Liming Jiang

Thanks for the great suggestion. I did not add this file because users may have different hardware environment, such as the GPU Driver, which may need compatible CUDA toolkit and...

You may refer to this [issue](https://github.com/NVlabs/stylegan2-ada-pytorch/issues/105) in the original [stylegan2-ada-pytorch](https://github.com/NVlabs/stylegan2-ada-pytorch) repo and the mentioned issue in it to generate images using CPU. It is not recommended to train the StyleGAN-based...

It seems working on my side. How many GPUs do you have?

For example, when you modify the `--gpu_ids 0` [here](https://github.com/EndlessSora/TSIT/blob/master/test_scripts/ast_summer2winteryosemite.sh#L18) to `--gpu_ids 1`, will it cause an error?

Thanks for the suggestion. May update the name later.

Have found some solutions: 1. https://askubuntu.com/questions/987344/unable-to-unzip-a-file-with-the-ubuntu-terminal 2. https://askubuntu.com/questions/54904/unzip-error-end-of-central-directory-signature-not-found

> May I ask whether the stacked hourglass of the structure extraction module in dfvae will be trained together, that is, whether the parameters of the stacked hourglass will undergo...

Thanks for your interest in our work. You might encounter this warning when using new PyTorch version. Please ignore this warning for the moment.

> https://github.com/EndlessSora/focal-frequency-loss/blob/5c34c2cb03bb9b26fa917fd9f032c009599290a5/focal_frequency_loss/focal_frequency_loss.py#L89 Thanks for your interest. We performed the square operation in Line 89. We use the squared Euclidean distance. You can consider printing each tensor out to understand the...

No. You can regard the square root of `freq_distance` in Line 90 as the Euclidean distance. We apply the squared Euclidean distance to ensure a smooth gradient.