Browse Source

Update Lecture5

main
Fafa-DL 2 years ago
parent
commit
cdca8cc65c
33 changed files with 2340 additions and 2 deletions
  1. +2337
    -0
      2022 ML/05 Sequence to sequence/HW05.ipynb
  2. BIN
      2022 ML/05 Sequence to sequence/HW05.pdf
  3. BIN
      2022 ML/05 Sequence to sequence/xformer (v6).pdf
  4. +3
    -2
      README.md
  5. +0
    -0
      选修 To Learn More/第一节/BP.pdf
  6. +0
    -0
      选修 To Learn More/第一节/Classification.pdf
  7. +0
    -0
      选修 To Learn More/第一节/DL.pdf
  8. +0
    -0
      选修 To Learn More/第一节/Logistic Regression.pdf
  9. +0
    -0
      选修 To Learn More/第一节/Regression.pdf
  10. +0
    -0
      选修 To Learn More/第七节/BERT train (v8).pdf
  11. +0
    -0
      选修 To Learn More/第七节/GPT3 (v6).pdf
  12. +0
    -0
      选修 To Learn More/第七节/Multi (v2).pdf
  13. BIN
      选修 To Learn More/第三节/Special Structure (v6).pdf
  14. +0
    -0
      选修 To Learn More/第二节/Optimization.pdf
  15. +0
    -0
      选修 To Learn More/第五节/Non-Autoregressive Sequence Generation.pdf
  16. +0
    -0
      选修 To Learn More/第五节/Pointer.pdf
  17. +0
    -0
      选修 To Learn More/第八节/dim reduction (v5).pdf
  18. +0
    -0
      选修 To Learn More/第八节/tsne (v2).pdf
  19. +0
    -0
      选修 To Learn More/第六节/FLOW.pdf
  20. BIN
      选修 To Learn More/第六节/GAN (v3).pdf
  21. +0
    -0
      选修 To Learn More/第六节/GANtheory (v2).pdf
  22. +0
    -0
      选修 To Learn More/第六节/VAE.pdf
  23. +0
    -0
      选修 To Learn More/第六节/WGAN (v2).pdf
  24. +0
    -0
      选修 To Learn More/第六节/fGAN.pdf
  25. +0
    -0
      选修 To Learn More/第十三节/PPO (v3).pdf
  26. +0
    -0
      选修 To Learn More/第十三节/QLearning (v2).pdf
  27. BIN
      选修 To Learn More/第十二节/RL (v6).pdf
  28. +0
    -0
      选修 To Learn More/第十五节/Meta1 (v6).pdf
  29. +0
    -0
      选修 To Learn More/第十五节/Meta2 (v4).pdf
  30. +0
    -0
      选修 To Learn More/第十节/Adversarial Attack.pptx
  31. +0
    -0
      选修 To Learn More/第四节/GNN.pdf
  32. +0
    -0
      选修 To Learn More/第四节/RNN.pdf
  33. +0
    -0
      选修 To Learn More/第四节/word2vec (v3).pdf

+ 2337
- 0
2022 ML/05 Sequence to sequence/HW05.ipynb
File diff suppressed because it is too large
View File


