NVlabs/edm

Issues

• Question about parameter tuning

  #21 opened a year ago by xiong-creator
  2

• Error in stride shape

  #30 opened 3 months ago by Mxdy21
  1

• Backward incompatible changes in huggingface_hub

  #29 opened 4 months ago by rafalkarczewski
  0

• Using EDM for end-to-end training of latent diffusion models

  #28 opened 7 months ago by pandekan
  0

• Question about Fourier Embedding and EDM Loss calculation

  #27 opened 8 months ago by zhengyu-su
  1

• Is it possible to also share the code for Figure 1b (ideal denoiser outputs) in the paper

  #26 opened 10 months ago by tien-th
  0

• ERROR when using multi-gpu training

  #18 opened a year ago by tsWen0309
  7

• AssertionError: Torch not compiled with CUDA enabled

  #25 opened a year ago by pierrelouisdelx
  0

• About hyper-parameters selection

  #24 opened a year ago by JincanDeng
  0

• Different Phases of Sampling

  #23 opened a year ago by coopersvajda
  0

• Can not reproduct the result in Table 2.

  #16 opened a year ago by Zyriix
  5

• On checkpoint for 256*256 images

  #22 opened a year ago by mengxiangming
  0

• Question about Augmentation

  #17 opened a year ago by hojaeklee
  1

• How to select options shown in Table2 for training VP/VE models

  #6 opened 2 years ago by MingyuKim87
  1

• VP checkpoints are trained using VE-scaling?

  #13 opened 2 years ago by huanranchen
  2

• params[x] in this process with sizes [x, x] appears not to match sizes of the same param in process 0

  #11 opened 2 years ago by forever208
  2

• Could you please provide us imagenet-uncond checkpoint?

  #15 opened a year ago by huanranchen
  0

• About the Checkpoint

  #14 opened a year ago by SYZhang0805
  0

• Is it possible to also share the code for Figure 3 in the paper

  #12 opened 2 years ago by yuanzhi-zhu
  1

• Is the original loss of VP/VE implemented (i.e. network predict epsilon)?

  #9 opened 2 years ago by forever208
  1

• DataLoader worker (pid xxxx) is killed by signal

  #7 opened 2 years ago by Zhendong-Wang
  1

• .

  #8 opened 2 years ago by philgzl
  0

• Wrong beta_D coefficient for VP during sampling?

  #5 opened 2 years ago by qiyan98
  1

• Deterministic sampling and sampling steps

  #4 opened 2 years ago by FutureXiang
  0

• Zero initialization of convolutions

  #3 opened 2 years ago by nicolas-dufour
  1

• Possible error in the up-sampling function

  #2 opened 2 years ago by Msadat97
  1

• Question about the parameterizations of VP-SDE

  #1 opened 2 years ago by FutureXiang
  1