Skip to content
Snippets Groups Projects
Commit 91d8fd0f authored by Carlos Ramisch's avatar Carlos Ramisch
Browse files

Remove very rare morphological features to simplify TP3

parent 3d0d99eb
Branches
No related tags found
No related merge requests found
......@@ -76,6 +76,17 @@ def simplify_supersense(sent):
token["frsemcor:noun"] = "Feeling"
return del_ssense_counter, mod_ssense_counter
def simplify_morphology(sent):
for token in sent:
if token["feats"] :
if "Typo" in token["feats"]:
del token['feats']['Typo']
elif token["upos"] in ["ADV", "NUM"]:
if "Number" in token["feats"]:
del token['feats']['Number']
elif "Gender" in token['feats']:
del token['feats']['Gender']
#########################################
def simplify_mwe_ne(sent):
......@@ -148,6 +159,7 @@ with open(sys.argv[1], "r", encoding="UTF=8") as f:
del_ssense_counter = del_ssense_counter + del_ssense_ci
mod_ssense_counter = mod_ssense_counter + mod_ssense_ci
del_ne_counter = del_ne_counter + simplify_mwe_ne(sent)
simplify_morphology(sent)
# subrel_counter = subrel_counter + remove_subrelations(sent)
if is_projective(sent) : # Returns false to remove sentence
if sent.metadata.get("global.columns", None): # Add header for new column
......
......@@ -1519,7 +1519,7 @@
17 le le DET _ Definite=Def|Gender=Masc|Number=Sing|PronType=Art 18 det _ _ 4 * *
18 myocarde myocarde NOUN _ Gender=Masc|Number=Sing 15 nmod _ _ 4 * *
19 sans sans ADP _ _ 20 case _ _ * * *
20 sus sus NOUN _ Typo=Yes 15 nmod _ CorrectForm=sus-décalage 5:_|MWE|LEX * *
20 sus sus NOUN _ _ 15 nmod _ CorrectForm=sus-décalage 5:_|MWE|LEX * *
21 décalage _ X _ _ 20 goeswith _ _ 5 * *
22 de de ADP _ _ 24 case _ _ 5 * *
23 le le DET _ Definite=Def|Gender=Masc|Number=Sing|PronType=Art 24 det _ _ 5 * *
......
......@@ -14461,7 +14461,7 @@
17 le le DET _ Definite=Def|Gender=Masc|Number=Sing|PronType=Art 18 det _ _ 4 * *
18 myocarde myocarde NOUN _ Gender=Masc|Number=Sing 15 nmod _ _ 4 * *
19 sans sans ADP _ _ 20 case _ _ * * *
20 sus sus NOUN _ Typo=Yes 15 nmod _ CorrectForm=sus-décalage 5:_|MWE|LEX * *
20 sus sus NOUN _ _ 15 nmod _ CorrectForm=sus-décalage 5:_|MWE|LEX * *
21 décalage _ X _ _ 20 goeswith _ _ 5 * *
22 de de ADP _ _ 24 case _ _ 5 * *
23 le le DET _ Definite=Def|Gender=Masc|Number=Sing|PronType=Art 24 det _ _ 5 * *
......@@ -17189,7 +17189,7 @@
# sent_id = emea-fr-dev_00222
# text = Troubles musculo squelettiques, du tissu conjonctif et des os
1 Troubles trouble NOUN _ Gender=Masc|Number=Plur 0 root _ _ * State *
2 musculo musculo ADJ _ Typo=Yes 1 amod _ CorrectForm=musculo-squelettiques * * *
2 musculo musculo ADJ _ _ 1 amod _ CorrectForm=musculo-squelettiques * * *
3 squelettiques _ X _ _ 2 goeswith _ _ * * *