BIN
2022 ML/05 Sequence to sequence/HW05.pdf View File


BIN
2022 ML/05 Sequence to sequence/xformer (v6).pdf View File


+ 3
- 2
README.md View File

@@ -17,7 +17,6 @@ UP将2021&2022所有作业的数据资料整理打包好了,由于文件太大

ppt/pdf支持直链下载。
```
[![BILIBILI](https://raw.githubusercontent.com/Fafa-DL/readme-data/main/gzh.jpg)](https://space.bilibili.com/46880349)

## 更新日志

@@ -41,6 +40,7 @@ ppt/pdf支持直链下载。
|2022/02/25|更新Lecture 2:What to do if my network fails to train补充内容与HW2|
|2022/03/05|更新Lecture 3:Images input,HW3|
|2022/03/13|更新Lecture 4 Sequence as input,HW4<br>UP将2021&2022所有作业的数据资料整理打包好放在公众号【啥都会一点的研究生】|
|2022/03/18|更新Lecture 5 Sequence to sequence,HW5,相应Data放在公众号维护的网盘中|

****
@@ -68,5 +68,6 @@ ppt/pdf支持直链下载。
|Lecture 2|[(一)局部最小值 (local minima) 与鞍点 (saddle point)](https://www.bilibili.com/video/BV1Wv411h7kN?p=19)<br/>[(二)批次 (batch) 与动量 (momentum)](https://www.bilibili.com/video/BV1Wv411h7kN?p=20)<br/>[(三)自动调整学习率 (Learning Rate)](https://www.bilibili.com/video/BV1Wv411h7kN?p=21)<br/>[(四)损失函数 (Loss) 也可能有影响](https://www.bilibili.com/video/BV1Wv411h7kN?p=22)|Video:<br/>[2022-再探宝可梦、数码宝贝分类器 — 浅谈机器学习原理](https://www.bilibili.com/video/BV1Wv411h7kN?p=23)<br/><br/>PDF:<br/>[Theory](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/theory%20(v7).pdf)|[Gradient Descent (Demo by AOE)](https://www.bilibili.com/video/BV1Wv411h7kN?p=24)<br/>[ Beyond Adam (part 1)](https://www.bilibili.com/video/BV1Wv411h7kN?p=26)<br/>[ Beyond Adam (part 2)](https://www.bilibili.com/video/BV1Wv411h7kN?p=27)|[Video](https://www.bilibili.com/video/BV1Wv411h7kN?p=28)<br/>[Slide](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/hw2_slides%202022.pdf)<br/>[Code](https://colab.research.google.com/drive/1hmTFJ8hdcnqRz_0oJSXjTGhZLVU-bS1a?usp=sharing)<br/>[Submission](https://www.kaggle.com/c/ml2022spring-hw2)|
|Lecture 3|[卷积神经网络CNN](https://www.bilibili.com/video/BV1Wv411h7kN?p=31)|Video:<br/>[为什么用了验证集还是过拟合](https://www.bilibili.com/video/BV1Wv411h7kN?p=32)<br/>[鱼与熊掌可以兼得的机器学习](https://www.bilibili.com/video/BV1Wv411h7kN?p=33)<br/><br/>PDF:<br/>[Validation](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/validation.pdf)<br/>[Why Deep](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/whydeep%20(v3).pdf)|[Spatial Transformer Layer](https://www.bilibili.com/video/BV1Wv411h7kN?p=34)|[Video](https://www.bilibili.com/video/BV1Wv411h7kN?p=35)<br/>[Slide](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/Machine%20Learning%20HW3%20-%20Image%20Classification.pdf)<br/>[Code](https://colab.research.google.com/drive/15hMu9YiYjE_6HY99UXon2vKGk2KwugWu)<br/>[Submission](https://www.kaggle.com/c/ml2022spring-hw3b)|
|Lecture 4|[自注意力机制(Self-attention)(上)](https://www.bilibili.com/video/BV1Wv411h7kN?p=41)<br/>[自注意力机制(Self-attention)(下)](https://www.bilibili.com/video/BV1Wv411h7kN?p=42)|Video:<br/>[None]<br/><br/>PDF:<br/>[None]|[RNN(part 1)](https://www.bilibili.com/video/BV1Wv411h7kN?p=40)<br/>[RNN(part 2)](https://www.bilibili.com/video/BV1Wv411h7kN?p=41)<br/>[GNN(part 1)](https://www.bilibili.com/video/BV1Wv411h7kN?p=42)<br/>[GNN(part 2)](https://www.bilibili.com/video/BV1Wv411h7kN?p=43)|[Video](https://www.bilibili.com/video/BV1Wv411h7kN?p=45)<br/>[Slide](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/Machine%20Learning%20HW4.pdf)<br/>[Code](https://colab.research.google.com/drive/1gC2Gojv9ov9MUQ1a1WDpVBD6FOcLZsog?usp=sharing)<br/>[Submission](https://www.kaggle.com/c/ml2022spring-hw4)|
