From 3629e2c8ec552cd2d7e0cd70dc8d09619dd8c8b5 Mon Sep 17 00:00:00 2001 From: Franck Dary <franck.dary@lis-lab.fr> Date: Wed, 27 Nov 2019 22:28:25 +0100 Subject: [PATCH] Updated tools features and hyperparameters --- .../data/feature_models/morpho_incremental.fm | 8 +- UD_any/data/feature_models/morpho_nofuture.fm | 8 +- UD_any/data/feature_models/parser_nofuture.fm | 8 +- .../data/feature_models/tagger_incremental.fm | 7 + UD_any/data/feature_models/tagger_nofuture.fm | 8 +- UD_any/lemmatizer/lemmatizer.dicts | 2 +- UD_any/tagparser/tagparser.dicts | 8 +- UD_any/tagparser_sequential/tagparser.dicts | 8 +- .../tagparser.dicts | 8 +- UD_any/tokeparser/lemmatizer_case.cla | 2 +- UD_any/tokeparser/lemmatizer_rules.cla | 2 +- UD_any/tokeparser/morpho.cla | 2 +- UD_any/tokeparser/normal.tm | 12 +- UD_any/tokeparser/parser.cla | 2 +- UD_any/tokeparser/segmenter.cla | 4 +- UD_any/tokeparser/tagger.cla | 2 +- UD_any/tokeparser/tokenizer.cla | 2 +- UD_any/tokeparser/tokeparser.dicts | 167 ++++++++--------- .../lemmatizer_case.cla | 2 +- .../lemmatizer_rules.cla | 2 +- UD_any/tokeparser_incremental/morpho.cla | 2 +- UD_any/tokeparser_incremental/parser.cla | 2 +- UD_any/tokeparser_incremental/segmenter.cla | 4 +- UD_any/tokeparser_incremental/tagger.cla | 2 +- UD_any/tokeparser_incremental/tokenizer.cla | 2 +- .../tokeparser_incremental/tokeparser.dicts | 170 ++++++++---------- .../tokeparser_sequential/lemmatizer_case.cla | 2 +- .../lemmatizer_rules.cla | 2 +- UD_any/tokeparser_sequential/morpho.cla | 4 +- UD_any/tokeparser_sequential/parser.cla | 4 +- UD_any/tokeparser_sequential/segmenter.cla | 4 +- UD_any/tokeparser_sequential/tagger.cla | 4 +- UD_any/tokeparser_sequential/tokenizer.cla | 2 +- UD_any/tokeparser_sequential/tokeparser.dicts | 167 ++++++++--------- 34 files changed, 322 insertions(+), 313 deletions(-) diff --git a/UD_any/data/feature_models/morpho_incremental.fm b/UD_any/data/feature_models/morpho_incremental.fm index 2086605..2e0ecc9 100644 --- a/UD_any/data/feature_models/morpho_incremental.fm +++ b/UD_any/data/feature_models/morpho_incremental.fm @@ -25,4 +25,10 @@ b.0#FORM.PART.0.0 b.0#FORM.PART.1.1 b.0#FORM.PART.2.2 b.0#FORM.PART.3.3 - +# RAW INPUT +raw.0 +raw.2 +raw.3 +raw.4 +raw.5 +raw.6 diff --git a/UD_any/data/feature_models/morpho_nofuture.fm b/UD_any/data/feature_models/morpho_nofuture.fm index 2086605..2e0ecc9 100644 --- a/UD_any/data/feature_models/morpho_nofuture.fm +++ b/UD_any/data/feature_models/morpho_nofuture.fm @@ -25,4 +25,10 @@ b.0#FORM.PART.0.0 b.0#FORM.PART.1.1 b.0#FORM.PART.2.2 b.0#FORM.PART.3.3 - +# RAW INPUT +raw.0 +raw.2 +raw.3 +raw.4 +raw.5 +raw.6 diff --git a/UD_any/data/feature_models/parser_nofuture.fm b/UD_any/data/feature_models/parser_nofuture.fm index d328787..e72a697 100644 --- a/UD_any/data/feature_models/parser_nofuture.fm +++ b/UD_any/data/feature_models/parser_nofuture.fm @@ -101,4 +101,10 @@ tc.1 tc.2 tc.3 tc.4 - +# RAW INPUT +raw.0 +raw.2 +raw.3 +raw.4 +raw.5 +raw.6 diff --git a/UD_any/data/feature_models/tagger_incremental.fm b/UD_any/data/feature_models/tagger_incremental.fm index 1bd749b..50ff0fc 100644 --- a/UD_any/data/feature_models/tagger_incremental.fm +++ b/UD_any/data/feature_models/tagger_incremental.fm @@ -23,3 +23,10 @@ b.0#FORM.PART.3.3 # MORPHO b.-2#MORPHO b.-1#MORPHO +# RAW INPUT +raw.0 +raw.2 +raw.3 +raw.4 +raw.5 +raw.6 diff --git a/UD_any/data/feature_models/tagger_nofuture.fm b/UD_any/data/feature_models/tagger_nofuture.fm index 062f955..98230a0 100644 --- a/UD_any/data/feature_models/tagger_nofuture.fm +++ b/UD_any/data/feature_models/tagger_nofuture.fm @@ -20,4 +20,10 @@ b.0#FORM.PART.0.0 b.0#FORM.PART.1.1 b.0#FORM.PART.2.2 b.0#FORM.PART.3.3 - +# RAW INPUT +raw.0 +raw.2 +raw.3 +raw.4 +raw.5 +raw.6 diff --git a/UD_any/lemmatizer/lemmatizer.dicts b/UD_any/lemmatizer/lemmatizer.dicts index 1bafa4c..f29e3ca 100644 --- a/UD_any/lemmatizer/lemmatizer.dicts +++ b/UD_any/lemmatizer/lemmatizer.dicts @@ -6,6 +6,6 @@ Lemmatizer_Rules_pos 30 Embeddings _ 25 Lemmatizer_Rules_morpho 30 Embeddings _ 50000 ############################################# Lemmatizer_Case_form 30 Embeddings _ 50000 -Lemmatizer_Case_letters 10 Embeddings _ 20000 +Lemmatizer_Case_letters 10 Embeddings _ 200000 Lemmatizer_Case_pos 30 Embeddings _ 25 Lemmatizer_Case_morpho 30 Embeddings _ 50000 diff --git a/UD_any/tagparser/tagparser.dicts b/UD_any/tagparser/tagparser.dicts index 9672e00..f428302 100644 --- a/UD_any/tagparser/tagparser.dicts +++ b/UD_any/tagparser/tagparser.dicts @@ -21,9 +21,9 @@ Morpho_eos 16 Embeddings _ 5 Morpho_gov 16 Embeddings _ 100 Morpho_pos 18 Embeddings _ 