metadata
license: mit
language:
- en
tags:
- medical
- radiology
model-index:
- name: rate-ner-rad
results: []
pipeline_tag: token-classification
RaTE-NER-Deberta
This model is a fine-tuned version of DeBERTa on the RaTE-NER dataset.
Model description
This model is trained to serve the RaTEScore metric, if you are interested in our pipeline, please refer to our paper and Github.
This model also can be used to extract Abnormality, Non-Abnormality, Anatomy, Disease, Non-Disease in medical radiology reports.
Usage
from transformers import AutoTokenizer, AutoModelForTokenClassification
ner_labels = ['B-ABNORMALITY', 'I-ABNORMALITY', 'B-NON-ABNORMALITY', 'I-NON-ABNORMALITY', 'B-DISEASE', 'I-DISEASE', 'B-NON-DISEASE', 'I-NON-DISEASE', 'B-ANATOMY', 'I-ANATOMY', 'O']
tokenizer = AutoTokenizer.from_pretrained("Angelakeke/RaTE-NER-Deberta")
model = AutoModelForTokenClassification.from_pretrained("Angelakeke/RaTE-NER-Deberta",
num_labels=len(ner_labels),
ignore_mismatched_sizes=True,
)
Author
Author: Weike Zhao
If you have any questions, please feel free to contact [email protected].
Citation