Commit 56c7064

committed

finish nn

1 parent a410d6a commit 56c7064Copy full SHA for 56c7064

File tree

3 files changed

+825

-183

lines changed

README.md
chapter3_NN
- linear-regression-gradient-descend.ipynb
- nn-sequential-module.ipynb

3 files changed

+825

-183

lines changed

`‎README.md`

Lines changed: 26 additions & 41 deletions

Original file line number	Diff line number	Diff line change
`@@ -26,19 +26,19 @@ Learn Deep Learning with PyTorch`
`26`	`26`	`- Chapter 3: 神经网络`
`27`	`27`	`- [线性模型与梯度下降](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/linear-regression-gradient-descend.ipynb)`
`28`	`28`	`- [Logistic 回归与优化器](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/logistic-regression/logistic-regression.ipynb)`
`29`		`- - Module和Sequential`
`30`		`- - 模型保存和读取`
`31`		`- - 自定义参数的初始化`
	`29`	`+ - [多层神经网络,Sequential 和 Module](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/nn-sequential-module.ipynb)`
	`30`	`+ - 参数初始化方法`
`32`	`31`	`- 优化算法`
`33`		`- - 多层神经网络`
`34`	`32`
`35`	`33`	`- Chapter 4: 卷积神经网络`
`36`		`- - 从0开始手动搭建卷积网络`
`37`		`- - 批标准化`
	`34`	`+ - 卷积网络介绍以及 PyTorch 中的卷积模块`
`38`	`35`	`- 使用重复元素的深度网络,VGG`
`39`	`36`	`- 更加丰富化结构的网络,GoogLeNet`
`40`	`37`	`- 深度残差网络,ResNet`
`41`	`38`	`- 稠密连接的卷积网络,DenseNet`
	`39`	`+ - 更好的训练卷积网络:数据增强、批标准化、dropout和正则化方法`
	`40`	`+ - 灵活的数据读取介绍`
	`41`	`+ - Fine-tuning: 通过微调进行迁移学习`
`42`	`42`
`43`	`43`	`- Chapter 5: 循环神经网络`
`44`	`44`	`- LSTM和GRU`
`@@ -47,48 +47,33 @@ Learn Deep Learning with PyTorch`
`47`	`47`	`- Word Embedding和N-Gram模型`
`48`	`48`	`- Seq-LSTM做词性预测`
`49`	`49`
`50`		`-- Chapter 6: PyTorch高级`
`51`		`- - tensorboard可视化`
`52`		`- - 各种优化算法`
`53`		`- - autograd.function的介绍`
`54`		`- - 数据并行和多GPU`
`55`		`- - PyTorch的分布式应用`
`56`		`- - 使用ONNX转化为Caffe2模型`
	`50`	`+- Chapter 6: 生成对抗网络`
	`51`	`+ - 自动编码器`
	`52`	`+ - 变分自动编码器`
	`53`	`+ - 生成对抗网络的介绍`
	`54`	`+ - 深度卷积对抗网络(DCGANs)`
	`55`	`+`
	`56`	`+- Chapter 7: PyTorch高级`
	`57`	`+ - tensorboard 可视化`
	`58`	`+ - 优化算法`
	`59`	`+ - autograd.function 的介绍`
	`60`	`+ - 数据并行和多 GPU`
	`61`	`+ - PyTorch 的分布式应用`
	`62`	`+ - 使用 ONNX 转化为 Caffe2 模型`
	`63`	`+ - PyTorch 写 C 扩展`
`57`	`64`
`58`	`65`	`### part2: 深度学习的应用`
`59`		`-- Chapter 7: 计算机视觉`
	`66`	`+- Chapter 8: 计算机视觉`
`60`	`67`	`- 图像增强的方法`
`61`	`68`	`- Fine-tuning: 通过微调进行迁移学习`
`62`		`- - 语义分割: 通过FCN实现像素级别的分类`
`63`		`- - 使用卷积网络进行目标检测`
`64`		`- - 使用triplet loss进行人脸识别`
	`69`	`+ - 语义分割: 通过 FCN 实现像素级别的分类`
`65`	`70`	`- Neural Transfer: 通过卷积网络实现风格迁移`
`66`	`71`	`- Deep Dream: 探索卷积网络眼中的世界`
`67`	`72`
`68`		`-- Chapter 8: 自然语言处理`
`69`		`- - char rnn实现文本生成`
`70`		`- - 联合卷积网络实现图片字幕`
`71`		`- - 使用rnn进行情感分析`
`72`		`- - seq2seq实现机器翻译`
`73`		`- - cnn+rnn+attention实现文本识别`
`74`		`- - Tree-lstm实现语义相关性分析`
`75`		`-`
`76`		`-### part3: 高级内容`
`77`		`-- Chapter 9: 生成对抗网络`
`78`		`- - 自动编码器`
`79`		`- - 变分自动编码器`
`80`		`- - 生成对抗网络的介绍`
`81`		`- - 深度卷积对抗网络(DCGANs)`
`82`		`- - Wasserstein-GANs`
`83`		`- - 条件生成对抗网络(Conditional`
`84`		`- GANs)`
`85`		`- - Pix2Pix`
`86`		`-`
`87`		`-- Chapter 10: 深度增强学习`
`88`		`- - 深度增强学习的介绍`
`89`		`- - Policy gradient`
`90`		`- - Actor-critic gradient`
`91`		`- - Deep Q-networks`
	`73`	`+- Chapter 9: 自然语言处理`
	`74`	`+ - char rnn 实现文本生成`
	`75`	`+ - seq2seq 实现机器翻译`
	`76`	`+ - cnn+rnn+attention 实现文本识别`
`92`	`77`
`93`	`78`	`## 一些别的资源`
`94`	`79`

0 commit comments

Comments

(0)

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit 56c7064

File tree

3 files changed

3 files changed

`‎README.md`

0 commit comments