21 Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 +Morpho_form.f 30 Embeddings _ 200000 Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 +Morpho_letters 30 Embeddings _ 200000 Morpho_labels 18 Embeddings _ 50 Morpho_morpho 22 Embeddings _ 50000 # LEMMATIZER @@ -44,9 +44,9 @@ Parser_eos 16 Embeddings _ 5 Parser_gov 16 Embeddings _ 100 Parser_pos 18 Embeddings _ 21 Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 +Parser_form.f 30 Embeddings _ 200000 Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 +Parser_letters 30 Embeddings _ 200000 Parser_labels 18 Embeddings _ 50 Parser_morpho 22 Embeddings _ 50000 # SEGMENTER diff --git a/UD_any/tagparser_sequential/tagparser.dicts b/UD_any/tagparser_sequential/tagparser.dicts index 9672e00..f428302 100644 --- a/UD_any/tagparser_sequential/tagparser.dicts +++ b/UD_any/tagparser_sequential/tagparser.dicts @@ -21,9 +21,9 @@ Morpho_eos 16 Embeddings _ 5 Morpho_gov 16 Embeddings _ 100 Morpho_pos 18 Embeddings _ 21 Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 +Morpho_form.f 30 Embeddings _ 200000 Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 +Morpho_letters 30 Embeddings _ 200000 Morpho_labels 18 Embeddings _ 50 Morpho_morpho 22 Embeddings _ 50000 # LEMMATIZER @@ -44,9 +44,9 @@ Parser_eos 16 Embeddings _ 5 Parser_gov 16 Embeddings _ 100 Parser_pos 18 Embeddings _ 21 Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 +Parser_form.f 30 Embeddings _ 200000 Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 +Parser_letters 30 Embeddings _ 200000 Parser_labels 18 Embeddings _ 50 Parser_morpho 22 Embeddings _ 50000 # SEGMENTER diff --git a/UD_any/tagparser_sequential_strong/tagparser.dicts b/UD_any/tagparser_sequential_strong/tagparser.dicts index 9672e00..f428302 100644 --- a/UD_any/tagparser_sequential_strong/tagparser.dicts +++ b/UD_any/tagparser_sequential_strong/tagparser.dicts @@ -21,9 +21,9 @@ Morpho_eos 16 Embeddings _ 5 Morpho_gov 16 Embeddings _ 100 Morpho_pos 18 Embeddings _ 21 Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 +Morpho_form.f 30 Embeddings _ 200000 Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 +Morpho_letters 30 Embeddings _ 200000 Morpho_labels 18 Embeddings _ 50 Morpho_morpho 22 Embeddings _ 50000 # LEMMATIZER @@ -44,9 +44,9 @@ Parser_eos 16 Embeddings _ 5 Parser_gov 16 Embeddings _ 100 Parser_pos 18 Embeddings _ 21 Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 +Parser_form.f 30 Embeddings _ 200000 Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 +Parser_letters 30 Embeddings _ 200000 Parser_labels 18 Embeddings _ 50 Parser_morpho 22 Embeddings _ 50000 # SEGMENTER diff --git a/UD_any/tokeparser/lemmatizer_case.cla b/UD_any/tokeparser/lemmatizer_case.cla index 0994d40..5ea11d7 100644 --- a/UD_any/tokeparser/lemmatizer_case.cla +++ b/UD_any/tokeparser/lemmatizer_case.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_case Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_case.as -Topology : (100,RELU,0.1) +Topology : (500,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/lemmatizer_rules.cla b/UD_any/tokeparser/lemmatizer_rules.cla index 9b59b62..97ae8f5 100644 --- a/UD_any/tokeparser/lemmatizer_rules.cla +++ b/UD_any/tokeparser/lemmatizer_rules.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_rules Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_rules.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/morpho.cla b/UD_any/tokeparser/morpho.cla index 024d1e9..d55f862 100644 --- a/UD_any/tokeparser/morpho.cla +++ b/UD_any/tokeparser/morpho.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : morpho Feature Model : data/feature_models/morpho_nofuture.fm Action Set : data/morpho_parts.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/normal.tm b/UD_any/tokeparser/normal.tm index 77ebff3..f591178 100644 --- a/UD_any/tokeparser/normal.tm +++ b/UD_any/tokeparser/normal.tm @@ -1,31 +1,39 @@ -Name : Tagger, Morpho, Lemmatizer and Parser Machine -Dicts : tagparser.dicts +Name : Tokenizer, Tagger, Morpho, Lemmatizer and Parser Machine +Dicts : tokeparser.dicts %CLASSIFIERS strategy strategy.cla +tokenizer tokenizer.cla tagger tagger.cla morpho morpho.cla lemmatizer_lookup lemmatizer_lookup.cla lemmatizer_rules lemmatizer_rules.cla lemmatizer_case lemmatizer_case.cla parser parser.cla +segmenter segmenter.cla %STATES strategy strategy +tokenizer tokenizer tagger tagger morpho morpho lemmatizer_lookup lemmatizer_lookup lemmatizer_rules lemmatizer_rules lemmatizer_case lemmatizer_case parser parser +segmenter segmenter %TRANSITIONS +strategy tokenizer MOVE tokenizer strategy tagger MOVE tagger strategy morpho MOVE morpho strategy lemmatizer_lookup MOVE lemmatizer_lookup strategy lemmatizer_rules MOVE lemmatizer_rules strategy lemmatizer_case MOVE lemmatizer_case strategy parser MOVE parser +strategy segmenter MOVE segmenter +tokenizer strategy * tagger strategy * morpho strategy * lemmatizer_lookup strategy * lemmatizer_case strategy * lemmatizer_rules strategy * parser strategy * +segmenter strategy * diff --git a/UD_any/tokeparser/parser.cla b/UD_any/tokeparser/parser.cla index 77714be..f54ffe9 100644 --- a/UD_any/tokeparser/parser.cla +++ b/UD_any/tokeparser/parser.