Helsinki-NLP/opus-mt-zh-en模型部署

最新推荐文章于 2025-01-20 10:31:49 发布

蓝净云

最新推荐文章于 2025-01-20 10:31:49 发布

阅读量983

点赞数 3

文章标签：自然语言处理机器翻译 python

本文链接：https://blog.youkuaiyun.com/weixin_45735391/article/details/145060949

版权

Helsinki-NLP

https://hf-mirror.com/Helsinki-NLP/opus-mt-zh-en

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import torch

model = AutoModelForSeq2SeqLM.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en").eval()
tokenizer = AutoTokenizer.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en")

def translate(text):
    with torch.no_grad():
        encoded = tokenizer([text], return_tensors="pt")
        sequences = model.generate(**encoded)
        return tokenizer.batch_decode(sequences, skip_special_tokens=True)[0]

input = "青春不能回头，所以青春没有终点。,.,/'[' ——《火影忍者》"
print(translate(input))

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en")
model = AutoModelForSeq2SeqLM.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en")

content = ['你好 世界', '哈哈哈哈']

inputs = tokenizer(content, return_tensors="pt", padding=True)
translated_tokens = model.generate(**inputs, )
for translated in tokenizer.batch_decode(translated_tokens, skip_special_tokens=True):
    print(translated)

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline

tokenizer = AutoTokenizer.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en")
model = AutoModelForSeq2SeqLM.from_pretrained("/data/models/Helsinki-NLP/opus-mt-zh-en")

translator = pipeline(
    'translation',
    model=model,
    tokenizer=tokenizer,
    src_lang='zho_Hans',
    tgt_lang='eng_Latn',
    max_length=512
)
print(translator(["你好 世界", "青春", ]))