【基础整理】attention:浅谈注意力机制与自注意力模型(附键值对注意力 + 多头注意力)
Vaswani, Ashish, et al. Attention is all you need. Advances in Neural Information Processing Systems. 2017.论文原文:https://arxiv.org/pdf/1706.03762v5.pdf源码:https://github.com/tensorflow/tensor2tensor (tensorflow / official)** https://github.com/facebookre.
原创
2021-03-09 22:41:30 ·
5394 阅读 ·
2 评论