Bowen Li

icon indicating a website link [email protected]

icon company Harbin Institute of Technology, Shenzhen icon location Shenzhen, Guangdong, China icon location MA.Eng Student in Harbin Institute of Technology, Shenzhen, China.

Results 3 comments of Bowen Li

adder 反卷积

你好，我按照卷积和反卷积的关系，将反卷积中的卷积操作替换成了adder，但是可视化中间的特征图为啥会出现比较严重的块效应。这是什么原因呢？感谢您的解答！ ![image](https://user-images.githubusercontent.com/68136466/123281799-8de0b500-d53c-11eb-9df4-aac7c78cb2c0.png)

gaussian1 = torch.distributions.laplace.Laplace(mean1, sigma1) ？

I also want to know.

gaussian1 = torch.distributions.laplace.Laplace(mean1, sigma1) ？

Thanks for your reply! I also test Gaussian and Laplace at the same time, and they have the similar performance as you said. I think this is because three Gaussian...