From df79cf1d2311fdd2da7900c84b4e018c5919c1ed Mon Sep 17 00:00:00 2001
From: Franck Dary <franck.dary@lis-lab.fr>
Date: Wed, 13 Apr 2022 16:51:55 +0200
Subject: [PATCH] updated templates machines
---
UD_any/templates/tokeparser_incr/machine.rm | 20 ++++++++++---------
UD_any/templates/tokeparser_seq/machine.rm | 22 +++++++++++----------
2 files changed, 23 insertions(+), 19 deletions(-)
diff --git a/UD_any/templates/tokeparser_incr/machine.rm b/UD_any/templates/tokeparser_incr/machine.rm
index e9e0a75..ad72b25 100644
--- a/UD_any/templates/tokeparser_incr/machine.rm
+++ b/UD_any/templates/tokeparser_incr/machine.rm
@@ -1,18 +1,21 @@
Name : Tokenizer, Tagger, Morpho, Lemmatizer, Parser and Segmenter Machine
Classifier : tokeparser
{
- Transitions : {tokenizer,data/tokenizer.ts tagger,data/tagger.ts morpho,data/morpho_whole.ts lemmatizer_rules,data/lemmatizer_rules.ts lemmatizer_case,data/lemmatizer_case.ts parser,data/parser_eager_rel_strict.ts segmenter,data/segmenter.ts}
+ Transitions : {tokenizer,data/tokenizer.ts tagger,data/tagger.ts morpho,data/morpho_whole.ts lemmatizer_rules,data/lemmatizer_rules.ts parser,data/parser_eager_rel_strict.ts segmenter,data/segmenter.ts}
LossMultiplier : {}
Network type : Modular
- Contextual : Window{-10 0} Columns{FORM} LSTM{1 1 0 1} In{64} Out{128} w2v{FORM,data/FORM.w2v} Targets{b.-3 b.-2 b.-1 b.0 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1}
- Context : Targets{b.-3 b.-2 b.-1 b.0 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{EOS ID UPOS FEATS DEPREL} LSTM{1 1 0 1} In{64} Out{64} w2v{}
+ Context : Targets{b.-3 b.-2 b.-1 b.0 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{FORM} Concat{1 1 0.0 1} In{300} Out{64} w2v{FORM,data/W2V/fasttextcleanfiltered.w2v}
+ Context : Targets{b.-3 b.-2 b.-1 b.0 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{EOS ID UPOS FEATS DEPREL} LSTM{1 1 0 1} In{128} Out{64} w2v{}
Focused : Column{prefix3:FORM} NbElem{3} Buffer{0} Stack{} LSTM{1 1 0 1} In{64} Out{64} w2v{}
Focused : Column{suffix3:FORM} NbElem{3} Buffer{0} Stack{} LSTM{1 1 0 1} In{64} Out{64} w2v{}
RawInput : Left{5} Right{10} LSTM{1 1 0.0 1} In{32} Out{32}
- History : NbElem{10} LSTM{1 1 0 1} In{32} Out{32}
- SplitTrans : LSTM{1 1 0.0 1} In{64} Out{64}
- InputDropout : 0.3
- MLP : {1600 0.3 1600 0.3}
+ History : NbElem{10} LSTM{1 1 0 1} In{128} Out{64}
+ HistoryMine : NbElem{4} LSTM{1 1 0 1} In{128} Out{64}
+ StateName : Out{64}
+ Distance : FromBuffer{} FromStack{0 1 2} ToBuffer{0} ToStack{} Threshold{15} LSTM{1 1 0.0 1} In{128} Out{64}
+ SplitTrans : LSTM{1 1 0.0 1} In{128} Out{64}
+ InputDropout : 0.5
+ MLP : {3200 0.4 1600 0.4}
End
Optimizer : Adagrad {0.01 0.000001 0 0.0000000001}
Type : classification
@@ -28,8 +31,7 @@ Strategy
tokenizer tokenizer * 0
tagger morpho * 0
morpho lemmatizer_rules * 0
- lemmatizer_rules lemmatizer_case * 0
