Skip to content

Latest commit

 

History

History
20 lines (11 loc) · 1.2 KB

File metadata and controls

20 lines (11 loc) · 1.2 KB

Token classification with BERT-like models

This repository contains code for token classification with BERT-like models.

The fine-tuned GeoEDdA CamemBERT model is available on the HuggingFace Hub: hhttps://huggingface.co/GEODE/camembert-base-edda-span-classification 🤗

For training and evaluation we used the GeoEDdA dataset available also on HuggingFace: https://huggingface.co/datasets/GEODE/GeoEDdA 🤗

For more details about the dataset and tagset used, see: https://github.com/GEODE-project/ner-spancat-edda

Acknowledgements

The authors are grateful to the ASLAN project (ANR-10-LABX-0081) of the Université de Lyon, for its financial support within the French program "Investments for the Future" operated by the National Research Agency (ANR).