Alpha-VL/ConvMAE

Issues

• Why not use the masked transformers directly in the first two stages?

  #36 opened 8 months ago by xwan0527
  0

• Question about VideoConvMAE

  #17 opened 2 years ago by JiaxinZhuang
  3

• When will you update the MR-MCMAE model?

  #35 opened 9 months ago by Gpysophie
  0

• Visualization VIT feature

  #34 opened a year ago by kimsekeun
  0

• about the training loss

  #33 opened a year ago by JasonWong30
  1

• Model Settings and checkpoint not match

  #32 opened a year ago by z-jiaming
  0

• Questions about convmae-v2

  #31 opened a year ago by DDKdakuan
  0

• full checkpoint

  #30 opened 2 years ago by LiewFeng
  0

• about pretrain model convmae.pth

  #12 opened 2 years ago by lywang76
  5

• hi，i need help

  #29 opened 2 years ago by wudizuixiaosa
  1

• image unpatchify related problems

  #28 opened 2 years ago by pengjuRen99
  2

• why still can't find the paper or details of ConvMAE-v2?

  #27 opened 2 years ago by smarengin
  2

• How can i train 200 epoches for DET ?

  #26 opened 2 years ago by ross-Hr
  15

• refactor hard coded numbers for more control over parameters (MaskedAutoencoderConvViT)

  #25 opened 2 years ago by DanTaranis
  1

• Hi

  #20 opened 2 years ago by ross-Hr
  1

• How long will the the pretraining stage takes in V100?

  #24 opened 2 years ago by russellllaputa
  0

• Total memory consumption for training with 32 batch size.

  #23 opened 2 years ago by IamYourAlpha
  6

• Train on

  #21 opened 2 years ago by zqyJason
  2

• mask convolution

  #7 opened 3 years ago by cathylao
  3

• Doubts about masking strategy

  #22 opened 2 years ago by aichifandefan
  0

• Question about ConvMAE-v2

  #19 opened 2 years ago by z-jiaming
  4

• Running pretrained convvit on larger image sizes

  #18 opened 2 years ago by JonathanBhimaniBurrows
  1

• When to release other model weights rather than base model?

  #16 opened 2 years ago by ewrfcas
  6

• How to pretrain CNN backbones like ResNet and EfficientNet?

  #15 opened 2 years ago by sarmientoj24
  1

• output of FastconvMAE

  #14 opened 2 years ago by lywang76
  2

• Can you give us guide how to run your downloaded convmae.pth model to finetune?

  #13 opened 2 years ago by lywang76
  2

• Hello, how to finetune own datasets

  #11 opened 2 years ago by wudizuixiaosa
  6

• ImageNet Evaluation

  #9 opened 3 years ago by SheldonHS
  6

• The results of LIN pretrained with ConvMAE-Based for 200 epochs.

  #10 opened 3 years ago by zhangxinyu-xyz
  0

• Pretraining implementation

  #6 opened 3 years ago by hao-pt
  9

• have you try pure convolution network?

  #3 opened 3 years ago by Dongshengjiang
  5

• Will the code of MAE pretrain be updated recently?

  #2 opened 3 years ago by KevinLongkk
  3

• Time required to train one epoch.

  #5 opened 3 years ago by charlesCXK
  6

• ConvMAE flops for classification ?

  #8 opened 3 years ago by sunsmarterjie
  2

• how many gpus and which type of gpu are needed for pretraining, down-stream task finetuning?

  #4 opened 3 years ago by wenmengzhou
  5

• When to update the code for object detection？

  #1 opened 3 years ago by lqz72
  2