664)
Nov 17, 2009 9:53:08 AM joshua.decoder.ff.tm.GrammarReader finalize
SEVERE: Grammar file stream was not closed, this indicates a coding
error: example/example.hiero.tm.gz
Nov 17, 2009 9:53:08 AM joshua.decoder.ff.tm.GrammarReader finalize
SEVERE: Grammar file stream was not closed, this indicates a coding
error: grammars/hiero.glue
"javalm.log" 46L, 3488C
Plz, tell me the way to solve it.