Feature | Description |
---|---|
Name | fr_ner4archives_v3_with_vectors |
Version | 0.0.0 |
spaCy | >=3.4.1,<3.5.0 |
Default Pipeline | tok2vec , ner |
Components | tok2vec , ner |
Vectors | 500000 keys, 500000 unique vectors (300 dimensions) |
Sources | French corpus for the NER task composed of finding aids in XML-EAD from the National Archives of France (v. 3.0) - Check corpus version on GitHub |
License | CC-BY-4.0 license |
Author | Archives nationales / Inria-Almanach |
Label Scheme
View label scheme (5 labels for 1 components)
Component | Labels |
---|---|
ner |
EVENT , LOCATION , ORGANISATION , PERSON , TITLE |
Accuracy
Type | Score |
---|---|
ENTS_F |
86.56 |
ENTS_P |
88.30 |
ENTS_R |
84.90 |
TOK2VEC_LOSS |
13527.63 |
NER_LOSS |
58805.82 |
- Downloads last month
- 3
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Space using ner4archives/fr_ner4archives_v3_with_vectors 1
Evaluation results
- NER Precisionself-reported0.883
- NER Recallself-reported0.849
- NER F Scoreself-reported0.866