Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

是否有计划将代码开源? #10

Closed
wangzhe258369 opened this issue Jun 25, 2019 · 6 comments
Closed

是否有计划将代码开源? #10

wangzhe258369 opened this issue Jun 25, 2019 · 6 comments

Comments

@wangzhe258369
Copy link

No description provided.

@ymcui
Copy link
Owner

ymcui commented Jun 25, 2019

感谢关注。
如报告中所提到的,我们尽量使用最简单的实现代码。
例如分类任务就是用谷歌官方的run_classifier.py,只需要参照已有的Processor,仿照写出对应的数据处理模块(只需修改其中的十几行甚至更少)即可,非常简单。
其他任务在GitHub中也有相关的开源实现(比如阅读理解,NER),故我们没有计划开源代码。

@ymcui ymcui closed this as completed Jun 25, 2019
@wangzhe258369
Copy link
Author

主要是在pretraining时做whole word mask的部分,想参考下是怎么把分词结果和FullTokenizer结合的。谢谢回复!

@ymcui
Copy link
Owner

ymcui commented Jun 27, 2019

你好,请参考我在这里的回复:#13

@wipen
Copy link

wipen commented Sep 18, 2019

可以试试我的实现:
https://github.com/wipen/bert_and_ernie

@lshowway
Copy link

主要是在pretraining时做whole word mask的部分,想参考下是怎么把分词结果和FullTokenizer结合的。谢谢回复!

有结果了嘛?知道预训练时WWM具体怎么处理的嘛?

@aiaiyueq11
Copy link

请问RBTL3模型为什么在预测的时候会出现结果随机的情况,我用RBTL3接了CRF作NER,对于同一条文本,每次的预测结果都不一致,当我换成其他预训练模型后就一致了。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants