Allenlzcoder
diff --git a/‎chap-序列生成模型.pdf
717 KB b/‎chap-序列生成模型.pdf
717 KB
diff --git a/‎index.html
Lines changed: 226 additions & 60 deletions b/‎index.html
Lines changed: 226 additions & 60 deletions
diff --git a/‎md/index.md
Lines changed: 3 additions & 3 deletions b/‎md/index.md
Lines changed: 3 additions & 3 deletions
diff --git a/‎v/sgm-seq2seq-cnn-mt.gif
2.54 MB b/‎v/sgm-seq2seq-cnn-mt.gif
2.54 MB
diff --git a/‎v/sgm-seq2seq-fnn-wavenet.gif
560 KB b/‎v/sgm-seq2seq-fnn-wavenet.gif
560 KB
diff --git a/‎v/sgm-seq2seq-rnn-mt.gif
1.23 MB b/‎v/sgm-seq2seq-rnn-mt.gif
1.23 MB
diff --git a/‎v/sgm-seq2seq-transformer.gif
1.18 MB b/‎v/sgm-seq2seq-transformer.gif
1.18 MB
diff --git a/‎v/sgm-seq2seq.html
Lines changed: 231 additions & 0 deletions b/‎v/sgm-seq2seq.html
Lines changed: 231 additions & 0 deletions
diff --git a/‎v/sgm-seq2seq.md
Lines changed: 30 additions & 0 deletions b/‎v/sgm-seq2seq.md
Lines changed: 30 additions & 0 deletions
@@ -43,12 +43,12 @@ published: true
 7. [网络优化与正则化](chap-网络优化与正则化.pdf)  [[ppt](./ppt/chap-网络优化与正则化.pptx)]  (updated 2018-3-12)
 8. [注意力机制与外部记忆](chap-注意力机制与外部记忆.pdf) [[ppt](./ppt/chap-注意力机制与外部记忆.pptx)] (updated 2018-11-13)
 9. [无监督学习](chap-无监督学习.pdf) [[ppt](./ppt/chap-无监督学习.pptx)] (updated 2018-9-28)
-10. [模型独立的学习方式](chap-模型独立的学习方式.pdf)
+10. [模型独立的学习方式](chap-模型独立的学习方式.pdf) (updated 2019-3-28)
 11. [概率图模型](chap-概率图模型.pdf) [[ppt](./ppt/chap-概率图模型.pptx)] (updated 2018-5-8)
 12. [深度信念网络](chap-深度信念网络.pdf) [[ppt](./ppt/chap-深度信念网络.pptx)] (updated 2018-6-7)
 13. [深度生成模型](chap-深度生成模型.pdf)[[ppt](./ppt/chap-深度生成模型.pptx)] (updated 2018-6-25)
-14. [序列生成模型](chap-序列生成模型.pdf)    一个过时版本：[词嵌入与语言模型](chap-语言模型与词嵌入.pdf)
-15. [深度强化学习](chap-深度强化学习.pdf)  [[ppt](./ppt/chap-深度强化学习.pptx)] (updated 2018-6-15)
+14. [深度强化学习](chap-深度强化学习.pdf)  [[ppt](./ppt/chap-深度强化学习.pptx)] (updated 2018-6-15)
+15. [序列生成模型](chap-序列生成模型.pdf)  [[ppt](./ppt/chap-序列生成模型.pptx)]  (updated 2018-4-4)  一个过时版本：[词嵌入与语言模型](chap-语言模型与词嵌入.pdf)
 16. [数学基础](chap-数学基础.pdf)  (updated 2018-6-12)
 
 
 
@@ -0,0 +1,30 @@
+# 序列到序列示例 
+
+## 基于循环神经网络的序列到序列模型
+![](.\sgm-seq2seq-rnn-mt.gif)
+
+图片来源： <https://github.com/google/seq2seq>
+
+
+
+## Transformer示例
+
+![](.\sgm-seq2seq-transformer.gif)
+
+图片来源： <https://ai.googleblog.com/2017/08/transformer-novel-neural-network.html>
+
+
+
+## 基于前馈神经网络的序列到序列示例
+
+![](./sgm-seq2seq-fnn-wavenet.gif)
+
+图片来源：<<https://deepmind.com/blog/wavenet-generative-model-raw-audio/>>
+
+
+
+## 基于卷积神经网络的序列到序列示例
+
+![](./sgm-seq2seq-cnn-mt.gif)
+
+图片来源：<https://github.com/facebookresearch/fairseq/blob/master/fairseq.gif>