fr_core_news_lg-2.3.0
·
1285 commits
to master
since this release
Details: https://spacy.io/models/fr#fr_core_news_lg
File checksum:
330283578d0eedb88290eba3927215ba3630e96f22bf342f94185ba4795591fa
French multi-task CNN trained on UD French Sequoia and WikiNER. Assigns word vectors, POS tags, dependency parse and named entities. Word vectors trained using FastText CBOW on Wikipedia and OSCAR (Common Crawl).
Feature | Description |
---|---|
Name | fr_core_news_lg |
Version | 2.3.0 |
spaCy | >=2.3.0,<2.4.0 |
Model size | 545 MB |
Pipeline | 聽tagger , parser , ner |
Vectors | 500000 keys, 500000 unique vectors (300 dimensions) |
Sources | UD French Sequoia v2.5 (Candito, Marie; Seddah, Djam茅; Perrier, Guy; Guillaume, Bruno) WikiNER OSCAR (Common Crawl) Wikipedia (20200301) |
License | LGPL |
Author | Explosion |
Label Scheme
Component | Labels |
---|---|
tagger |
聽ADJ , ADJ__Gender=Fem|Number=Plur , ADJ__Gender=Fem|Number=Plur|NumType=Ord , ADJ__Gender=Fem|Number=Sing , ADJ__Gender=Fem|Number=Sing|NumType=Ord , ADJ__Gender=Masc , ADJ__Gender=Masc|Number=Plur , ADJ__Gender=Masc|Number=Plur|NumType=Ord , ADJ__Gender=Masc|Number=Sing , ADJ__Gender=Masc|Number=Sing|NumType=Ord , ADJ__NumType=Ord , ADJ__Number=Plur , ADJ__Number=Sing , ADJ__Number=Sing|NumType=Ord , ADP , ADP_DET__Definite=Def|Gender=Masc|Number=Sing|PronType=Art , ADP_DET__Definite=Def|Number=Plur|PronType=Art , ADP_PRON__Gender=Fem|Number=Plur , ADP_PRON__Gender=Masc|Number=Plur , ADP_PRON__Gender=Masc|Number=Sing , ADV , ADV__Gender=Fem , ADV__Polarity=Neg , ADV__PronType=Int , AUX__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part , AUX__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , AUX__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , AUX__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin , AUX__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin , AUX__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , AUX__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin , AUX__Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin , AUX__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , AUX__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , AUX__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , AUX__Tense=Past|VerbForm=Part , AUX__Tense=Pres|VerbForm=Part , AUX__VerbForm=Inf , CCONJ , DET , DET__Definite=Def|Gender=Fem|Number=Sing|PronType=Art , DET__Definite=Def|Gender=Masc|Number=Sing|PronType=Art , DET__Definite=Def|Number=Plur|PronType=Art , DET__Definite=Def|Number=Sing|PronType=Art , DET__Definite=Ind|Gender=Fem|Number=Plur|PronType=Art , DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Art , DET__Definite=Ind|Gender=Masc|Number=Plur|PronType=Art , DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Art , DET__Definite=Ind|Number=Plur|PronType=Art , DET__Definite=Ind|Number=Sing|PronType=Art , DET__Gender=Fem|Number=Plur , DET__Gender=Fem|Number=Plur|PronType=Int , DET__Gender=Fem|Number=Sing , DET__Gender=Fem|Number=Sing|Poss=Yes , DET__Gender=Fem|Number=Sing|PronType=Dem , DET__Gender=Fem|Number=Sing|PronType=Int , DET__Gender=Masc|Number=Plur , DET__Gender=Masc|Number=Sing , DET__Gender=Masc|Number=Sing|PronType=Dem , DET__Gender=Masc|Number=Sing|PronType=Int , DET__Number=Plur , DET__Number=Plur|Poss=Yes , DET__Number=Plur|PronType=Dem , DET__Number=Sing , DET__Number=Sing|Poss=Yes , INTJ , NOUN , NOUN__Gender=Fem , NOUN__Gender=Fem|Number=Plur , NOUN__Gender=Fem|Number=Sing , NOUN__Gender=Masc , NOUN__Gender=Masc|Number=Plur , NOUN__Gender=Masc|Number=Plur|NumType=Card , NOUN__Gender=Masc|Number=Sing , NOUN__Gender=Masc|Number=Sing|NumType=Card , NOUN__NumType=Card , NOUN__Number=Plur , NOUN__Number=Sing , NUM , NUM__Gender=Masc|NumType=Card , NUM__NumType=Card , PART , PRON , PRON__Gender=Fem , PRON__Gender=Fem|Number=Plur , PRON__Gender=Fem|Number=Plur|Person=3 , PRON__Gender=Fem|Number=Plur|Person=3|PronType=Prs , PRON__Gender=Fem|Number=Plur|PronType=Dem , PRON__Gender=Fem|Number=Plur|PronType=Rel , PRON__Gender=Fem|Number=Sing , PRON__Gender=Fem|Number=Sing|Person=3 , PRON__Gender=Fem|Number=Sing|Person=3|PronType=Prs , PRON__Gender=Fem|Number=Sing|PronType=Dem , PRON__Gender=Fem|Number=Sing|PronType=Rel , PRON__Gender=Masc , PRON__Gender=Masc|Number=Plur , PRON__Gender=Masc|Number=Plur|Person=3 , PRON__Gender=Masc|Number=Plur|Person=3|PronType=Prs , PRON__Gender=Masc|Number=Plur|PronType=Dem , PRON__Gender=Masc|Number=Plur|PronType=Rel , PRON__Gender=Masc|Number=Sing , PRON__Gender=Masc|Number=Sing|Person=3 , PRON__Gender=Masc|Number=Sing|Person=3|PronType=Dem , PRON__Gender=Masc|Number=Sing|Person=3|PronType=Prs , PRON__Gender=Masc|Number=Sing|PronType=Dem , PRON__Gender=Masc|Number=Sing|PronType=Rel , PRON__NumType=Card , PRON__Number=Plur , PRON__Number=Plur|Person=1 , PRON__Number=Plur|Person=1|PronType=Prs , PRON__Number=Plur|Person=1|Reflex=Yes , PRON__Number=Plur|Person=2 , PRON__Number=Plur|Person=2|PronType=Prs , PRON__Number=Plur|Person=2|Reflex=Yes , PRON__Number=Plur|Person=3 , PRON__Number=Sing , PRON__Number=Sing|Person=1 , PRON__Number=Sing|Person=1|PronType=Prs , PRON__Number=Sing|Person=1|Reflex=Yes , PRON__Number=Sing|Person=2|PronType=Prs , PRON__Number=Sing|Person=3 , PRON__Number=Sing|PronType=Dem , PRON__Person=3 , PRON__Person=3|Reflex=Yes , PRON__PronType=Int , PRON__PronType=Rel , PROPN , PROPN__Gender=Fem|Number=Plur , PROPN__Gender=Fem|Number=Sing , PROPN__Gender=Masc , PROPN__Gender=Masc|Number=Plur , PROPN__Gender=Masc|Number=Sing , PROPN__Number=Plur , PROPN__Number=Sing , PUNCT , SCONJ , SYM , VERB__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part , VERB__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part , VERB__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part , VERB__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part , VERB__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Gender=Masc|Tense=Past|VerbForm=Part , VERB__Gender=Masc|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Mood=Cnd|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Cnd|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin , VERB__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Imp|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Imp|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin , VERB__Mood=Imp|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=2|Tense=Imp|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin , VERB__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin , VERB__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Ind|Person=3|VerbForm=Fin , VERB__Mood=Ind|VerbForm=Fin , VERB__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin , VERB__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin , VERB__Mood=Sub|Number=Sing|Person=3|Tense=Past|VerbForm=Fin , VERB__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin , VERB__Number=Plur|Tense=Past|VerbForm=Part , VERB__Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Number=Sing|Tense=Past|VerbForm=Part , VERB__Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass , VERB__Tense=Past|VerbForm=Part , VERB__Tense=Past|VerbForm=Part|Voice=Pass , VERB__Tense=Pres|VerbForm=Part , VERB__VerbForm=Inf , X , _SP |
parser |
聽ROOT , acl , acl:relcl , advcl , advmod , amod , appos , aux:pass , aux:tense , case , cc , ccomp , conj , cop , dep , det , expl:comp , expl:pass , expl:subj , fixed , flat:foreign , flat:name , iobj , mark , nmod , nsubj , nsubj:pass , nummod , obj , obl:agent , obl:arg , obl:mod , parataxis , punct , vocative , xcomp |
ner |
聽LOC , MISC , ORG , PER |
Accuracy
Type | Score |
---|---|
LAS |
聽85.78 |
UAS |
聽89.30 |
TOKEN_ACC |
聽98.52 |
TAGS_ACC |
聽96.23 |
ENTS_F |
聽85.63 |
ENTS_P |
聽85.78 |
ENTS_R |
聽85.48 |
Because the model is trained on Wikipedia, it may perform inconsistently on many genres, such as social media text. The NER accuracy refers to the "silver standard" annotations in the WikiNER corpus. Accuracy on these annotations tends to be higher than correct human annotations.
Installation
pip install spacy
python -m spacy download fr_core_news_lg