吴恩达
深度学习工程师微专业 - 一线人工智能大师吴恩达亲研-网易云课堂 - 网易云课堂
https://github.com/cheneyoung/deeplearningAI
https://github.com/Kulbear/deep-learning-coursera
Transformer神经网络可视化指南
Transformer神经网络可视化指南_哔哩哔哩_bilibili
汉语自然语言处理-从零解读碾压循环神经网络的transformer模型(一)-注意力机制-位置编码-attention is all you need
汉语自然语言处理-从零解读碾压循环神经网络的transformer模型(一)-注意力机制-位置编码-attention is all you need_哔哩哔哩_bilibili
seq2seq,attention,transformer(keras代码实现)
seq2seq,attention,transformer(keras代码实现)_哔哩哔哩_bilibili
数学家是我理想
Transformer的PyTorch实现
Transformer的PyTorch实现_哔哩哔哩_bilibili
seq2seq的PyTorch实现
seq2seq的PyTorch实现_哔哩哔哩_bilibili
Seq2Seq(attention)的PyTorch实现
Seq2Seq(attention)的PyTorch实现_哔哩哔哩_bilibili
BERT的PyTorch实现
几个博客:
Transformer 详解
https://wmathor.com/index.php/archives/1438/
附加:
Bert (Bi-directional Encoder Representations from Transformers) Pytorch 源码解读(二)_ZJRN1027的博客-CSDN博客
The Illustrated Transformer – Jay Alammar – Visualizing machine learning one concept at a time.
代码:
http://nlp.seas.harvard.edu/2018/04/03/attention.html
https://wmathor.com/index.php/archives/1438/的其它泛化博客地址:
Transformers from scratch | peterbloem.nl
The Illustrated Transformer – Jay Alammar – Visualizing machine learning one concept at a time.
英文博客
https://jalammar.github.io/illustrated-transformer/
中文翻译:
可视化理解Transformer结构
https://zhuanlan.zhihu.com/p/59629215
https://blog.csdn.net/longxinchen_ml/article/details/86533005
一些可视化展示:
https://colab.research.google.com/github/tensorflow/tensor2tensor/blob/master/tensor2tensor/notebooks/hello_t2t.ipynb#scrollTo=OJKU36QAfqOC
https://www.youtube.com/playlist?list=PL86uXYUJ7999zE8u2-97i4KG_2Zpufkfb
https://zhuanlan.zhihu.com/p/398408338
https://zhuanlan.zhihu.com/p/105493618
其他:
68 Transformer【动手学深度学习v2】静姐你的李老师
https://www.bilibili.com/video/BV1Kq4y1H7FL?from=search&seid=12274291120981888663&spm_id_from=333.337.0.0
transformer原理全家桶与应用
transformer原理全家桶与应用_哔哩哔哩_bilibili
Transformer的PyTorch实现
Transformer的PyTorch实现_哔哩哔哩_bilibili
Transformer从零详细解读(可能是你见过最通俗易懂的讲解)
Transformer从零详细解读(可能是你见过最通俗易懂的讲解)_哔哩哔哩_bilibili
台大李宏毅21年机器学习课程 self-attention和transformer
台大李宏毅21年机器学习课程 self-attention和transformer_哔哩哔哩_bilibili
Transformer模型详解
Transformer模型详解 - Welcome to AI World
代码:
Attention 机制 – Transformer
Attention 机制 -- Transformer - 程序员大本营