OREPA_CVPR2022
OREPA_CVPR2022 copied to clipboard
JUGGHM
CVPR 2022 "Online Convolutional Re-parameterization"
Hi, I'm wondering if you've run into any issues with numerical stability or know what may be the cause. With normal RepVGG, I get differences as high as 4e-4 comparing...
作者你好，非常喜歡你這篇文章的idea，然後我現在是想extend到3D，但是我沒有太看懂在OREPA function裡的self.fre_init的作用，可以再解釋一下ma? 然後對於prior_tensor，我們能變成3D的嗎?
Hi, thanks for making your code public. It is really great work! I ran your code, and i think there is a minor error in your code. On line 217...
Proposition 1 A single-branch linear mapping, when re-parameterizing parts or all of it by over-two-layer multi-branch topologies, the entire end-to-end weight matrix will be differently optimized. If one layer of...
```python class OREPA_LargeConvBase(nn.Module): def __init__(self, in_channels, out_channels, kernel_size, stride, padding, dilation=1, groups=1, deploy=False, nonlinear=None): …… internal_channels = out_channels ``` 代码中为什么将3x3conv之间channel设置成与输出一致，而不是输出的倍数，进一步增加参数量，提升模型性能？
你好： 我看文章里OREPA和RepVGG结合时，是直接在conv_3*3上加OREPA， 而不是直接将conv_3*3/conv_1*1/identity三个分支换成OREPA的形式。请问这样做的是因为直接将conv_3*3/conv_1*1/identity三个分支换成OREPA进行训练效果不好么？
Hello, I'm very confused about accuracy of ResNet34. Specifically, I train ResNet34 many time, but accuracy of ResNet34 is about 74.40. I found that this paper and RepVGG both report...
Hi, thanks for your great work! I tried to reproduce the visualization of branch-level similarity of OREPA blocks, but the unexpected results emerged. Could you share details about it?
