PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 训练
Posted SpikeKing
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 训练相关的知识,希望对你有一定的参考价值。
欢迎关注我的CSDN:https://blog.csdn.net/caroline_wendy
本文地址:https://blog.csdn.net/caroline_wendy/article/details/128243621
Paper:A ConvNet for the 2020s,FAIR,Berkeley
Code:https://github.com/facebookresearch/ConvNeXt
Isotropic ConvNeXt vs. ViT
-
ConvNeXt去掉下采样层,就是与ViT各项同性
-
ViT来源于DeiT -> ViT-S/B,MAE -> ViT-L
-
warm-up的epoch增多
源码:
class ConvNeXtIsotropic(nn.Module)
以上是关于PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 训练的主要内容,如果未能解决你的问题,请参考以下文章
PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 训练
PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 训练
PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 网络
PyTorch笔记 - A ConvNet for the 2020s (ConvNeXt) 网络