focal-frequency-loss

focal-frequency-loss copied to clipboard

The warning code is【C:\Users\PC\.conda\envs\paGAN\lib\site-packages\torch\autograd\__init__.py:173: UserWarning: Casting complex values to real discards the imaginary part (Triggered internally at C:\cb\pytorch_1000000000000\work\aten\src\ATen\native\Copy.cpp:239.) Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass】...

hejs9603

Could you please provide the method of reproducing other methods in the paper? This repo Only provide vanilla AE .

peylnog

Train problem

2

Hi, thanks for your work. I got a problem when I use the focal frequency loss for training. This sentence appears above the log file(but the network is still on...

wwang0107

About the calculation of distance

5

https://github.com/EndlessSora/focal-frequency-loss/blob/5c34c2cb03bb9b26fa917fd9f032c009599290a5/focal_frequency_loss/focal_frequency_loss.py#L90 Thanks for your work I want to know the calculation of the distance. Euclidean distance needs square operation and why here just "tmp[...,0],tmp[...,1]" what does it mean?

IItaly

Thank you for a nice and handy implementation. I would like to ask you to provide some kind of stylegan2 training config, e.g. like [this one](https://github.com/XPixelGroup/BasicSR/blob/master/options/train/StyleGAN/train_StyleGAN2_256_Cmul2_FFHQ.yml) so it would be...

dearkafka

LiHuan1216

Hi, thank you for providing the code. It was really helpful. One thing I am curious about is that when training VAE, unlike VanillaAE, the KL loss weight can affect...

wubowen416

Why do we clamp [here](https://github.com/EndlessSora/focal-frequency-loss/blob/c0d94cad8876d750cb9034f1b846035b0436decd/focal_frequency_loss/focal_frequency_loss.py#L81)?

schmidt-ai

About use

1

Hello, will the gradient of loss be calculated when ffl is used as loss? I try to put ffl on cuda, but it is always a scalar, and mse is...

zy-charon

# Welcome update to OpenMMLab 2.0 I am Vansin, the technical operator of OpenMMLab. In September of last year, we announced the release of OpenMMLab 2.0 at the World Artificial...

vansin