NER4Archives pipeline optimized for GPU and specialized on French National Archives findings aids (XML-EAD). Components: transformer, ner. Based on camembert-base model.
Feature |
Description |
Name |
fr_ner4archives_camembert_base |
Version |
0.0.0 |
spaCy |
>=3.3.1,<3.4.0 |
Default Pipeline |
transformer , ner |
Components |
transformer , ner |
Vectors |
0 keys, 0 unique vectors (0 dimensions) |
Sources |
French corpus for the NER task composed of finding aids in XML-EAD from the National Archives of France (v. 2.0) - 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 |
86.31 |
ENTS_P |
87.25 |
ENTS_R |
85.40 |
TRANSFORMER_LOSS |
159157.68 |
NER_LOSS |
27979.50 |