Mingbao Lin 林明宝

你的问题不属于我们代码的bug

Hi, Thanks for your interest. In our CVPR version, we haven't prune the network for detection. However, we have observed the discovered rank property using YOLOv3. I think any task...

Zero rank denotes a zero matrix (feature map), which implies a zero filter that should be removed. I think ReLU layer is very important in this paper since it zeroizes...

In other words, our discovery might be limited to networks equipped with ReLU layers.

It has been illustrated in readme.md

We do not perform these models currently.

Thanks for your interest. I have just checked it on my server and everything goes smooth. So, please check the environment of your workstation, or have your correctly downloaded the...

> I have another problem, when I test the accuracy of pretrained resnet-50 on imagenet you provided, but the accuracy is just 0.0899, it's too low? Make sure you are...

> Thanks for your quick reply! > I run your code just change the pretrained model path and dataset path, my imagenet testset was downloaded from official website named 'ILSVRC2012_img_val.tar'(size:...

> I have found my mistake: when I untar my validation dataset from official website(6.3GB) , just a single folder, under the folder just many images, it's different from train...