README.md

metadata

language:
  - da
tags:
  - ner
  - bert
  - pytorch
  - transformers
license: CC BY-SA 4.0
datasets:
  - DaNE
metrics:
  - f1
widget:
  - text: Jens Peter Hansen kommer fra Danmark

BERT fine-tuned for Named Entity Recognition in Danish

The model tags tokens (in Danish sentences) with named entity tags (BIO format) [PER, ORG, LOC, MISC]. The pretrained language model used for fine-tuning is the Danish BERT by BotXO.

See the DaNLP documentation for more details.

Here is how to use the model:

from transformers import BertTokenizer, BertForTokenClassification

model = BertForTokenClassification.from_pretrained("DaNLP/da-bert-ner")
tokenizer = BertTokenizer.from_pretrained("DaNLP/da-bert-ner")

Training Data

The model has been trained on the DaNE.