LTH14/mar

Issues

• Increasing token size

  #71 opened a month ago by jinnan-chen
  3

• Reproduce the results of Table 1

  #72 opened a month ago by Doraemonzzz
  1

• noise_schedule setting

  #70 opened a month ago by wangherr
  3

• Loss is nan, stopping training

  #69 opened a month ago by tau-yihouxiang
  3

• Doesn't work well in speech generation task.

  #35 opened 2 months ago by FacePoluke
  10

• Scaling diffusion MLP for learning higher dimensional VAE feature

  #55 opened 2 months ago by zythenoob
  7

• Mismatch state_dict when resuming from pre-trained MAR-B

  #68 opened a month ago by tau-yihouxiang
  2

• Comparison of results before and after using diff_loss

  #66 opened a month ago by shuowang666
  2

• Differences between Latent Space and Pixel Space

  #67 opened a month ago by OK932001
  1

• MAR used for image restoration

  #65 opened a month ago by chengliang0109
  1

• Request for Your Email Address

  #64 opened a month ago by ziwei-cui
  1

• How is 0.2325 calculated?

  #62 opened a month ago by shuowang666
  4

• About diffloss

  #63 opened a month ago by RohollahHS
  2

• image of the training loss change in TensorBoard

  #61 opened a month ago by wangherr
  1

• Why put a VAE to encode and decode outside the MAR module?

  #58 opened a month ago by DeepDuke
  6

• [Question] The inference speed of MAR and MAR-Diffusion

  #60 opened a month ago by EPIC-Lab-sjtu
  1

• Will mar work when img_size=64 or img_size=128?

  #59 opened a month ago by fengyang0317
  2

• mask generation for training and inference with MAR

  #57 opened 2 months ago by YOU-k
  1

• Clarification of speed

  #45 opened 2 months ago by zehongs
  6

• About VAE channels

  #56 opened 2 months ago by pokameng
  10

• Questions about causal methods

  #27 opened 3 months ago by Tom-zgt
  6

• About the positional encodings of diffusion MLP.

  #54 opened 2 months ago by whwjdqls
  2

• The influence of VAE feature dim

  #53 opened 2 months ago by Tom-zgt
  1

• About training with cached vae latents

  #52 opened 2 months ago by RohollahHS
  2

• Challenges in Memorizing Single or Few Images

  #49 opened 2 months ago by kifarid
  3

• About Encoder and Decoder in MAE

  #42 opened 2 months ago by RohollahHS
  2

• About Train

  #36 opened 2 months ago by pokameng
  31

• Faster training with fp16 or bf16

  #29 opened 3 months ago by shaochenze
  5

• Request for Causal AR Version Release

  #39 opened 2 months ago by aengusng8
  3

• Diffusion MLP initialization and training problem

  #46 opened 2 months ago by Kevin-thu
  1

• Reproducing the BASE model.

  #48 opened 2 months ago by cxxgtxy
  18

• model and training code for the AR variant

  #34 opened 2 months ago by MikeWangWZHL
  5

• Dataset used for training...

  #50 opened 2 months ago by Nobody-Zhang
  2

• Reconstruction loss in ELBO

  #51 opened 2 months ago by Paulmzr
  2

• Small Issue/error in the code

  #47 opened 2 months ago by niklasbubeck
  1

• Add HF integration to MAR

  #32 opened 2 months ago by jadechoghari
  14

• About masking ratio

  #43 opened 2 months ago by RohollahHS
  1

• Training Problems

  #44 opened 2 months ago by drx-code
  1

• Unable to calculate FID

  #41 opened 2 months ago by xbyym
  2

• Difference Between MAR and MAGE

  #30 opened 3 months ago by JeremyCJM
  5

• CFG for cross-entropy

  #38 opened 2 months ago by shaochenze
  2

• About Training Loss

  #37 opened 2 months ago by Ferry1231
  4

• Is Autoencoder ok?

  #33 opened 3 months ago by Ferry1231
  1

• The CFG strategy - linear. vs constant

  #31 opened 3 months ago by yuhuUSTC
  9

• generate images with arbitrary resolutions,

  #28 opened 3 months ago by Leiii-Cao
  4

• Inference details of an ablation experiment.

  #26 opened 3 months ago by tgxs002
  1

• VAE decoded as NaN in early stages of training

  #25 opened 3 months ago by xiazhi1
  2

• Why main_cache do not use flip augment?

  #24 opened 3 months ago by xiazhi1
  2

• How should inference be performed when using VQ-16 (discrete)? During decoding, should we use the AR output for VQ and then decode?

  #23 opened 3 months ago by Tom-zgt
  2

• About the mask schedule during training

  #22 opened 3 months ago by zythenoob
  4