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : parser Feature Model : data/feature_models/parser_nofuture.fm Action Set : data/parser.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/segmenter.cla b/UD_any/tokeparser/segmenter.cla index 6a817d1..67b2394 100644 --- a/UD_any/tokeparser/segmenter.cla +++ b/UD_any/tokeparser/segmenter.cla @@ -1,7 +1,7 @@ Name : Segmenter Type : Prediction Oracle : segmenter -Feature Model : data/feature_models/tagger.fm +Feature Model : data/feature_models/parser_nofuture.fm Action Set : data/segmenter.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/tagger.cla b/UD_any/tokeparser/tagger.cla index c98cb92..75c785a 100644 --- a/UD_any/tokeparser/tagger.cla +++ b/UD_any/tokeparser/tagger.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : tagger Feature Model : data/feature_models/tagger_nofuture.fm Action Set : data/tagger.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser/tokenizer.cla b/UD_any/tokeparser/tokenizer.cla index cdd5a2e..febcc49 100644 --- a/UD_any/tokeparser/tokenizer.cla +++ b/UD_any/tokeparser/tokenizer.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : tokenizer Feature Model : data/feature_models/tokenizer.fm Action Set : data/tokenizer.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : no diff --git a/UD_any/tokeparser/tokeparser.dicts b/UD_any/tokeparser/tokeparser.dicts index c9235b9..9d84402 100644 --- a/UD_any/tokeparser/tokeparser.dicts +++ b/UD_any/tokeparser/tokeparser.dicts @@ -1,102 +1,93 @@ #Name Dimension Mode # ############################ # TOKENIZER -Tokenizer_bool 02 Embeddings _ 5 -Tokenizer_int 05 Embeddings _ 200 -Tokenizer_letters 30 Embeddings _ 200000 -Tokenizer_form 30 Embeddings _ 50000 -Tokenizer_form.f 30 Embeddings _ 200000 -Tokenizer_actions 05 Embeddings _ 200 +Tokenizer_actions 18 Embeddings _ 200 +Tokenizer_bool 16 Embeddings _ 20 +Tokenizer_int 16 Embeddings _ 200 +Tokenizer_eos 16 Embeddings _ 20 +Tokenizer_gov 16 Embeddings _ 100 +Tokenizer_pos 18 Embeddings _ 30 +Tokenizer_form 80 Embeddings _ 20 +Tokenizer_form.f 80 Embeddings _ 200000 +Tokenizer_lemma 80 Embeddings _ 20 +Tokenizer_letters 80 Embeddings _ 5000 +Tokenizer_labels 18 Embeddings _ 200 +Tokenizer_morpho 22 Embeddings _ 2000 # TAGGER -Tagger_actions 18 Embeddings _ 20 -Tagger_bool 16 Embeddings _ 5 +Tagger_actions 18 Embeddings _ 200 +Tagger_bool 16 Embeddings _ 20 Tagger_int 16 Embeddings _ 200 -Tagger_eos 16 Embeddings _ 5 +Tagger_eos 16 Embeddings _ 20 Tagger_gov 16 Embeddings _ 100 -Tagger_pos 18 Embeddings _ 21 -Tagger_form 30 Embeddings _ 50000 -Tagger_form.f 30 Embeddings _ 200000 -Tagger_lemma 30 Embeddings _ 50000 -Tagger_letters 30 Embeddings _ 200000 -Tagger_labels 18 Embeddings _ 50 -Tagger_morpho 22 Embeddings _ 50000 +Tagger_pos 18 Embeddings _ 30 +Tagger_form 80 Embeddings _ 20 +Tagger_form.f 80 Embeddings _ 200000 +Tagger_lemma 80 Embeddings _ 20 +Tagger_letters 80 Embeddings _ 5000 +Tagger_labels 18 Embeddings _ 200 +Tagger_morpho 22 Embeddings _ 2000 # MORPHO -Morpho_actions 18 Embeddings _ 1000 -Morpho_bool 16 Embeddings _ 5 +Morpho_actions 18 Embeddings _ 200 +Morpho_bool 16 Embeddings _ 20 Morpho_int 16 Embeddings _ 200 -Morpho_eos 16 Embeddings _ 5 +Morpho_eos 16 Embeddings _ 20 Morpho_gov 16 Embeddings _ 100 -Morpho_pos 18 Embeddings _ 21 -Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 -Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 -Morpho_labels 18 Embeddings _ 50 -Morpho_morpho 22 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Rules_form 30 Embeddings _ 50000 -Lemmatizer_Rules_letters 10 Embeddings _ 200000 -Lemmatizer_Rules_pos 30 Embeddings _ 21 -Lemmatizer_Rules_morpho 30 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Case_form 30 Embeddings _ 50000 -Lemmatizer_Case_letters 10 Embeddings _ 200000 -Lemmatizer_Case_pos 30 Embeddings _ 21 -Lemmatizer_Case_morpho 30 Embeddings _ 50000 +Morpho_pos 18 Embeddings _ 30 +Morpho_form 80 Embeddings _ 20 +Morpho_form.f 80 Embeddings _ 200000 +Morpho_lemma 