KinWaiCheuk

Is it possible for you to write a dummy code, for example creating a random batch of input via `x = torch.randn(16, 44100)`, where 16 is the batch size, or...

It seems to be a GPU bottleneck to me. Can you check when you use torch.istft, what is the GPU utility? Does it reaches 100%? If torch.istft does not reach...

> I'm using GeForce RTX 2080, the same code everything is fine. If this problem only appears when using RTX 3080, but not on RTX 2080, then it seems this...

Currently this problem happens in RTX 3090 and RTX 3080, but not on RTX 2080? I am not sure if there is something in the RTX 3000 series that is...

I did a quick search, and I found that other people are facing the same problem using PyTorch on a RTX 3090. Since nnAudio is implemented using convolutional layers from...

I am afraid that there is nothing we can do at the moment. We might need to wait for NVIDIA to release new version of CUDA or cudnn. I hope...

After studying the code line by line, I also have this doubt. I was wondering if the element-wise product used in this code is another type of attention mechanism. But...

I realized that this issue is the duplicate of #10

> nnAudios implmentation seems to be onnx exportable. We might want to check the differences... [KinWaiCheuk/nnAudio#23 (comment)](https://github.com/KinWaiCheuk/nnAudio/issues/23#issuecomment-768954731) I used two nn.Conv1d to write my STFT class in nnAudio. One Conv1d...