Transformer优秀博客
知乎讲解self-attention
https://zhuanlan.zhihu.com/p/48508221
多头注意力机制及pytorch实现
https://blog.csdn.net/HappyCtest/article/details/109847449
知乎讲解self-attention
https://zhuanlan.zhihu.com/p/48508221
多头注意力机制及pytorch实现
https://blog.csdn.net/HappyCtest/article/details/109847449