AILab-CVC/UniRepLKNet

Issues

• unireplknet_s is slower than convnext_b

  #22 opened 2 months ago by VERYBC
  2

• Install install 19_large_depthwise_conv2d_torch_extension but ”bash: ./setup.py: Permission denied“

  #18 opened 8 months ago by CYYJL
  2

• How to convert TesorRT

  #21 opened 4 months ago by monologuesmw
  0

• How could i use reparameterization to convtranspose

  #20 opened 5 months ago by SherryYu33
  0

• It's stuck in the./setup.py install --user step

  #19 opened 6 months ago by Changee-git
  3

• Why UniRepLKNet does not using drop path.

  #17 opened 9 months ago by powermano
  0

• 关于UniRepLKNetBlock中的self.gamma的作用

  #16 opened 9 months ago by wsy-yjys
  0

• input_size 384*128 tends to be nan

  #14 opened 10 months ago by Zoesxw
  3

• The training is very slow

  #15 opened 10 months ago by tinggh
  3

• 图像分割

  #13 opened 10 months ago by songwan1
  0

• 作者您好，在Time-series子项目中，张量的格式似乎出错

  #10 opened a year ago by prestyan
  7

• Layer Decay impl only supports convnext_small/base/large/xlarge

  #12 opened a year ago by ZackSong0319
  1

• cp: 无法获取'ModelNet40' 的文件状态(stat): 没有那个文件或目录

  #11 opened a year ago by TKroos18
  1

• error: command 'C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.4\\bin\\nvcc.exe'

  #6 opened a year ago by ll8657
  1

• can it convert to onnx and trt to boost the inference speed?

  #5 opened a year ago by zimenglan-sysu-512
  1

• depthwise conv 似乎没加速

  #7 opened a year ago by ZhuShengchen
  1

• 您好，无法使用timm创建模型

  #9 opened a year ago by xzx0554
  1

• 论文里面的一个问题

  #4 opened a year ago by EveningLin
  4

• A small bug in reparameterize()

  #8 opened a year ago by elokuuM
  2

• The parameter loading is incomplete

  #3 opened a year ago by Manningchan
  2