80 Embeddings _ 20 +Morpho_letters 80 Embeddings _ 5000 +Morpho_labels 18 Embeddings _ 200 +Morpho_morpho 22 Embeddings _ 2000 +# LEMMATIZER_RULE +Lemmatizer_Rules_actions 18 Embeddings _ 2000 +Lemmatizer_Rules_bool 16 Embeddings _ 20 +Lemmatizer_Rules_int 16 Embeddings _ 200 +Lemmatizer_Rules_eos 16 Embeddings _ 20 +Lemmatizer_Rules_gov 16 Embeddings _ 20 +Lemmatizer_Rules_pos 18 Embeddings _ 30 +Lemmatizer_Rules_form 80 Embeddings _ 100000 +Lemmatizer_Rules_form.f 80 Embeddings _ 20 +Lemmatizer_Rules_lemma 80 Embeddings _ 20 +Lemmatizer_Rules_letters 80 Embeddings _ 5000 +Lemmatizer_Rules_labels 18 Embeddings _ 20 +Lemmatizer_Rules_morpho 22 Embeddings _ 2000 +# LEMMATIZER_CASE +Lemmatizer_Case_actions 18 Embeddings _ 2000 +Lemmatizer_Case_bool 16 Embeddings _ 20 +Lemmatizer_Case_int 16 Embeddings _ 200 +Lemmatizer_Case_eos 16 Embeddings _ 20 +Lemmatizer_Case_gov 16 Embeddings _ 20 +Lemmatizer_Case_pos 18 Embeddings _ 30 +Lemmatizer_Case_form 80 Embeddings _ 100000 +Lemmatizer_Case_form.f 80 Embeddings _ 20 +Lemmatizer_Case_lemma 80 Embeddings _ 20 +Lemmatizer_Case_letters 80 Embeddings _ 5000 +Lemmatizer_Case_labels 18 Embeddings _ 20 +Lemmatizer_Case_morpho 22 Embeddings _ 2000 # PARSER Parser_actions 18 Embeddings _ 200 -Parser_bool 16 Embeddings _ 5 +Parser_bool 16 Embeddings _ 20 Parser_int 16 Embeddings _ 200 -Parser_eos 16 Embeddings _ 5 +Parser_eos 16 Embeddings _ 20 Parser_gov 16 Embeddings _ 100 -Parser_pos 18 Embeddings _ 21 -Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 -Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 -Parser_labels 18 Embeddings _ 50 -Parser_morpho 22 Embeddings _ 50000 +Parser_pos 18 Embeddings _ 30 +Parser_form 80 Embeddings _ 20 +Parser_form.f 80 Embeddings _ 200000 +Parser_lemma 80 Embeddings _ 20 +Parser_letters 80 Embeddings _ 5000 +Parser_labels 18 Embeddings _ 200 +Parser_morpho 22 Embeddings _ 2000 # SEGMENTER -Segmenter_bool 02 Embeddings _ 5 -Segmenter_int 05 Embeddings _ 200 -Segmenter_letters 30 Embeddings _ 200000 -Segmenter_pos 15 Embeddings _ 21 -Segmenter_form 30 Embeddings _ 50000 -Segmenter_form.f 30 Embeddings _ 200000 -Segmenter_actions 05 Embeddings _ 21 -# ERROR_TAGGER -Error_Tagger_actions 18 Embeddings _ -Error_Tagger_bool 16 Embeddings _ -Error_Tagger_int 16 Embeddings _ -Error_Tagger_eos 16 Embeddings _ -Error_Tagger_gov 16 Embeddings _ -Error_Tagger_pos 18 Embeddings _ -Error_Tagger_form 30 Embeddings _ -Error_Tagger_lemma 30 Embeddings _ -Error_Tagger_letters 30 Embeddings _ -Error_Tagger_labels 18 Embeddings _ -Error_Tagger_morpho 22 Embeddings _ -# ERROR_MORPHO -Error_Morpho_actions 18 Embeddings _ -Error_Morpho_bool 16 Embeddings _ -Error_Morpho_int 16 Embeddings _ -Error_Morpho_eos 16 Embeddings _ -Error_Morpho_gov 16 Embeddings _ -Error_Morpho_pos 18 Embeddings _ -Error_Morpho_form 30 Embeddings _ -Error_Morpho_lemma 30 Embeddings _ -Error_Morpho_letters 30 Embeddings _ -Error_Morpho_labels 18 Embeddings _ -Error_Morpho_morpho 22 Embeddings _ -# ERROR_PARSER -Error_Parser_actions 18 Embeddings _ -Error_Parser_bool 16 Embeddings _ -Error_Parser_int 16 Embeddings _ -Error_Parser_eos 16 Embeddings _ -Error_Parser_gov 16 Embeddings _ -Error_Parser_pos 18 Embeddings _ -Error_Parser_form 30 Embeddings _ -Error_Parser_lemma 30 Embeddings _ -Error_Parser_letters 30 Embeddings _ -Error_Parser_labels 18 Embeddings _ -Error_Parser_morpho 22 Embeddings _ +Segmenter_actions 18 Embeddings _ 200 +Segmenter_bool 16 Embeddings _ 20 +Segmenter_int 16 Embeddings _ 200 +Segmenter_eos 16 Embeddings _ 20 +Segmenter_gov 16 Embeddings _ 100 +Segmenter_pos 18 Embeddings _ 30 +Segmenter_form 80 Embeddings _ 20 +Segmenter_form.f 80 Embeddings _ 200000 +Segmenter_lemma 80 Embeddings _ 20 +Segmenter_letters 80 Embeddings _ 5000 +Segmenter_labels 18 Embeddings _ 200 +Segmenter_morpho 22 Embeddings _ 2000 diff --git a/UD_any/tokeparser_incremental/lemmatizer_case.cla b/UD_any/tokeparser_incremental/lemmatizer_case.cla index 0994d40..912296a 100644 --- a/UD_any/tokeparser_incremental/lemmatizer_case.cla +++ b/UD_any/tokeparser_incremental/lemmatizer_case.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_case Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_case.as -Topology : (100,RELU,0.1) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/lemmatizer_rules.cla b/UD_any/tokeparser_incremental/lemmatizer_rules.cla index 9b59b62..97ae8f5 100644 --- a/UD_any/tokeparser_incremental/lemmatizer_rules.cla +++ b/UD_any/tokeparser_incremental/lemmatizer_rules.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_rules Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_rules.