Ma-Lab-Berkeley/CRATE

Issues

• 为什么我做分割效果和论文说的不一样，还没有vit好

  #29 opened 6 months ago by YanGe1105
  1

• computing rate reduction in CRATE

  #28 opened 6 months ago by heeseokjung
  1

• the final task-specific architecture is classification head as the paper informed，but why you show demo in segmentation task？

  #25 opened 6 months ago by sanwei111
  1

• can i implement this idea into some open source llm？such as qwen

  #26 opened 6 months ago by sanwei111
  1

• where is the inference code？

  #27 opened 6 months ago by sanwei111
  1

• How should I extract the features from unclassified data?

  #24 opened 8 months ago by zhrli
  0

• more pretrained weights

  #8 opened a year ago by idonashino
  1

• Experiment on Diffusion Models

  #15 opened 10 months ago by yuzheyao22
  1

• Confusion about the Code Implementation

  #18 opened 10 months ago by HenryLau7
  1

• Is there any example for language?

  #12 opened 10 months ago by subercui
  1

• How CREAT differs from Transformer

  #11 opened 10 months ago by moon2yue
  1

• Linear projection instead of convolution

  #17 opened a year ago by LukasMahieu
  0

• Can this be applied to languages?

  #20 opened a year ago by ElrondL
  2

• 关于attention中部分代码的问题

  #19 opened a year ago by 01vanilla
  0

• ask for Figure13、14 code

  #16 opened a year ago by 01vanilla
  0

• Difference between crate-demo.pth and model_best.pth.tar (from CRATE-base)

  #14 opened a year ago by FJGEODEV
  0

• pretrained CRATE weight?

  #1 opened 2 years ago by EveningLin
  6

• KeyError:'model'

  #5 opened a year ago by yiichu03
  3

• Taking one further step of whitebox approach

  #9 opened a year ago by ngkel
  1

• The white-box explannation of CLS token

  #10 opened a year ago by ngkel
  1

• 预训练模型

  #3 opened 2 years ago by 01vanilla
  0

• requirement

  #2 opened 2 years ago by suyou5
  2