el_core_news_md-2.1.0

Details: https://spacy.io/models/el#el_core_news_md

File checksum: 7ae074564e262a954ae5c0be18949058eb7e10f421b5c917855589b4284cfee5

Greek pipeline with word vectors, POS tags, dependencies and named entities. Word vectors use Facebook's FastText Common Crawl vectors, pruned to a vocabulary of 20,000 items. Words outside the most frequent were mapped to the nearest neighbouring vector within the 20,000 rows retained. Syntax (dependencies and POS tags) trained from the Universal Dependencies conversion of the Greek Dependency Treebank (v2.2). Named entity annotations were created by Giannis Daras using Prodigy, using the OntoNotes 5 annotation schema.

Feature	Description
Name	el_core_news_md
Version	2.1.0
spaCy	>=2.1.0
Model size	126 MB
Pipeline	tagger, parser, ner
Vectors	1999938 keys, 20000 unique vectors (300 dimensions)
Sources	Common Crawl, Greek Dependency Treebank, Daras GSOC 2018
License	CC BY-NC 4.0
Author	Giannis Daras

Accuracy

Type	Score
ENTS_F	81.06
ENTS_P	78.66
ENTS_R	83.61
LAS	85.00
TAGS_ACC	96.56
TOKEN_ACC	100.00
UAS	88.29

Installation

pip install spacy
spacy download el_core_news_md

de_core_news_sm-2.1.0

Details: https://spacy.io/models/de#de_core_news_sm

File checksum: 25f1140eef95dccb124ca74b28db83f27787c210bcf6a97db4269690146e443a

German multi-task CNN trained on the TIGER and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	de_core_news_sm
Version	2.1.0
spaCy	>=2.1.0
Model size	10 MB
Pipeline	tagger, parser, ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	TIGER Corpus, Wikipedia
License	MIT
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	83.10
ENTS_P	84.09
ENTS_R	82.13
LAS	88.58
TAGS_ACC	96.27
TOKEN_ACC	95.88
UAS	90.66

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
spacy download de_core_news_sm

de_core_news_md-2.1.0

Details: https://spacy.io/models/de#de_core_news_md

File checksum: c1ab2f9e95b084dd895ac8f76b12c880d20766e886d3364350125971b3a72e60

German multi-task CNN trained on the TIGER and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	de_core_news_md
Version	2.1.0
spaCy	>=2.1.0
Model size	210 MB
Pipeline	tagger, parser, ner
Vectors	276087 keys, 20000 unique vectors (300 dimensions)
Sources	TIGER Corpus, Wikipedia
License	MIT
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	83.78
ENTS_P	84.25
ENTS_R	83.31
LAS	89.41
TAGS_ACC	96.55
TOKEN_ACC	95.88
UAS	91.23

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
spacy download de_core_news_md

en_vectors_web_lg-2.1.0a0

Details: https://spacy.io/models/en#en_vectors_web_lg

File checksum: 832304e25f0a35cdaed9638eca1d09a265ea1fb0c63ca90d74232f48c817d5ec

1.2m 300d vectors trained on Common Crawl with GloVe

Feature	Description
Name	en_vectors_web_lg
Version	2.1.0a0
spaCy	>=2.1.0.a7,<3.0.0
Model size	631 MB
Vectors	1070971 keys, 1070971 unique vectors (300 dimensions)
Sources	Common Crawl
License	CC BY-SA 3.0
Author	Jeffrey Pennington, Richard Socher, and Christopher D. Manning. Repackaged by Explosion AI

Installation

pip install spacy-nightly
spacy download en_vectors_web_lg

xx_ent_wiki_sm-2.1.0a7

Details: https://spacy.io/models/xx#xx_ent_wiki_sm

File checksum: 1d07e7a7acbc1750c680c687725eabc836f094b8eeaf1a1b7c7901d5e4b67fdc

Multi-lingual CNN trained on Nothman et al. (2010) Wikipedia corpus. Assigns named entities. Supports identification of PER, LOC, ORG and MISC entities for English, German, Spanish, French, Italian, Portuguese and Russian.