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/morpho.cla b/UD_any/tokeparser_incremental/morpho.cla index 63340a0..040537d 100644 --- a/UD_any/tokeparser_incremental/morpho.cla +++ b/UD_any/tokeparser_incremental/morpho.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : morpho Feature Model : data/feature_models/morpho_incremental.fm Action Set : data/morpho_parts.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/parser.cla b/UD_any/tokeparser_incremental/parser.cla index 77714be..f54ffe9 100644 --- a/UD_any/tokeparser_incremental/parser.cla +++ b/UD_any/tokeparser_incremental/parser.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : parser Feature Model : data/feature_models/parser_nofuture.fm Action Set : data/parser.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/segmenter.cla b/UD_any/tokeparser_incremental/segmenter.cla index 6a817d1..67b2394 100644 --- a/UD_any/tokeparser_incremental/segmenter.cla +++ b/UD_any/tokeparser_incremental/segmenter.cla @@ -1,7 +1,7 @@ Name : Segmenter Type : Prediction Oracle : segmenter -Feature Model : data/feature_models/tagger.fm +Feature Model : data/feature_models/parser_nofuture.fm Action Set : data/segmenter.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/tagger.cla b/UD_any/tokeparser_incremental/tagger.cla index 9aa35ed..0d1b83a 100644 --- a/UD_any/tokeparser_incremental/tagger.cla +++ b/UD_any/tokeparser_incremental/tagger.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : tagger Feature Model : data/feature_models/tagger_incremental.fm Action Set : data/tagger.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_incremental/tokenizer.cla b/UD_any/tokeparser_incremental/tokenizer.cla index 127a62e..9dccbfe 100644 --- a/UD_any/tokeparser_incremental/tokenizer.cla +++ b/UD_any/tokeparser_incremental/tokenizer.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : tokenizer Feature Model : data/feature_models/tokenizer_incremental.fm Action Set : data/tokenizer.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : no diff --git a/UD_any/tokeparser_incremental/tokeparser.dicts b/UD_any/tokeparser_incremental/tokeparser.dicts index fd78502..9d84402 100644 --- a/UD_any/tokeparser_incremental/tokeparser.dicts +++ b/UD_any/tokeparser_incremental/tokeparser.dicts @@ -1,105 +1,93 @@ #Name Dimension Mode # ############################ # TOKENIZER -Tokenizer_bool 02 Embeddings _ 5 -Tokenizer_int 05 Embeddings _ 200 -Tokenizer_letters 30 Embeddings _ 200000 -Tokenizer_form 30 Embeddings _ 50000 -Tokenizer_form.f 30 Embeddings _ 200000 -Tokenizer_actions 05 Embeddings _ 200 -Tokenizer_pos 05 Embeddings _ 25 -Tokenizer_morpho 05 Embeddings _ 50000 -Tokenizer_eos 05 Embeddings _ 5 +Tokenizer_actions 18 Embeddings _ 200 +Tokenizer_bool 16 Embeddings _ 20 +Tokenizer_int 16 Embeddings _ 200 +Tokenizer_eos 16 Embeddings _ 20 +Tokenizer_gov 16 Embeddings _ 100 +Tokenizer_pos 18 Embeddings _ 30 +Tokenizer_form 80 Embeddings _ 20 +Tokenizer_form.f 80 Embeddings _ 200000 +Tokenizer_lemma 80 Embeddings _ 20 +Tokenizer_letters 80 Embeddings _ 5000 +Tokenizer_labels 18 Embeddings _ 200 +Tokenizer_morpho 22 Embeddings _ 2000 # TAGGER -Tagger_actions 18 Embeddings _ 20 -Tagger_bool 16 Embeddings _ 5 +Tagger_actions 18 Embeddings _ 200 +Tagger_bool 16 Embeddings _ 20 Tagger_int 16 Embeddings _ 200 -Tagger_eos 16 Embeddings _ 5 +Tagger_eos 16 Embeddings _ 20 Tagger_gov 16 Embeddings _ 100 -Tagger_pos 18 Embeddings _ 21 -Tagger_form 30 Embeddings _ 50000 -Tagger_form.f 30 Embeddings _ 200000 -Tagger_lemma 30 Embeddings _ 50000 -Tagger_letters 30 Embeddings _ 200000 -Tagger_labels 18 Embeddings _ 50 -Tagger_morpho 22 Embeddings _ 50000 +Tagger_pos 18 Embeddings _ 30 +Tagger_form 80 Embeddings _ 20 +Tagger_form.f 80 Embeddings _ 200000 +Tagger_lemma 80 Embeddings _ 20 +Tagger_letters 80 Embeddings _ 5000 +Tagger_labels 18 Embeddings _ 200 +Tagger_morpho 22 Embeddings _ 2000 # MORPHO -Morpho_actions 18 Embeddings _ 1000 -Morpho_bool 16 Embeddings _ 5 +Morpho_actions 18 Embeddings _ 200 +Morpho_bool 16 Embeddings _ 20 Morpho_int 16 Embeddings _ 200 -Morpho_eos 16 Embeddings _ 5 +Morpho_eos 16 Embeddings _ 20 Morpho_gov 16 Embeddings _ 100 -Morpho_pos 18 Embeddings _ 21 -Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 -Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 -Morpho_labels 18 Embeddings _ 50 -Morpho_morpho 22 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Rules_form 30 Embeddings _ 50000 -Lemmatizer_Rules_letters 10 Embeddings _ 200000 -Lemmatizer_Rules_pos 30 Embeddings _ 21 -Lemmatizer_Rules_morpho 30 