dome272/VQGAN-pytorch
Pytorch implementation of VQGAN (Taming Transformers for High-Resolution Image Synthesis) (https://arxiv.org/pdf/2012.09841.pdf)
PythonMIT
Issues
- 8
Missing projection of Attention values in non-local block and incorrect position of the Upsampling block in the decoder
#23 opened by hgupta01 - 17
The issue with training results.
#10 opened by JunZhan2000 - 0
Bug fixed version.
#25 opened by aa1234241 - 0
LPIPS loss can produce negative values
#24 opened by jklimmek - 0
I try to train the Transformers, and I get this errors: RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.
#21 opened by F4k3r22 - 0
- 1
Environment Setup
#19 opened by nooralahzadeh - 0
This is a great Repo
#18 opened by wdayang - 0
- 1
Grid artifacts in vqgan
#15 opened by darthjaja6 - 0
- 0
download vqgan_last_ckpt.pt
#14 opened by Gannn12138 - 3
add label conditions problem
#11 opened by zhaoyk1986 - 0
Does transformer forward only predict the last index, z_i? If so, how to obtain the sequence of z_q?
#9 opened by bruce2233 - 1
- 0
- 0
conditional image generation
#5 opened by Ivvvvvvvvvvy - 2
Conditional image generation problem
#6 opened by Ivvvvvvvvvvy - 2
real image in save results looks strange
#4 opened by uyo9ko - 1
- 1
Black border
#1 opened by gogobd