T5-EN-VI-SMALL：预训练文本到文本转换变压器，用于英文越南文翻译

数据集

使用了IWSLT'15英文-越南文数据，来自 Stanford NLP group 。对于所有实验，语料库被分为训练集、开发集和测试集：

Data set	Sentences	Download
Training	133,317	via 1233321 or located in data/train-en-vi.tgz
Development	1,553	via 1234321 or located in data/dev-2012-en-vi.tgz
Test	1,268	via 1235321 or located in data/test-2013-en-vi.tgz

结果

测试集上的结果。

Model	BLEU (Beam Search)
1236321	23.30
Sequence-to-sequence model with attention	26.10
Neural Phrase-based Machine Translation 1237321	27.69
Neural Phrase-based Machine Translation + LM 1237321	28.07
t5-en-vi-small (pretraining, without training data)	28.46 (cased) / 29.23 (uncased)
t5-en-vi-small (fineturning with training data)	32.38 (cased) / 33.19 (uncased)

示例使用

import torch

from transformers import T5ForConditionalGeneration, T5Tokenizer
import torch
if torch.cuda.is_available():       
    device = torch.device("cuda")

    print('There are %d GPU(s) available.' % torch.cuda.device_count())

    print('We will use the GPU:', torch.cuda.get_device_name(0))
else:
    print('No GPU available, using the CPU instead.')
    device = torch.device("cpu")

model = T5ForConditionalGeneration.from_pretrained("NlpHUST/t5-en-vi-small")
tokenizer = T5Tokenizer.from_pretrained("NlpHUST/t5-en-vi-small")
model.to(device)

src = "In school , we spent a lot of time studying the history of Kim Il-Sung , but we never learned much about the outside world , except that America , South Korea , Japan are the enemies ."
tokenized_text = tokenizer.encode(src, return_tensors="pt").to(device)
model.eval()
summary_ids = model.generate(
                    tokenized_text,
                    max_length=128, 
                    num_beams=5,
                    repetition_penalty=2.5, 
                    length_penalty=1.0, 
                    early_stopping=True
                )
output = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
print(output)

输出

Ở trường, chúng tôi dành nhiều thời gian để nghiên cứu về lịch sử Kim Il-Sung, nhưng chúng tôi chưa bao giờ học được nhiều về thế giới bên ngoài, ngoại trừ Mỹ, Hàn Quốc, Nhật Bản là kẻ thù.

联系信息

关于此项目的个人交流，请联系Nha Nguyen Van（nha282@gmail.com）。

作者:

NLP HUST

数据集大小:

2.24 GB