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Case_form 30 Embeddings _ 50000 -Lemmatizer_Case_letters 10 Embeddings _ 200000 -Lemmatizer_Case_pos 30 Embeddings _ 21 -Lemmatizer_Case_morpho 30 Embeddings _ 50000 +Morpho_pos 18 Embeddings _ 30 +Morpho_form 80 Embeddings _ 20 +Morpho_form.f 80 Embeddings _ 200000 +Morpho_lemma 80 Embeddings _ 20 +Morpho_letters 80 Embeddings _ 5000 +Morpho_labels 18 Embeddings _ 200 +Morpho_morpho 22 Embeddings _ 2000 +# LEMMATIZER_RULE +Lemmatizer_Rules_actions 18 Embeddings _ 2000 +Lemmatizer_Rules_bool 16 Embeddings _ 20 +Lemmatizer_Rules_int 16 Embeddings _ 200 +Lemmatizer_Rules_eos 16 Embeddings _ 20 +Lemmatizer_Rules_gov 16 Embeddings _ 20 +Lemmatizer_Rules_pos 18 Embeddings _ 30 +Lemmatizer_Rules_form 80 Embeddings _ 100000 +Lemmatizer_Rules_form.f 80 Embeddings _ 20 +Lemmatizer_Rules_lemma 80 Embeddings _ 20 +Lemmatizer_Rules_letters 80 Embeddings _ 5000 +Lemmatizer_Rules_labels 18 Embeddings _ 20 +Lemmatizer_Rules_morpho 22 Embeddings _ 2000 +# LEMMATIZER_CASE +Lemmatizer_Case_actions 18 Embeddings _ 2000 +Lemmatizer_Case_bool 16 Embeddings _ 20 +Lemmatizer_Case_int 16 Embeddings _ 200 +Lemmatizer_Case_eos 16 Embeddings _ 20 +Lemmatizer_Case_gov 16 Embeddings _ 20 +Lemmatizer_Case_pos 18 Embeddings _ 30 +Lemmatizer_Case_form 80 Embeddings _ 100000 +Lemmatizer_Case_form.f 80 Embeddings _ 20 +Lemmatizer_Case_lemma 80 Embeddings _ 20 +Lemmatizer_Case_letters 80 Embeddings _ 5000 +Lemmatizer_Case_labels 18 Embeddings _ 20 +Lemmatizer_Case_morpho 22 Embeddings _ 2000 # PARSER Parser_actions 18 Embeddings _ 200 -Parser_bool 16 Embeddings _ 5 +Parser_bool 16 Embeddings _ 20 Parser_int 16 Embeddings _ 200 -Parser_eos 16 Embeddings _ 5 +Parser_eos 16 Embeddings _ 20 Parser_gov 16 Embeddings _ 100 -Parser_pos 18 Embeddings _ 21 -Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 -Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 -Parser_labels 18 Embeddings _ 50 -Parser_morpho 22 Embeddings _ 50000 +Parser_pos 18 Embeddings _ 30 +Parser_form 80 Embeddings _ 20 +Parser_form.f 80 Embeddings _ 200000 +Parser_lemma 80 Embeddings _ 20 +Parser_letters 80 Embeddings _ 5000 +Parser_labels 18 Embeddings _ 200 +Parser_morpho 22 Embeddings _ 2000 # SEGMENTER -Segmenter_bool 02 Embeddings _ 5 -Segmenter_int 05 Embeddings _ 200 -Segmenter_letters 30 Embeddings _ 200000 -Segmenter_pos 15 Embeddings _ 21 -Segmenter_form 30 Embeddings _ 50000 -Segmenter_form.f 30 Embeddings _ 200000 -Segmenter_actions 05 Embeddings _ 21 -# ERROR_TAGGER -Error_Tagger_actions 18 Embeddings _ -Error_Tagger_bool 16 Embeddings _ -Error_Tagger_int 16 Embeddings _ -Error_Tagger_eos 16 Embeddings _ -Error_Tagger_gov 16 Embeddings _ -Error_Tagger_pos 18 Embeddings _ -Error_Tagger_form 30 Embeddings _ -Error_Tagger_lemma 30 Embeddings _ -Error_Tagger_letters 30 Embeddings _ -Error_Tagger_labels 18 Embeddings _ -Error_Tagger_morpho 22 Embeddings _ -# ERROR_MORPHO -Error_Morpho_actions 18 Embeddings _ -Error_Morpho_bool 16 Embeddings _ -Error_Morpho_int 16 Embeddings _ -Error_Morpho_eos 16 Embeddings _ -Error_Morpho_gov 16 Embeddings _ -Error_Morpho_pos 18 Embeddings _ -Error_Morpho_form 30 Embeddings _ -Error_Morpho_lemma 30 Embeddings _ -Error_Morpho_letters 30 Embeddings _ -Error_Morpho_labels 18 Embeddings _ -Error_Morpho_morpho 22 Embeddings _ -# ERROR_PARSER -Error_Parser_actions 18 Embeddings _ -Error_Parser_bool 16 Embeddings _ -Error_Parser_int 16 Embeddings _ -Error_Parser_eos 16 Embeddings _ -Error_Parser_gov 16 Embeddings _ -Error_Parser_pos 18 Embeddings _ -Error_Parser_form 30 Embeddings _ -Error_Parser_lemma 30 Embeddings _ -Error_Parser_letters 30 Embeddings _ -Error_Parser_labels 18 Embeddings _ -Error_Parser_morpho 22 Embeddings _ +Segmenter_actions 18 Embeddings _ 200 +Segmenter_bool 16 Embeddings _ 20 +Segmenter_int 16 Embeddings _ 200 +Segmenter_eos 16 Embeddings _ 20 +Segmenter_gov 16 Embeddings _ 100 +Segmenter_pos 18 Embeddings _ 30 +Segmenter_form 80 Embeddings _ 20 +Segmenter_form.f 80 Embeddings _ 200000 +Segmenter_lemma 80 Embeddings _ 20 +Segmenter_letters 80 Embeddings _ 5000 +Segmenter_labels 18 Embeddings _ 200 +Segmenter_morpho 22 Embeddings _ 2000 diff --git a/UD_any/tokeparser_sequential/lemmatizer_case.cla b/UD_any/tokeparser_sequential/lemmatizer_case.cla index 0994d40..912296a 100644 --- a/UD_any/tokeparser_sequential/lemmatizer_case.cla +++ b/UD_any/tokeparser_sequential/lemmatizer_case.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_case Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_case.as -Topology : (100,RELU,0.1) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/lemmatizer_rules.cla b/UD_any/tokeparser_sequential/lemmatizer_rules.cla index 9b59b62..97ae8f5 100644 --- a/UD_any/tokeparser_sequential/lemmatizer_rules.cla +++ b/UD_any/tokeparser_sequential/lemmatizer_rules.