README.md

metadata

license: apache-2.0
language:
  - vi
metrics:
  - perplexity
pipeline_tag: fill-mask

Bkcare-base-pretrained: Pre-trained Language Models for Vietnamese in Health Text Mining

bkcare-bert-pretrained is the a strong baseline language models for Vietnamese in Healthcare domain.

Example usage

import torch
from transformers import AutoModel, AutoTokenizer

vihealthbert = AutoModel.from_pretrained("BookingCare/bkcare-bert-pretrained")
tokenizer = AutoTokenizer.from_pretrained("BookingCare/bkcare-bert-pretrained")

line = "Bệnh viện chợ rẫy ở Thành phố Hồ Chí Minh"

input_ids = torch.tensor([tokenizer.encode(line)])
with torch.no_grad():
    features = vihealthbert(input_ids)  # Models outputs are now tuples