finish cnn architecture

Author: L1aoXingyu

README.md

@@ -28,23 +28,14 @@ Learn Deep Learning with PyTorch
     - [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)
     - [深度神经网络](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/deep-nn.ipynb)
     - [参数初始化方法](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/param_initialize.ipynb)
-    - 优化算法
-        - [SGD](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/sgd.ipynb)
-        - [动量法](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/momentum.ipynb)
-        - [Adagrad](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adagrad.ipynb)
-        - [RMSProp](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/rmsprop.ipynb)
-        - [Adadelta](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adadelta.ipynb)
-        - [Adam](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adam.ipynb)
-
+ 
 - Chapter 4: 卷积神经网络
-    - PyTorch 中的卷积模块
-    - 使用重复元素的深度网络，VGG
-    - 更加丰富化结构的网络，GoogLeNet
-    - 深度残差网络，ResNet
-    - 稠密连接的卷积网络，DenseNet
-    - 更好的训练卷积网络：数据增强、批标准化、dropout和正则化方法
-    - 灵活的数据读取介绍
-    - Fine-tuning: 通过微调进行迁移学习
+    - [PyTorch 中的卷积模块](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter4_CNN/basic_conv.ipynb)
+    - [使用重复元素的深度网络，VGG](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_CNN/vgg.ipynb)
+    - [更加丰富化结构的网络，GoogLeNet](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_CNN/googlenet.ipynb)
+    - [深度残差网络，ResNet](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_CNN/resnet.ipynb)
+    - [稠密连接的卷积网络，DenseNet](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_CNN/densenet.ipynb)
+    - 更好的训练卷积网络：数据增强、批标准化、dropout、正则化方法以及学习率衰减
 
 - Chapter 5: 循环神经网络
     - LSTM 和 GRU
@@ -62,6 +53,13 @@ Learn Deep Learning with PyTorch
 - Chapter 7: PyTorch高级
     - tensorboard 可视化
     - 优化算法
+        - [SGD](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/sgd.ipynb)
+        - [动量法](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/momentum.ipynb)
+        - [Adagrad](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adagrad.ipynb)
+        - [RMSProp](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/rmsprop.ipynb)
+        - [Adadelta](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adadelta.ipynb)
+        - [Adam](https://github.com/SherlockLiao/code-of-learn-deep-learning-with-pytorch/blob/master/chapter3_NN/optimizer/adam.ipynb)
+    - 灵活的数据读取介绍
     - autograd.function 的介绍
     - 数据并行和多 GPU
     - PyTorch 的分布式应用
@@ -70,14 +68,14 @@ Learn Deep Learning with PyTorch
 
 ### part2: 深度学习的应用
 - Chapter 8: 计算机视觉
-    - 图像增强的方法
     - Fine-tuning: 通过微调进行迁移学习
     - 语义分割: 通过 FCN 实现像素级别的分类
     - Neural Transfer: 通过卷积网络实现风格迁移
     - Deep Dream: 探索卷积网络眼中的世界
 
 - Chapter 9: 自然语言处理
     - char rnn 实现文本生成
+    - Image Caption: 实现图片字幕生成
     - seq2seq 实现机器翻译
     - cnn+rnn+attention 实现文本识别