locuslab/convmixer

Issues

• something about loss

  #16 opened 3 years ago by tjl1
  0

• CIFAR-10 training settings

  #15 opened 3 years ago by fugokidi
  0

• Could you please release the training logs of convmixer on imagenet and cifar?

  #14 opened 3 years ago by zw615
  0

• Training scheme modifications for small GPUs

  #13 opened 3 years ago by justanhduc
  4

• Training time

  #12 opened 3 years ago by bonlime
  1

• Cifar10 baseline doesn't reach 95%

  #11 opened 3 years ago by K-H-Ismail
  13

• License

  #10 opened 3 years ago by Aspie96
  2

• Request more experiment results to compare to other architecture.

  #8 opened 3 years ago by Luciennnnnnn
  1

• Segmentation ConvMixer architecture ?

  #9 opened 3 years ago by divideconcept
  0

• padding=same？

  #7 opened 3 years ago by linhaoqi027
  1

• Do you use this model on other downstream work? (like semantic segmentation )

  #6 opened 3 years ago by BboyHanat
  3

• What's new about this model?

  #3 opened 3 years ago by vztu
  5

• Experiments with full convolutional layers instead of patch embedding？

  #4 opened 3 years ago by forjiuzhou
  2

• 你好，convmixer能否直接提取图片的2D特征？

  #5 opened 3 years ago by zmxsss
  0

• is single gpu training possible?

  #1 opened 3 years ago by flydragon2018
  2

• weight location

  #2 opened 3 years ago by flydragon2018
  1