Feature	Description
Name	xx_ent_wiki_sm
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	3 MB
Pipeline	ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	Wikipedia
License	MIT
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	81.64
ENTS_P	82.23
ENTS_R	81.06

Because the model is trained on Wikipedia, it may perform inconsistently on many genres, such as social media text.

Installation

pip install spacy-nightly
spacy download xx_ent_wiki_sm

pt_core_news_sm-2.1.0a7

Details: https://spacy.io/models/pt#pt_core_news_sm

File checksum: 75e9d7ec03b5be522124a8b4e504cf636860c0466fc24a59c0b00928b29e0df1

Portuguese multi-task CNN trained on the Universal Dependencies and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	pt_core_news_sm
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	12 MB
Pipeline	tagger, parser, ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	Universal Dependencies, Wikipedia
License	CC BY-SA 4.0
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	88.98
ENTS_P	89.10
ENTS_R	88.86
LAS	86.20
TAGS_ACC	80.65
TOKEN_ACC	100.00
UAS	89.45

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-nightly
spacy download pt_core_news_sm

nl_core_news_sm-2.1.0a7

Details: https://spacy.io/models/nl#nl_core_news_sm

File checksum: 0ef11af5cac084f5d5110bf163e1f6ddb3575fd1e6ff64aee4e09140dbf7e1ea

Dutch multi-task CNN trained on the Universal Dependencies and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	nl_core_news_sm
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	10 MB
Pipeline	tagger, parser, ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	Universal Dependencies, Wikipedia
License	CC BY-SA 4.0
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	86.98
ENTS_P	86.46
ENTS_R	87.50
LAS	77.62
TAGS_ACC	91.53
TOKEN_ACC	100.00
UAS	83.85

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-nightly
spacy download nl_core_news_sm

it_core_news_sm-2.1.0a7

Details: https://spacy.io/models/it#it_core_news_sm

File checksum: 87c9839a78977388d0d2503779c1cbda1373237e9b434f72623eb19b5c3cc32c

Italian multi-task CNN trained on the Universal Dependencies and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	it_core_news_sm
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	10 MB
Pipeline	tagger, parser, ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	Universal Dependencies, Wikipedia
License	CC BY-NC-SA 3.0
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	86.31
ENTS_P	86.48
ENTS_R	86.13
LAS	87.20
TAGS_ACC	95.97
TOKEN_ACC	100.00
UAS	91.09

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-nightly
spacy download it_core_news_sm

fr_core_news_sm-2.1.0a7

Details: https://spacy.io/models/fr#fr_core_news_sm

File checksum: 5ac388c58ed1b2f9b306a24ae770a292c7919f8a5298eb2ab14f351a3dca8207

French multi-task CNN trained on the French Sequoia (Universal Dependencies) and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	fr_core_news_sm
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	12 MB
Pipeline	tagger, parser, ner
Vectors	0 keys, 0 unique vectors (0 dimensions)
Sources	Sequoia Corpus (UD), Wikipedia
License	LGPL
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	83.02
ENTS_P	83.10
ENTS_R	82.95
LAS	84.41
TAGS_ACC	94.66
TOKEN_ACC	100.00
UAS	87.28

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-nightly
spacy download fr_core_news_sm

fr_core_news_md-2.1.0a7

Details: https://spacy.io/models/fr#fr_core_news_md

File checksum: b18a388fae63d0dac45b0bb2aef6ce6f6833af4b29e11030325c3c58972afd21

French multi-task CNN trained on the French Sequoia (Universal Dependencies) and WikiNER corpus. Assigns context-specific token vectors, POS tags, dependency parse and named entities. Supports identification of PER, LOC, ORG and MISC entities.

Feature	Description
Name	fr_core_news_md
Version	2.1.0a7
spaCy	>=2.1.0a4
Model size	81 MB
Pipeline	tagger, parser, ner
Vectors	579447 keys, 20000 unique vectors (300 dimensions)
Sources	Sequoia Corpus (UD), Wikipedia
License	LGPL
Author	Explosion AI

Accuracy

Type	Score
ENTS_F	83.34
ENTS_P	83.43
ENTS_R	83.25
LAS	86.20
TAGS_ACC	95.28
TOKEN_ACC	100.00
UAS	89.10

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-nightly
spacy download fr_core_news_md