|Lecture 5|[类神经网络训练不起来怎么办(五)批次标准化](https://www.bilibili.com/video/BV1Wv411h7kN?p=48)<br/>[Transformer(上)](https://www.bilibili.com/video/BV1Wv411h7kN?p=49)<br/>[Transformer(下)](https://www.bilibili.com/video/BV1Wv411h7kN?p=50)|Video:<br/>[None]<br/><br/>PDF:<br/>[xformer](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/xformer%20(v8).pdf)|[NAT model](https://www.bilibili.com/video/BV1Wv411h7kN?p=51)<br/>[Pointer network](https://www.bilibili.com/video/BV1Wv411h7kN?p=52)|[Video]<br/>[Slide](https://speech.ee.ntu.edu.tw/~hylee/ml/ml2022-course-data/HW05.pdf)<br/>[Code](https://colab.research.google.com/drive/1Tlyk2vCBQ8ZCuDQcCSEWTLzr1_xYF9CL#scrollTo=Le4RFWXxjmm0)<br/>[Submission]|
****
[![BILIBILI](https://raw.githubusercontent.com/Fafa-DL/readme-data/main/gzh.jpg)](https://space.bilibili.com/46880349)

选修 To Learn More/BP.pdf → 选修 To Learn More/第一节/BP.pdf View File


选修 To Learn More/Classification.pdf → 选修 To Learn More/第一节/Classification.pdf View File


选修 To Learn More/DL.pdf → 选修 To Learn More/第一节/DL.pdf View File


选修 To Learn More/Logistic Regression.pdf → 选修 To Learn More/第一节/Logistic Regression.pdf View File


选修 To Learn More/Regression.pdf → 选修 To Learn More/第一节/Regression.pdf View File


选修 To Learn More/BERT train (v8).pdf → 选修 To Learn More/第七节/BERT train (v8).pdf View File


选修 To Learn More/GPT3 (v6).pdf → 选修 To Learn More/第七节/GPT3 (v6).pdf View File


选修 To Learn More/Multi (v2).pdf → 选修 To Learn More/第七节/Multi (v2).pdf View File


BIN
选修 To Learn More/第三节/Special Structure (v6).pdf View File


选修 To Learn More/Optimization.pdf → 选修 To Learn More/第二节/Optimization.pdf View File


选修 To Learn More/Non-Autoregressive Sequence Generation.pdf → 选修 To Learn More/第五节/Non-Autoregressive Sequence Generation.pdf View File


选修 To Learn More/Pointer.pdf → 选修 To Learn More/第五节/Pointer.pdf View File


选修 To Learn More/dim reduction (v5).pdf → 选修 To Learn More/第八节/dim reduction (v5).pdf View File


选修 To Learn More/tsne (v2).pdf → 选修 To Learn More/第八节/tsne (v2).pdf View File


选修 To Learn More/FLOW.pdf → 选修 To Learn More/第六节/FLOW.pdf View File


BIN
选修 To Learn More/第六节/GAN (v3).pdf View File


选修 To Learn More/GANtheory (v2).pdf → 选修 To Learn More/第六节/GANtheory (v2).pdf View File


选修 To Learn More/VAE.pdf → 选修 To Learn More/第六节/VAE.pdf View File


选修 To Learn More/WGAN (v2).pdf → 选修 To Learn More/第六节/WGAN (v2).pdf View File


选修 To Learn More/fGAN.pdf → 选修 To Learn More/第六节/fGAN.pdf View File


选修 To Learn More/PPO (v3).pdf → 选修 To Learn More/第十三节/PPO (v3).pdf View File


选修 To Learn More/QLearning (v2).pdf → 选修 To Learn More/第十三节/QLearning (v2).pdf View File


BIN
选修 To Learn More/第十二节/RL (v6).pdf View File


选修 To Learn More/Meta1 (v6).pdf → 选修 To Learn More/第十五节/Meta1 (v6).pdf View File


选修 To Learn More/Meta2 (v4).pdf → 选修 To Learn More/第十五节/Meta2 (v4).pdf View File


选修 To Learn More/Adversarial Attack.pptx → 选修 To Learn More/第十节/Adversarial Attack.pptx View File


选修 To Learn More/GNN.pdf → 选修 To Learn More/第四节/GNN.pdf View File


选修 To Learn More/RNN.pdf → 选修 To Learn More/第四节/RNN.pdf View File


选修 To Learn More/word2vec (v3).pdf → 选修 To Learn More/第四节/word2vec (v3).pdf View File


Loading…
Cancel
Save