
This repository provides a small-sized Japanese GPT-2 model. The model was trained using code from Github repository rinnakk/japanese-pretrained-models by rinna Co., Ltd.
# coding = utf-8
import torch
import torch_npu
from transformers import T5Tokenizer, RobertaForMaskedLM
import argparse
from openmind import pipeline, is_torch_npu_available
parser = argparse.ArgumentParser(description='manual to this script')
parser.add_argument("--model_name_or_path", type=str, default="./")
args = parser.parse_args()
model_path = args.model_name_or_path
device = None
if is_torch_npu_available():
device = "npu:0"
else:
device = "cpu"
# load tokenizer
tokenizer = T5Tokenizer.from_pretrained(model_path)
tokenizer.do_lower_case = True # due to some bug of tokenizer config loading
# load model
model = RobertaForMaskedLM.from_pretrained(model_path)
model = model.to(device)
model = model.eval()
# original text
text = "4年に1度オリンピックは開かれる。"
# prepend [CLS]
text = "[CLS]" + text
# tokenize
tokens = tokenizer.tokenize(text)
print(tokens) # output: ['[CLS]', '▁4', '年に', '1', '度', 'オリンピック', 'は', '開かれる', '。']']
# mask a token
masked_idx = 5
tokens[masked_idx] = tokenizer.mask_token
print(tokens) # output: ['[CLS]', '▁4', '年に', '1', '度', '[MASK]', 'は', '開かれる', '。']
# convert to ids
token_ids = tokenizer.convert_tokens_to_ids(tokens)
print(token_ids) # output: [4, 1602, 44, 24, 368, 6, 11, 21583, 8]
# convert to tensor
token_tensor = torch.LongTensor([token_ids])
# provide position ids explicitly
position_ids = list(range(0, token_tensor.size(1)))
print(position_ids) # output: [0, 1, 2, 3, 4, 5, 6, 7, 8]
position_id_tensor = torch.LongTensor([position_ids])
# get the top 10 predictions of the masked token
with torch.no_grad():
outputs = model(input_ids=token_tensor.to(device), position_ids=position_id_tensor.to(device))
predictions = outputs[0][0, masked_idx].topk(10)
for i, index_t in enumerate(predictions.indices):
index = index_t.item()
token = tokenizer.convert_ids_to_tokens([index])[0]
print(i, token)A 12-layer, 768-hidden-size transformer-based language model.
The model was trained on Japanese CC-100 and Japanese Wikipedia to optimize a traditional language modelling objective on 8\*V100 GPUs for around 15 days. It reaches around 21 perplexity on a chosen validation set from CC-100.
The model uses a sentencepiece-based tokenizer, the vocabulary was trained on the Japanese Wikipedia using the official sentencepiece training script.
@misc{rinna-japanese-gpt2-small,
title = {rinna/japanese-gpt2-small},
author = {Zhao, Tianyu and Sawada, Kei},
url = {https://https://hf-mirror.com/rinna/japanese-gpt2-small}
}
@inproceedings{sawada2024release,
title = {Release of Pre-Trained Models for the {J}apanese Language},
author = {Sawada, Kei and Zhao, Tianyu and Shing, Makoto and Mitsui, Kentaro and Kaga, Akio and Hono, Yukiya and Wakatsuki, Toshiaki and Mitsuda, Koh},
booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
month = {5},
year = {2024},
pages = {13898--13905},
url = {https://aclanthology.org/2024.lrec-main.1213},
note = {\url{https://arxiv.org/abs/2404.01657}}
}