nmaac / acon Goto Github PK

您好,我是东北大学的在读硕士,做的显著目标检测方向.
我看了您在CVPR2021发的这篇关于ACON激活函数的文章,想尝试在显著目标检测使用ACON,但是网站里只提供了ResNet的预训练模型,所以想请问下您那边方便提供下VGG的预训练模型吗?
十分感谢.

我把网络relu全换为为AconC后发现速度慢了一倍，是因为没优化还是计算量增大了？

感谢您的研究工作，请问论文中 Fig 4是将1.2x和-0.8x分别作为p1x,p2x带入ACON-C函数运算后得到的图像吗，我在计算的时候绘图出现了问题。

看了源代码，meta-acon要自动学习self.p，self.q，以及生成β的conv每层权重？

Hello,my networks is basd on conv1d,could I use the acon?

您好，我这两天在backbone是resnet50的centerNet网络中，，参照MetaACON中的程序，在resnet50 bn2后也加了一行self.acon=MetaAconC(planes),， forward中也将out = self.relu(out)改为：out = self.acon(out)，，，训练的时候bach_size设为16，使用了nms，学习率0.001，权值衰减是0.0005，然后预训练权重分别加载了res50.acon.pth和res50.metaacon.pth，用的voc数据集进行的训练、验证，最后结果map特别小，相比原来的centernet网络特别小。
我不太懂是哪里出了问题，或者是不是不能直接用你提供的预训练权值文件？我弄了一天多还是不行。所以来叨扰大佬了。谢谢。

hello! i want to use it in Hyperspectral images.pleace teach me how to use acon in 3d datasets?

ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 16, 1, 1])
会在这一行，beta = self.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(x.mean(dim=2, keepdims=True).mean(dim=3, keepdims=True))))))
会莫名其妙出现这个错误，然后早dataloader中使用了drop_last=True仍然没用

Hi, thanks for your amazing work!
How to use ACON in the pre-trained model? Can I directly replace all activation functions in the ImageNet pre-trained network with ACON, and then finetune it in the downstream task?

Hi, nmaac! 非常感谢您的工作，请问调用AconC的参数是输出图像的channel吗？

Hi, Thanks for your nice work.

From the paper, the improvements on ImageNet are obvious, which is attractive to me.
Recently, I conduct experiments on CIFAR-100 with ResNet-18 and the meta-ACON. However, the performance is not satisfactory.
Here, I want to know if you can provide some results on CIFAR-100 for reference.

Very thanks.

Aconc激活函数有tensorflow版本的吗？

MetaAconC:
self.fc1 = nn.Conv2d(width, max(r,width//r), kernel_size=1, stride=1, bias=True)
self.bn1 = nn.BatchNorm2d(max(r,width//r))
self.fc2 = nn.Conv2d(max(r,width//r), width, kernel_size=1, stride=1, bias=True)
self.bn2 = nn.BatchNorm2d(width)

It should be “nn.Conv2d(width, max(r,width//r), kernel_size=1, stride=1, bias=False)”？

How do I apply meta-ACON to the network structure of other tasks？for example super-resolution。Could you give me some advice？