-
Hi Chris,
Thanks for sharing the implementation. I want to use your rotation trick on some other datasets but I'm not sure what might be the best practice to modify the hyper-parameter accordingly. F…
-
Do we need target images to run test? I want to upscale my own data.
Also please fix Basicsr:
`python inference_realsrt.py --input input --output results --model_path experiments/pretrained_models…
-
Instead of using frozen VQGAN tokens, we also train/finetune VQGAN end-to-end
-
Hi! I'm receiving an error while trying to train the model on a custom dataset in colab. I created test.txt and training.txt and modified the paths at custom_vqgan.yaml. I don't know if the images dim…
-
Hello,
Where can the ckpt file of ocr-vqgan be downloaded from?
-
I'm playing around trying to run this on my 12 GB gpu. It just barely fits, however the VQGAN takes up enough memory to overflow. It would be nice if I could use another GPU for vqgan and maybe tokeni…
-
Hello, I reproduced your method in our own dataset, but the loss didn't decrease. Is pre-trained VQGAN necessary for the training of specific datasets？
-
Hi,
I was training on my own data set with taming's VQ-GAN.
There's some error with the dimension with the vae model
![image](https://user-images.githubusercontent.com/85055246/124013852-68a8e680-…
-
Looks like the dedicated host for the VQGAN is getting hit pretty hard right now.
https://github.com/CompVis/taming-transformers/issues/53
-
# List of VQGAN+CLIP Implementations
Here's a list of all VQGAN implementations I found on the internet.
[https://ljvmiranda921.github.io/notebook/2021/08/11/vqgan-list/](https://ljvmiranda921.githu…