Skip to content
This repository has been archived by the owner on Jul 24, 2024. It is now read-only.

Commit

Permalink
fix
Browse files Browse the repository at this point in the history
Signed-off-by: Jianliang Shen <[email protected]>
  • Loading branch information
Jianliang-Shen committed Jun 30, 2024
1 parent 558441e commit a579742
Showing 1 changed file with 2 additions and 0 deletions.
2 changes: 2 additions & 0 deletions source/_posts/AGI-Startup.md
Original file line number Diff line number Diff line change
Expand Up @@ -152,6 +152,8 @@ PyTorch内部实现了神经网络中绝大多数的layer,这些layer都继承

图像nn包括,卷积层(Conv)、池化层(Pool),池化方式又分为平均池化(AvgPool)、最大值池化(MaxPool)、自适应池化(AdaptiveAvgPool)等。而卷积层除了常用的前向卷积之外,还有逆卷积(TransposeConv)。卷积神经网络的本质就是卷积层、池化层、激活层以及其他层的叠加。池化层可以看作是一种特殊的卷积层,其主要用于下采样,增加池化层可以在保留主要特征的同时降低参数量,从而一定程度上防止了过拟合。池化层没有可学习参数,它的weight是固定的。在`torch.nn`工具箱中封装好了各种池化层,常见的有最大池化(MaxPool)和平均池化(AvgPool)。

</details>

## 安装 Pytorch 1.6 环境(CPU版本)

https://pytorch.org/get-started/previous-versions/
Expand Down

0 comments on commit a579742

Please sign in to comment.