yechanp

Hello , it might be too late. I found that learning rate is decreasing too fast. Change gamma 0.1 to big number such as 0.9. It will be better

I think it is okay. It should be kept as conv3d. but it actually performs like conv2d because one of kernel size is 1.

Hello, I think I found the answer. The problem occurred due to name mismatch. You need to change the output so file name in setup.py: change ``` package = Extension('bbox',...

I also tried this. But, in my opinion, this is not the matter of differentiability. I also tried SiLU = (x*sigmoid) activation function which is smooth, but I failed to...

After a few experiments yesterday, I found that initial values of the model parameters are important. If the activation function changes, stds of initializer should be adjusted. In my setting,...

Have you succeeded in deeper networks? In my experiments, loss is decreasing, but it stops at some points and does not converge.

It works when I changed the activation functions. It is hard for ReLU, but it works for other invertible activations