- lemmatizer_case parser * 0
+ lemmatizer_rules parser * 0
parser segmenter eager_SHIFT 0
parser segmenter eager_RIGHT_rel 0
parser parser * 0
diff --git a/UD_any/templates/tokeparser_seq/machine.rm b/UD_any/templates/tokeparser_seq/machine.rm
index 93d9bec..f63c06a 100644
--- a/UD_any/templates/tokeparser_seq/machine.rm
+++ b/UD_any/templates/tokeparser_seq/machine.rm
@@ -1,18 +1,21 @@
Name : Tokenizer, Tagger, Morpho, Lemmatizer, Parser and Segmenter Machine
Classifier : tokeparser
{
- Transitions : {tokenizer,data/tokenizer.ts tagger,data/tagger.ts morpho,data/morpho_whole.ts lemmatizer_rules,data/lemmatizer_rules.ts lemmatizer_case,data/lemmatizer_case.ts parser,data/parser_eager_rel_strict.ts segmenter,data/segmenter.ts}
- LossMultiplier : {}
+ Transitions : {tokenizer,data/tokenizer.ts tagger,data/tagger.ts morpho,data/morpho_whole.ts lemmatizer_rules,data/lemmatizer_rules.ts parser,data/parser_eager_rel_strict.ts segmenter,data/segmenter.ts}
+ LossMultiplier : {}
Network type : Modular
- Contextual : Window{-10 10} Columns{FORM} LSTM{1 1 0 1} In{64} Out{128} w2v{FORM,data/FORM.w2v} Targets{b.-3 b.-2 b.-1 b.0 b.1 b.2 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1}
- Context : Targets{b.-3 b.-2 b.-1 b.0 b.1 b.2 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{EOS ID UPOS FEATS DEPREL} LSTM{1 1 0 1} In{64} Out{64} w2v{}
+ Context : Targets{b.-3 b.-2 b.-1 b.0 b.1 b.2 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{FORM} Concat{1 1 0.0 1} In{300} Out{64} w2v{FORM,data/W2V/fasttextcleanfiltered.w2v}
+ Context : Targets{b.-3 b.-2 b.-1 b.0 b.1 b.2 s.0 s.1 s.2 b.0.0 s.0.0 s.0.-1 s.1.0 s.1.-1 s.2.0 s.2.-1} Columns{EOS ID UPOS FEATS DEPREL} LSTM{1 1 0 1} In{128} Out{64} w2v{}
Focused : Column{prefix3:FORM} NbElem{3} Buffer{0} Stack{} LSTM{1 1 0 1} In{64} Out{64} w2v{}
Focused : Column{suffix3:FORM} NbElem{3} Buffer{0} Stack{} LSTM{1 1 0 1} In{64} Out{64} w2v{}
RawInput : Left{5} Right{10} LSTM{1 1 0.0 1} In{32} Out{32}
- History : NbElem{10} LSTM{1 1 0 1} In{32} Out{32}
- SplitTrans : LSTM{1 1 0.0 1} In{64} Out{64}
- InputDropout : 0.3
- MLP : {1600 0.3 1600 0.3}
+ History : NbElem{10} LSTM{1 1 0 1} In{128} Out{64}
+ HistoryMine : NbElem{4} LSTM{1 1 0 1} In{128} Out{64}
+ StateName : Out{64}
+ Distance : FromBuffer{} FromStack{0 1 2} ToBuffer{0} ToStack{} Threshold{15} LSTM{1 1 0.0 1} In{128} Out{64}
+ SplitTrans : LSTM{1 1 0.0 1} In{128} Out{64}
+ InputDropout : 0.5
+ MLP : {3200 0.4 1600 0.4}
End
Optimizer : Adagrad {0.01 0.000001 0 0.0000000001}
Type : classification
@@ -31,8 +34,7 @@ Strategy
Block : End{cannotMove}
morpho morpho * 1
Block : End{cannotMove}
- lemmatizer_rules lemmatizer_case * 0
- lemmatizer_case lemmatizer_rules * 1
+ lemmatizer_rules lemmatizer_rules * 1
Block : End{cannotMove}
parser segmenter eager_SHIFT 0
parser segmenter eager_RIGHT_rel 0
--
GitLab