如何通过代理从Gluonnlp下载预训练模型？

Question

我在公司代理后面。我正在尝试下载BERT的词汇表。

_, vocab_bert = gluonnlp.model.bert_12_768_12(dataset_name='wiki_multilingual_uncased',
                                         pretrained=False, root='./model')

HTTPSConnectionPool(host='apache-mxnet.s3-accelerate.dualstack.amazonaws.com', 
port=443): Max retries exceeded with url: 
/gluon/dataset/vocab/wiki_multilingual_uncased-2b2514cc.zip 
(Caused by NewConnectionError('<urllib3.connection.VerifiedHTTPSConnection 
object at 0x7f43c2650b38>: 
Failed to establish a new connection: [Errno -2] Name or service not known',))

如何在Jupyter Notebook中将代理数据提供给gluonnlp？

Answer 1

我找到的解决方案是在Jupyter笔记本中设置代理：

import os 
os.environ['http_proxy'] = "http://user:password01ge@proxy.com:8080/"  
os.environ['https_proxy'] = "https://user:password01ge@proxy.com:8080/"