lucidrains/vit-pytorch

Issues

• train navit nest 3d error when backward

  #332 opened 16 days ago by HaloTrouvaille
  5

• RegionViT - Local token embedding

  #330 opened 2 months ago by minhquoc0712
  1

• The Total params: and Params size (MB) of the model printed by summary are different from the bit_base model in timm library. Theoretically, the same settings should be the same. What is the reason?

  #329 opened 2 months ago by lucker26
  1

• MS-COCO training from Imagenet pretrained checkpoint

  #328 opened 2 months ago by prateekiiest
  1

• Multi-GPU training of NaViT model

  #322 opened 4 months ago by b5y
  1

• Weight Initialization

  #321 opened 5 months ago by simonaay
  0

• SimpleViT misleading summary

  #320 opened 5 months ago by asusdisciple
  0

• anyone knows why _freeze_stages() starts from block[0]?

  #319 opened 5 months ago by abc5z7
  0

• Questions about distill_loss

  #289 opened a year ago by haoren55555
  3

• Validation accuracy higher than training accuracy

  #311 opened 6 months ago by yoder460
  1

• Choice for reduced order model / latent space

  #315 opened 6 months ago by ramdhan1989
  0

• [MaxViT] Block/Grid Attention question

  #314 opened 6 months ago by sonderlau
  0

• Whether to include pre-trained models

  #307 opened 7 months ago by KawaiiAsh
  1

• Swin UNet

  #312 opened 6 months ago by sibi-venti
  0

• Why Remove PreNorm?

  #309 opened 6 months ago by tonyyunyang
  0

• Patch Embedding Design Choice?

  #310 opened 6 months ago by tonyyunyang
  0

• Cuda memory for 3D VIT

  #300 opened 8 months ago by JesseZZZZZ
  2

• Request for Pre-trained Weights for Vit

  #308 opened 7 months ago by ZSLsherly
  0

• Non-deterministic results based on group_max_seq_len in NaViT

  #306 opened 7 months ago by dempsey-ryan
  3

• CrossViT does not handle other than three channel images

  #304 opened 7 months ago by Yash-10
  2

• PyPi page markdown render

  #302 opened 7 months ago by soumya1729
  1

• A question with ViT 3d

  #298 opened 8 months ago by JesseZZZZZ
  0

• how to train

  #288 opened a year ago by lingxitong
  2

• Add implementation of LongVit

  #297 opened 9 months ago by jpfeil
  4

• Problems regarding training 3D Vision transformer : model does not converge

  #296 opened 10 months ago by Uljibuh
  0

• Multi-target Regression Question

  #295 opened 10 months ago by stethemJ
  0

• can we use CvT model for segmentation?

  #294 opened 10 months ago by HawkingRadiation42
  0

• Masking attention with batches

  #293 opened 10 months ago by ashrafflh
  0

• Question regarding 1d fft use

  #292 opened a year ago by chengengliu
  1

• Trouble loading ViT - Dino structure for channels>3?

  #291 opened a year ago by AgentM-GEG
  0

• Layernorm in Cross attention

  #287 opened a year ago by turtleman99
  4

• CvT with 1 channel input data

  #286 opened a year ago by tranlg99
  2

• Using vision transformers for different image resolutions

  #280 opened a year ago by Oussamab21
  1

• Not correctly understanding the Multi Head Attention part of the ViT implementation...

  #282 opened a year ago by JavierUrenaPhDProjects
  3

• Potential regression with PT 2.0 and CUDA 12.2/CuDNN 8.9.4

  #281 opened a year ago by roywei
  1

• Saving and loading model seems to be regressing to lower performance

  #278 opened a year ago by aperiamegh
  1

• vit_pytorch -> cross_vit.py(mistake)

  #279 opened a year ago by RufusRubin
  1

• structural 3D ViT

  #277 opened a year ago by aperiamegh
  4

• This ViT implementation as generative network

  #276 opened a year ago by MrCorsair3
  1

• TVM compilation failed on SimpleViT

  #275 opened a year ago by yangxin0926
  0

• Dimension issues in Masked Patch Prediction

  #269 opened a year ago by KananVyas
  4

• MAE Training

  #271 opened a year ago by mw9385
  0

• Is it possible to use the "Accessing Attention" of the vit-pytorch on the timm models?

  #270 opened a year ago by Shima-shoki
  0

• ViT for regression task such as Real Estate Price Prediction or Stock Exchange Datasets, any regression dataset.

  #259 opened a year ago by saifhassan
  7

• When running python train_cifar10.py, RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase.

  #267 opened 2 years ago by laserljy
  0

• ViVit pos encoding

  #266 opened 2 years ago by eyalmazuz
  2

• PyTorch 2.0 support

  #262 opened 2 years ago by kxzxvbk
  2

• Integrate Aim - an open-source experiment tracker

  #263 opened 2 years ago by tatyusha
  1

• Multi-head attention part on ViT

  #261 opened 2 years ago by andreYoo
  0

• Small Typo in CCT description

  #258 opened 2 years ago by DSARichard
  1