README.md

metadata

language:
  - da
tags:
  - bert
  - pytorch
  - emotion
license: cc-by-sa-4.0
datasets:
  - social media
metrics:
  - f1
widget:
  - text: Der er et træ i haven.

Danish BERT for emotion detection

The BERT Emotion model detects whether a Danish text is emotional or not. It is based on the pretrained Danish BERT model by BotXO which has been fine-tuned on social media data.

See the DaNLP documentation for more details.

Here is how to use the model:

from transformers import BertTokenizer, BertForSequenceClassification

model = BertForSequenceClassification.from_pretrained("DaNLP/da-bert-emotion-binary")
tokenizer = BertTokenizer.from_pretrained("DaNLP/da-bert-emotion-binary")

Training data

The data used for training has not been made publicly available. It consists of social media data manually annotated in collaboration with Danmarks Radio.