INIST-CNRS/rapido

Rapido's project Model

This model was trained with and meant to be used with datas from Rapido project. The package flair was used for the training. The following versions of Flair and Transformers are required :

• transformers == 4.41.2
• flair == 0.10

How to use

from flair.models import SequenceTagger
from flair.data import Sentence

tagger = SequenceTagger.load('rapido-model.pt')
your_sentence = "Put your sentence here"
sentence = Sentence(your_sentence)
self.tagger.predict(sentences)
for sentence in sentences:
    for entity in sentence.get_spans('ner'):
        if entity.tag == "LOC":
            entity_txt = entity.text #get the entity
            entity_score = '%.3f' % entity.score #get the score
            print(entity_txt,score)

Validation score

          precision    recall  f1-score   support

     LOC     0.8790    0.8563    0.8675       891 

For more information

• https://github.com/Inist-CNRS/web-services/blob/main/services/data-rapido/README.md