metadata

language: ja
license: cc-by-nc-sa-4.0
tags:
  - roberta
  - medical
mask_token: '[MASK]'
widget:
  - text: この患者は[MASK]と診断された。

alabnii/jmedroberta-base-sentencepiece

Model description

This is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).

This model is released under the Creative Commons 4.0 International License (CC BY-NC-SA 4.0).

Datasets used for pre-training

abstracts (train: 1.6GB (10M sentences), validation: 0.2GB (1.3M sentences))
abstracts & body texts (train: 0.2GB (1.4M sentences))

How to use

Input text must be converted to full-width characters（全角）in advance.

You can use this model for masked language modeling as follows:

from transformers import AutoModelForMaskedLM, AutoTokenizer

model = AutoModelForMaskedLM.from_pretrained("alabnii/jmedroberta-base-sentencepiece")
model.eval()
tokenizer = AutoTokenizer.from_pretrained("alabnii/jmedroberta-base-sentencepiece")

texts = ['この患者は[MASK]と診断された。']
inputs = tokenizer.batch_encode_plus(texts, return_tensors='pt')
outputs = model(**inputs)
tokenizer.convert_ids_to_tokens(outputs.logits[0][1:-1].argmax(axis=-1))
# ['▁この', '患者は', 'ＡＭＬ', '▁', 'と診断された', '。']

Alternatively, you can employ Fill-mask pipeline.

from transformers import pipeline
fill = pipeline("fill-mask", model="alabnii/jmedroberta-base-sentencepiece", top_k=10)
fill("この患者は[MASK]と診断された。")
#[{'score': 0.04239409416913986,
#  'token': 7698,
#  'token_str': 'ＡＭＬ',
#  'sequence': 'この患者はＡＭＬ と診断された。'},
# {'score': 0.03562006726861,
#  'token': 3298,
#  'token_str': 'ＳＬＥ',
#  'sequence': 'この患者はＳＬＥ と診断された。'},
# {'score': 0.025064188987016678,
#  'token': 10303,
#  'token_str': 'ＭＤＳ',
#  'sequence': 'この患者はＭＤＳ と診断された。'},
# ...

You can fine-tune this model on downstream tasks.

See also sample Colab notebooks: https://colab.research.google.com/drive/1BUD3DKOUMqcwIO3X5bYUOsR_wDzgOJcd?usp=sharing

Tokenization

Each sentence is tokenized into tokens by SentencePiece (Unigram).

Vocabulary

The vocabulary consists of 30000 tokens induced by SentencePiece (Unigram).

Training procedure

The following hyperparameters were used during pre-training:

learning_rate: 0.0001
train_batch_size: 32
eval_batch_size: 32
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 256
total_eval_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 20000
training_steps: 2000000
mixed_precision_training: Native AMP

Note: Why do we call our model RoBERTa, not BERT?

As the config file suggests, our model is based on HuggingFace's BertForMaskedLM class. However, we consider our model as RoBERTa for the following reasons:

We kept training only with max sequence length (= 512) tokens.
We removed the next sentence prediction (NSP) training objective.
We introduced dynamic masking (changing the masking pattern in each training iteration).

Acknowledgements

This work was supported by Japan Japan Science and Technology Agency (JST) AIP Trilateral AI Research (Grant Number: JPMJCR20G9), and Joint Usage/Research Center for Interdisciplinary Large-scale Information Infrastructures (JHPCN) (Project ID: jh221004), in Japan.
In this research work, we used the "mdx: a platform for the data-driven future".