From 6607f37acade5f87fc90e6367095670b03d146e6 Mon Sep 17 00:00:00 2001
From: Franck Dary <franck.dary@lis-lab.fr>
Date: Sun, 3 May 2020 18:59:32 +0200
Subject: [PATCH] updated tagparser_seq for new strategy
---
UD_any/data/Makefile | 4 ++--
UD_any/data/getTransitionSets.py | 20 --------------------
UD_any/tagparser_seq/machine.rm | 28 +++++++++++++++++-----------
3 files changed, 19 insertions(+), 33 deletions(-)
diff --git a/UD_any/data/Makefile b/UD_any/data/Makefile
index 49aea81..926906c 100644
--- a/UD_any/data/Makefile
+++ b/UD_any/data/Makefile
@@ -37,14 +37,14 @@ tokenizer.ts: all_no_test.conllu $(MCD)
segmenter.ts:
echo "EOS b.0" > $@
- echo "REWRITE b.0 EOS _" >> $@
+ echo "NOTHING" >> $@
+ sed -i -e 's/^/<segmenter> /' $@
columns: all_no_test.conllu $(MCD)
for number in 1 2 3 4 5 6 7 8 9 10 ; do \
cat all_no_test.conllu | sed '/^#/ d' | cut -f$$number | sort --unique > col_$$number.txt ; \
done
./getTransitionSets.py $(MCD) col_*\.txt
- cat tagger.ts parser.ts > taggerparser.ts
texts:
./getRawText.py $(CONLL2TXT) $(TRAIN_FILES) $(DEV_FILES) $(TEST_FILES)
diff --git a/UD_any/data/getTransitionSets.py b/UD_any/data/getTransitionSets.py
index 34b9551..dd5ad30 100755
--- a/UD_any/data/getTransitionSets.py
+++ b/UD_any/data/getTransitionSets.py
@@ -70,25 +70,6 @@ if __name__ == "__main__" :
output.close()
elif nameCol == "DEPREL" :
- output = open("parser_legacy.ts", 'w', encoding='utf-8')
- print("REDUCE", file=output)
- labels = set()
- labelsList = []
- for line in open(colFile, "r", encoding='utf-8') :
- striped = line.strip()
- if len(striped) == 0 or striped == "root" or striped == "_" :
- continue
- label = striped.split(':')[0]
- if label not in labels :
- labels.add(striped)
- labelsList.append(striped)
- labelsList.sort()
- for label in labelsList :
- print("LEFT " + label, file=output)
- print("RIGHT " + label, file=output)
- print("EOS s.0", file=output)
- print("SHIFT", file=output)
- output.close()
output = open("parser.ts", 'w', encoding='utf-8')
print("<parser> REDUCE", file=output)
labels = set()
@@ -105,7 +86,6 @@ if __name__ == "__main__" :
for label in labelsList :
print("<parser> LEFT " + label, file=output)
print("<parser> RIGHT " + label, file=output)
- print("<parser> EOS", file=output)
print("<parser> SHIFT", file=output)
output.close()
diff --git a/UD_any/tagparser_seq/machine.rm b/UD_any/tagparser_seq/machine.rm
index a6ef1f1..9efe192 100644
--- a/UD_any/tagparser_seq/machine.rm
+++ b/UD_any/tagparser_seq/machine.rm
@@ -1,9 +1,9 @@
-Name : Tagger, Feats and Parser Machine
-Classifier : tagmorpho
+Name : Tagger, Feats , Parser and Segmenter sequential Machine
+Classifier : tagparser
{
- Transitions : {tagger,data/tagger.ts morpho,data/morpho_parts.ts parser,data/parser.ts}
+ Transitions : {tagger,data/tagger.ts morpho,data/morpho_parts.ts parser,data/parser.ts segmenter,data/segmenter.ts}
Network type : Modular
- StateName : States{tagger morpho parser} Out{128}
+ StateName : States{tagger morpho parser segmenter} Out{128}
Context : Buffer{-3 -2 -1 0 1 2} Stack{} Columns{FORM UPOS} GRU{1 1 0 1} In{64} Out{128}
Focused : Column{ID} NbElem{1} Buffer{-1 0 1 2} Stack{2 1 0} GRU{1 1 0 1} In{64} Out{128}
Focused : Column{FORM} NbElem{10} Buffer{-1 0 1 2} Stack{2 1 0} GRU{1 1 0 1} In{64} Out{128}
@@ -17,10 +17,16 @@ Classifier : tagmorpho
Optimizer : Adam {0.0002 0.9 0.999 0.00000001 0.00001 true}
}
Predictions : UPOS FEATS HEAD DEPREL EOS
-Strategy : sequential
- tagger morpho 1
- morpho parser NOTHING 1
- morpho parser 0
- parser tagger SHIFT 1
- parser tagger RIGHT 1
- parser tagger 0
+Strategy
+{
+ Block : End{cannotMove}
+ tagger tagger * 1
+ Block : End{cannotMove}
+ morpho morpho NOTHING 1
+ morpho morpho * 0
+ Block : End{cannotMove}
+ parser segmenter SHIFT 0
+ parser segmenter RIGHT 0
+ parser parser * 0
+ segmenter parser * 1
+}
--
GitLab