[Deep learning] Detailed explanation of Transformer model

Foreword

  This article is a learning record, and the content and pictures in it are mostly borrowed from other articles. Links to related blog posts are given in the references.

Overall Architecture

Encoder

Decoder

References

[1]Self-Attention and Transformer
[2]Detailed explanation of Transformer (Attention Is All You Need)
[3]Highly recommended!NTU Li Hongyi's self-attention mechanism and Transformer explained in detail!
[4]The Illustrated Transformer
[5] Understanding of Q, K, V in Transformer
[6]Why is the V in the (KQV) of the transformer's self_attention also multiplied by a Wv matrix?
[9]The Annotated Transformer