HiTZ
/

Token Classification
Transformers
Safetensors
bert
Inference Endpoints
ragerri's picture
Update README.md
b899520 verified
metadata
license: apache-2.0
base_model: bert-base-multilingual-cased
datasets:
  - HiTZ/multilingual-abstrct
language:
  - en
  - es
  - fr
  - it
metrics:
  - f1
pipeline_tag: token-classification
library_name: transformers
widget:
  - text: >-
      The dysuria resolved faster in patients implanted with 103Pd but was
      unaffected by the use of supplemental radiotherapy and/or androgen
      deprivation therapy.
  - text: >-
      La disuria se resolvió más rápidamente en los pacientes implantados con
      103Pd, pero no se vio afectada por el uso de radioterapia suplementaria
      y/o terapia de privación de andrógenos.
  - text: >-
      La dysurie s'est résorbée plus rapidement chez les patients implantés avec
      du 103Pd, mais n'a pas été affectée par l'utilisation d'une radiothérapie
      complémentaire et/ou d'une thérapie de privation d'androgènes.
  - text: >-
      La disuria si è risolta più rapidamente nei pazienti impiantati con 103Pd,
      ma non è stata influenzata dall'uso della radioterapia supplementare e/o
      della terapia di deprivazione androgenica.


mBERT for multilingual Argument Detection in the Medical Domain

This model is a fine-tuned version of bert-base-multilingual-cased for the argument component detection task on AbstRCT data in English, Spanish, French and Italian (https://huggingface.co/datasets/HiTZ/multilingual-abstrct).

Performance

F1-macro scores (at sequence level) and their averages per test set from the argument component detection results of monolingual, monolingual automatically post-processed, multilingual, multilingual automatically post-processed, and crosslingual experiments.

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3.0

Framework versions

  • Transformers 4.40.0.dev0
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.2

Contact: Anar Yeginbergen and Rodrigo Agerri HiTZ Center - Ixa, University of the Basque Country UPV/EHU