transformer

详解Transformer （Attention Is All You Need） - 知乎
https://zhuanlan.zhihu.com/p/48508221

[整理] 聊聊 Transformer - 知乎
https://zhuanlan.zhihu.com/p/47812375

2. Attention Is All You Need（Transformer）算法原理解析 - hyc339408769 - 博客园
   https://www.cnblogs.com/huangyc/p/9813907.html

Transformer 模型的 PyTorch 实现 - 掘金
https://juejin.im/post/5b9f1af0e51d450e425eb32d

bert

bojone/bert4keras: keras implement of transformers for humans
https://github.com/bojone/bert4keras

原文：https://www.cnblogs.com/gxli/p/13159645.html