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : lemma_rules Feature Model : data/feature_models/lemmatizer_rules.fm Action Set : data/lemmatizer_rules.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/morpho.cla b/UD_any/tokeparser_sequential/morpho.cla index 024d1e9..a8d5211 100644 --- a/UD_any/tokeparser_sequential/morpho.cla +++ b/UD_any/tokeparser_sequential/morpho.cla @@ -1,7 +1,7 @@ Name : Morpho Type : Prediction Oracle : morpho -Feature Model : data/feature_models/morpho_nofuture.fm +Feature Model : data/feature_models/morpho.fm Action Set : data/morpho_parts.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/parser.cla b/UD_any/tokeparser_sequential/parser.cla index 77714be..ab7b82c 100644 --- a/UD_any/tokeparser_sequential/parser.cla +++ b/UD_any/tokeparser_sequential/parser.cla @@ -1,7 +1,7 @@ Name : Parser Type : Prediction Oracle : parser -Feature Model : data/feature_models/parser_nofuture.fm +Feature Model : data/feature_models/parser.fm Action Set : data/parser.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/segmenter.cla b/UD_any/tokeparser_sequential/segmenter.cla index 6a817d1..0057fd3 100644 --- a/UD_any/tokeparser_sequential/segmenter.cla +++ b/UD_any/tokeparser_sequential/segmenter.cla @@ -1,7 +1,7 @@ Name : Segmenter Type : Prediction Oracle : segmenter -Feature Model : data/feature_models/tagger.fm +Feature Model : data/feature_models/parser.fm Action Set : data/segmenter.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/tagger.cla b/UD_any/tokeparser_sequential/tagger.cla index c98cb92..7b1bfb3 100644 --- a/UD_any/tokeparser_sequential/tagger.cla +++ b/UD_any/tokeparser_sequential/tagger.cla @@ -1,7 +1,7 @@ Name : Tagger Type : Prediction Oracle : tagger -Feature Model : data/feature_models/tagger_nofuture.fm +Feature Model : data/feature_models/tagger.fm Action Set : data/tagger.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : yes diff --git a/UD_any/tokeparser_sequential/tokenizer.cla b/UD_any/tokeparser_sequential/tokenizer.cla index cdd5a2e..febcc49 100644 --- a/UD_any/tokeparser_sequential/tokenizer.cla +++ b/UD_any/tokeparser_sequential/tokenizer.cla @@ -3,5 +3,5 @@ Type : Prediction Oracle : tokenizer Feature Model : data/feature_models/tokenizer.fm Action Set : data/tokenizer.as -Topology : (500,RELU,0.3) +Topology : (800,RELU,0.3) Dynamic : no diff --git a/UD_any/tokeparser_sequential/tokeparser.dicts b/UD_any/tokeparser_sequential/tokeparser.dicts index c9235b9..9d84402 100644 --- a/UD_any/tokeparser_sequential/tokeparser.dicts +++ b/UD_any/tokeparser_sequential/tokeparser.dicts @@ -1,102 +1,93 @@ #Name Dimension Mode # ############################ # TOKENIZER -Tokenizer_bool 02 Embeddings _ 5 -Tokenizer_int 05 Embeddings _ 200 -Tokenizer_letters 30 Embeddings _ 200000 -Tokenizer_form 30 Embeddings _ 50000 -Tokenizer_form.f 30 Embeddings _ 200000 -Tokenizer_actions 05 Embeddings _ 200 +Tokenizer_actions 18 Embeddings _ 200 +Tokenizer_bool 16 Embeddings _ 20 +Tokenizer_int 16 Embeddings _ 200 +Tokenizer_eos 16 Embeddings _ 20 +Tokenizer_gov 16 Embeddings _ 100 +Tokenizer_pos 18 Embeddings _ 30 +Tokenizer_form 80 Embeddings _ 20 +Tokenizer_form.f 80 Embeddings _ 200000 +Tokenizer_lemma 80 Embeddings _ 20 +Tokenizer_letters 80 Embeddings _ 5000 +Tokenizer_labels 18 Embeddings _ 200 +Tokenizer_morpho 22 Embeddings _ 2000 # TAGGER -Tagger_actions 18 Embeddings _ 20 -Tagger_bool 16 Embeddings _ 5 +Tagger_actions 18 Embeddings _ 200 +Tagger_bool 16 Embeddings _ 20 Tagger_int 16 Embeddings _ 200 -Tagger_eos 16 Embeddings _ 5 +Tagger_eos 16 Embeddings _ 20 Tagger_gov 16 Embeddings _ 100 -Tagger_pos 18 Embeddings _ 21 -Tagger_form 30 Embeddings _ 50000 -Tagger_form.f 30 Embeddings _ 200000 -Tagger_lemma 30 Embeddings _ 50000 -Tagger_letters 30 Embeddings _ 200000 -Tagger_labels 18 Embeddings _ 50 -Tagger_morpho 22 Embeddings _ 50000 +Tagger_pos 18 Embeddings _ 30 +Tagger_form 80 Embeddings _ 20 +Tagger_form.f 80 Embeddings _ 200000 +Tagger_lemma 80 Embeddings _ 20 +Tagger_letters 80 Embeddings _ 5000 +Tagger_labels 18 Embeddings _ 200 +Tagger_morpho 22 Embeddings _ 2000 # MORPHO -Morpho_actions 18 Embeddings _ 1000 -Morpho_bool 16 Embeddings _ 5 +Morpho_actions 18 Embeddings _ 200 +Morpho_bool 16 Embeddings _ 20 Morpho_int 16 Embeddings _ 200 -Morpho_eos 16 Embeddings _ 5 +Morpho_eos 16 Embeddings _ 20 Morpho_gov 16 Embeddings _ 100 -Morpho_pos 18 Embeddings _ 21 -Morpho_form 30 Embeddings _ 50000 -Morpho_form.f 30 Embeddings _ 20000 -Morpho_lemma 30 Embeddings _ 50000 -Morpho_letters 30 Embeddings _ 20000 -Morpho_labels 18 Embeddings _ 50 -Morpho_morpho 22 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Rules_form 30 Embeddings _ 50000 -Lemmatizer_Rules_letters 10 Embeddings _ 200000 -Lemmatizer_Rules_pos 30 Embeddings _ 21 -Lemmatizer_Rules_morpho 30 Embeddings _ 50000 -# LEMMATIZER -Lemmatizer_Case_form 30 Embeddings _ 50000 -Lemmatizer_Case_letters 10 Embeddings _ 200000 -Lemmatizer_Case_pos 30 Embeddings _ 21 -Lemmatizer_Case_morpho 30 Embeddings _ 50000 +Morpho_pos 18 Embeddings _ 30 +Morpho_form 80 Embeddings _ 20 +Morpho_form.f 80 Embeddings _ 200000 +Morpho_lemma 80 Embeddings _ 20 +Morpho_letters 80 Embeddings _ 5000 +Morpho_labels 18 Embeddings _ 200 +Morpho_morpho 22 Embeddings _ 2000 +# LEMMATIZER_RULE +Lemmatizer_Rules_actions 18 Embeddings _ 2000 +Lemmatizer_Rules_bool 16 Embeddings _ 20 +Lemmatizer_Rules_int 16 Embeddings _ 200 +Lemmatizer_Rules_eos 16 Embeddings _ 20 +Lemmatizer_Rules_gov 16 Embeddings _ 20 +Lemmatizer_Rules_pos 18 Embeddings _ 30 +Lemmatizer_Rules_form 80 Embeddings _ 100000 +Lemmatizer_Rules_form.f 80 Embeddings _ 20 +Lemmatizer_Rules_lemma 80 Embeddings _ 20 +Lemmatizer_Rules_letters 80 Embeddings _ 5000 +Lemmatizer_Rules_labels 18 Embeddings _ 20 +Lemmatizer_Rules_morpho 22 Embeddings _ 2000 +# LEMMATIZER_CASE +Lemmatizer_Case_actions 18 Embeddings _ 2000 +Lemmatizer_Case_bool 16 Embeddings _ 20 +Lemmatizer_Case_int 16 Embeddings _ 200 +Lemmatizer_Case_eos 16 Embeddings _ 20 +Lemmatizer_Case_gov 16 Embeddings _ 20 +Lemmatizer_Case_pos 18 Embeddings _ 30 +Lemmatizer_Case_form 80 Embeddings _ 100000 +Lemmatizer_Case_form.f 80 Embeddings _ 20 +Lemmatizer_Case_lemma 80 Embeddings _ 20 +Lemmatizer_Case_letters 80 Embeddings _ 5000 +Lemmatizer_Case_labels 18 Embeddings _ 20 +Lemmatizer_Case_morpho 22 Embeddings _ 2000 # PARSER Parser_actions 18 Embeddings _ 200 -Parser_bool 16 Embeddings _ 5 +Parser_bool 16 Embeddings _ 20 Parser_int 16 Embeddings _ 200 -Parser_eos 16 Embeddings _ 5 +Parser_eos 16 Embeddings _ 20 Parser_gov 16 Embeddings _ 100 -Parser_pos 18 Embeddings _ 21 -Parser_form 30 Embeddings _ 50000 -Parser_form.f 30 Embeddings _ 20000 -Parser_lemma 30 Embeddings _ 50000 -Parser_letters 30 Embeddings _ 20000 -Parser_labels 18 Embeddings _ 50 -Parser_morpho 22 Embeddings _ 50000 +Parser_pos 18 Embeddings _ 30 +Parser_form 80 Embeddings _ 20 +Parser_form.f 80 Embeddings _ 200000 +Parser_lemma 80 Embeddings _ 20 +Parser_letters 80 Embeddings _ 5000 +Parser_labels 18 Embeddings _ 200 +Parser_morpho 22 Embeddings _ 2000 # SEGMENTER -Segmenter_bool 02 Embeddings _ 5 -Segmenter_int 05 Embeddings _ 200 -Segmenter_letters 30 Embeddings _ 200000 -Segmenter_pos 15 Embeddings _ 21 -Segmenter_form 30 Embeddings _ 50000 -Segmenter_form.f 30 Embeddings _ 200000 -Segmenter_actions 05 Embeddings _ 21 -# ERROR_TAGGER -Error_Tagger_actions 18 Embeddings _ -Error_Tagger_bool 16 Embeddings _ -Error_Tagger_int 16 Embeddings _ -Error_Tagger_eos 16 Embeddings _ -Error_Tagger_gov 16 Embeddings _ -Error_Tagger_pos 18 Embeddings _ -Error_Tagger_form 30 Embeddings _ -Error_Tagger_lemma 30 Embeddings _ -Error_Tagger_letters 30 Embeddings _ -Error_Tagger_labels 18 Embeddings _ -Error_Tagger_morpho 22 Embeddings _ -# ERROR_MORPHO -Error_Morpho_actions 18 Embeddings _ -Error_Morpho_bool 16 Embeddings _ -Error_Morpho_int 16 Embeddings _ -Error_Morpho_eos 16 Embeddings _ -Error_Morpho_gov 16 Embeddings _ -Error_Morpho_pos 18 Embeddings _ -Error_Morpho_form 30 Embeddings _ -Error_Morpho_lemma 30 Embeddings _ -Error_Morpho_letters 30 Embeddings _ -Error_Morpho_labels 18 Embeddings _ -Error_Morpho_morpho 22 Embeddings _ -# ERROR_PARSER -Error_Parser_actions 18 Embeddings _ -Error_Parser_bool 16 Embeddings _ -Error_Parser_int 16 Embeddings _ -Error_Parser_eos 16 Embeddings _ -Error_Parser_gov 16 Embeddings _ -Error_Parser_pos 18 Embeddings _ -Error_Parser_form 30 Embeddings _ -Error_Parser_lemma 30 Embeddings _ -Error_Parser_letters 30 Embeddings _ -Error_Parser_labels 18 Embeddings _ -Error_Parser_morpho 22 Embeddings _ +Segmenter_actions 18 Embeddings _ 200 +Segmenter_bool 16 Embeddings _ 20 +Segmenter_int 16 Embeddings _ 200 +Segmenter_eos 16 Embeddings _ 20 +Segmenter_gov 16 Embeddings _ 100 +Segmenter_pos 18 Embeddings _ 30 +Segmenter_form 80 Embeddings _ 20 +Segmenter_form.f 80 Embeddings _ 200000 +Segmenter_lemma 80 Embeddings _ 20 +Segmenter_letters 80 Embeddings _ 5000 +Segmenter_labels 18 Embeddings _ 200 +Segmenter_morpho 22 Embeddings _ 2000 -- GitLab