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 Click to expand the usage of this model. from transformers import AutoTokenizer, AutoModelForTokenClassification import torch def post process(tokenized text, predicted entities, tokenizer): entity spans = [] start = end = None entity type = None for i, (token, label) in enumerate(zip(tokenized text, predicted entities[:len(tokenized text)])): if token in ["[CLS]", "[SEP]"]: continue if label != "O" and i Author Author: Weike Zhao If you have any questions, please feel free to contact zwk0629@sjtu.edu.cn. Citation
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy