Bert Base Chinese Medical Ner

🚀 Chinese Medical Named Entity Recognition

This project focuses on named entity recognition in the Chinese medical field, providing a practical solution for extracting medical entities from text.

🚀 Quick Start

The project is available at: https://github.com/iioSnail/chinese_medical_ner

💻 Usage Examples

Basic Usage

from transformers import AutoModelForTokenClassification, BertTokenizerFast

tokenizer = BertTokenizerFast.from_pretrained('iioSnail/bert-base-chinese-medical-ner')
model = AutoModelForTokenClassification.from_pretrained("iioSnail/bert-base-chinese-medical-ner")

sentences = ["瘦脸针、水光针和玻尿酸详解！", "半月板钙化的病因有哪些？"]
inputs = tokenizer(sentences, return_tensors="pt", padding=True, add_special_tokens=False)
outputs = model(**inputs)
outputs = outputs.logits.argmax(-1) * inputs['attention_mask']

print(outputs)

Output Results

tensor([[1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4, 4, 4],
        [1, 2, 2, 2, 3, 4, 4, 4, 4, 4, 4, 4, 0, 0]])

Here, 1=B, 2=I, 3=E, 4=O. 1, 3 represents a two - character medical entity, 1,2,3 represents a three - character medical entity, 1,2,2,3 represents a four - character medical entity, and so on.

You can use MedicalNerModel.format_outputs(sentences, outputs) in the project to convert the output.

Advanced Usage

The effect is as follows:

[
  [
    {'start': 0, 'end': 3, 'word': '瘦脸针'},
    {'start': 4, 'end': 7, 'word': '水光针'},
    {'start': 8, 'end': 11, 'word': '玻尿酸'}
  ],
  [
    {'start': 0, 'end': 5, 'word': '半月板钙化'}
  ]
]

For more information, please refer to the project: https://github.com/iioSnail/chinese_medical_ner

📄 License

This project is licensed under the AFL-3.0 license.