YimianDai/open-aff

Issues

• ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 256, 1, 1])

  #43 opened a year ago by war0927
  2

• 关于BN层

  #44 opened a year ago by war0927
  2

• 请问怎么在pytorch直接使用AFF呀？

  #45 opened 10 months ago by XinQ-Yu
  1

• aff使用于fpn或者Unet结构

  #39 opened 2 years ago by 760677482
  2

• Do you have the code of Pytorch version?

  #2 opened 4 years ago by ma3252788
  15

• 如何利用已经训练好的模型来测试

  #4 opened 4 years ago by CZMHHH
  2

• 您好可以发一下iAFF-ResNeXt-50-32x4d训练ImageNet的权重文件吗

  #42 opened a year ago by LeePei1
  0

• 可否进行实现三种特征图的融合

  #35 opened 2 years ago by yuyunlong2002
  1

• 为何将iaff加入到resnet第一层会报如下错误：TypeError: forward() missing 1 required positional argument: 'residual'

  #41 opened a year ago by cbn3
  0

• TypeError: forward() missing 1 required positional argument: 'residual'

  #29 opened 3 years ago by haideralimughal
  2

• resnet50预训练模型

  #30 opened 3 years ago by fukangzhang
  3

• 关于应用于FPN

  #10 opened 4 years ago by Pluto1314
  7

• 代码中AFF模块中，为什么要用2×呢？

  #33 opened 3 years ago by YoungLNB
  1

• When the website will be ready?

  #40 opened 2 years ago by Zhiyuan-R
  0

• iAFF类的forward函数中, xg2 = self.global_att(xi) 这一行是不是笔误

  #37 opened 2 years ago by CaA23187
  1

• 运行模块时报错请教

  #38 opened 2 years ago by GeniusLittleFlyingPig
  1

• Concat and fusion instead of add in iAFF and AFF

  #36 opened 2 years ago by quannv-1204
  0

• 请教 这种特征融合思路能否应用到 NLP 中？

  #34 opened 2 years ago by SupritYoung
  0

• __init__() takes from 1 to 3 positional arguments but 6 were given

  #32 opened 3 years ago by Thar-un
  0

• module -> self.global_att2

  #31 opened 3 years ago by BoYangXDU
  0

• 是否可以应用到3D目标检测中

  #24 opened 4 years ago by liulingfeng123
  1

• 您好，请问AFF模块中的最后一步为什么要乘以2？有什么意义呢？希望收到您的回复

  #28 opened 3 years ago by ElegantNorlin
  2

• AFF-FPN

  #25 opened 4 years ago by lab-gpu
  3

• ASKCResNetFPN

  #26 opened 4 years ago by YimianDai
  0

• Can you share inference time data for different AFF/iAFF models?

  #23 opened 4 years ago by cmsflash
  2

• mxnet parameters to pytorch

  #22 opened 4 years ago by Trouble404
  1

• 关于 Network visualization 的问题

  #21 opened 4 years ago by hzlbbfrog
  6

• 关于 Network visualization

  #20 opened 4 years ago by hzlbbfrog
  1

• 注意力机制的预训练模型有restnet18的吗？

  #8 opened 4 years ago by roger-cv
  25

• 关于localization和small object

  #19 opened 4 years ago by yinyuhangi
  0

• Batch norm问题

  #15 opened 4 years ago by Pluto1314
  7

• 问题

  #16 opened 4 years ago by yinyuhangi
  4

• CIFAR-100 dataset in train_cifar.py has only 20 classes.

  #14 opened 4 years ago by takedarts
  6

• 关于 Channel Attention 的 Scale

  #7 opened 4 years ago by YimianDai
  1

• Training from scratch - log report

  #11 opened 4 years ago by moabarar
  9

• iAFF为什么要用两次注意力

  #9 opened 4 years ago by swjtulinxi
  4

• 请问AFF的代码中2 * F.broadcast_mul(x, wei) + 2 * F.broadcast_mul(residual, 1-wei)

  #13 opened 4 years ago by swjtulinxi
  1

• 作者您好，我想请教下关于通道注意力的尺度问题

  #12 opened 4 years ago by lightersss
  6

• Typo?

  #5 opened 4 years ago by Ha0Tang
  1

• 一些问题请教一下

  #1 opened 4 years ago by qdd1234
  7

• Excellent work, looking forward to your pytorch version！

  #6 opened 4 years ago by mogeko2333
  0

• code implementation: multiplication of 2 in the final layer output

  #3 opened 4 years ago by Crane-YU
  2