4 , , PUNCT _ _ 7 punct _ _ * * *
5 de de ADP _ _ 7 case _ _ * * *
......@@ -42396,7 +42396,7 @@
27 l' le DET _ Definite=Def|Number=Sing|PronType=Art 28 det _ _ * * *
28 embargo embargo NOUN _ Gender=Masc|Number=Sing 25 nmod _ _ * Act *
29 , , PUNCT _ _ 22 punct _ _ * * *
30 faute faute ADV _ Gender=Fem|Number=Sing 22 advmod _ _ 2:ADP|MWE|IRREG * *
30 faute faute ADV _ Gender=Fem 22 advmod _ _ 2:ADP|MWE|IRREG * *
31 de de ADP _ _ 32 case _ _ 2 * *
32 produits produit NOUN _ Gender=Masc|Number=Plur 30 obl:arg _ _ * Food *
33 alimentaires alimentaire ADJ _ Number=Plur 32 amod _ _ * * *
......@@ -75442,7 +75442,7 @@
3 août août NOUN _ Gender=Masc|Number=Sing 2 nmod _ _ * * *
4 , , PUNCT _ _ 13 punct _ _ * * *
5 les le DET _ Definite=Def|Number=Plur|PronType=Art 9 det _ _ * * *
6 neuf neuf NUM _ Gender=Masc|NumType=Card 9 amod _ ExtPos=ADJ * * *
6 neuf neuf NUM _ NumType=Card 9 amod _ ExtPos=ADJ * * *
7 cent cent NUM _ NumType=Card 6 fixed _ _ * * *
8 vingt vingt NUM _ NumType=Card 6 fixed _ _ * * *
9 photos photo NOUN _ Gender=Fem|Number=Plur 13 nsubj:pass _ _ * Artifact *
This diff is collapsed.
......@@ -1272,7 +1272,7 @@
27 l' le DET _ Definite=Def|Number=Sing|PronType=Art 28 det _ _ * * *
28 embargo embargo NOUN _ Gender=Masc|Number=Sing 25 nmod _ _ * Act *
29 , , PUNCT _ _ 22 punct _ _ * * *
30 faute faute ADV _ Gender=Fem|Number=Sing 22 advmod _ _ 2:ADP|MWE|IRREG * *
30 faute faute ADV _ Gender=Fem 22 advmod _ _ 2:ADP|MWE|IRREG * *
31 de de ADP _ _ 32 case _ _ 2 * *
32 produits produit NOUN _ Gender=Masc|Number=Plur 30 obl:arg _ _ * Food *
33 alimentaires alimentaire ADJ _ Number=Plur 32 amod _ _ * * *
......
......@@ -12114,7 +12114,7 @@
# sent_id = emea-fr-dev_00222
# text = Troubles musculo squelettiques, du tissu conjonctif et des os
1 Troubles trouble NOUN _ Gender=Masc|Number=Plur 0 root _ _ * State *
2 musculo musculo ADJ _ Typo=Yes 1 amod _ CorrectForm=musculo-squelettiques * * *
2 musculo musculo ADJ _ _ 1 amod _ CorrectForm=musculo-squelettiques * * *
3 squelettiques _ X _ _ 2 goeswith _ _ * * *
4 , , PUNCT _ _ 7 punct _ _ * * *
5 de de ADP _ _ 7 case _ _ * * *
......@@ -53929,7 +53929,7 @@
3 août août NOUN _ Gender=Masc|Number=Sing 2 nmod _ _ * * *
4 , , PUNCT _ _ 13 punct _ _ * * *
5 les le DET _ Definite=Def|Number=Plur|PronType=Art 9 det _ _ * * *
6 neuf neuf NUM _ Gender=Masc|NumType=Card 9 amod _ ExtPos=ADJ * * *
6 neuf neuf NUM _ NumType=Card 9 amod _ ExtPos=ADJ * * *
7 cent cent NUM _ NumType=Card 6 fixed _ _ * * *
8 vingt vingt NUM _ NumType=Card 6 fixed _ _ * * *
9 photos photo NOUN _ Gender=Fem|Number=Plur 13 nsubj:pass _ _ * Artifact *
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment