HaloTrouvaille/YOLO-Multi-Backbones-Attention

Issues

• Can someone update the datasets? URL invalid

  #33 opened 8 months ago by HongyuZhu999
  0

• 2020.9.26 (1) Add VoVNet as the backbone. The result is excellent.

  #19 opened 4 years ago by joe660
  2

• #数据集

  #32 opened 3 years ago by zhang629
  1

• where in your code CBAM and eca blocks are defined.

  #16 opened 4 years ago by zunairaR
  6

• 关于cfg

  #31 opened 3 years ago by yaoyaoyao-1019
  1

• Custom Data in YAML form?

  #30 opened 3 years ago
  0

• How to train this on custom dataset?

  #29 opened 3 years ago
  0

• Train erro

  #15 opened 4 years ago by xxd1999
  4

• How do you set the kernel size of ECA Layer?

  #28 opened 4 years ago by developer0hye
  0

• train.py

  #27 opened 4 years ago by gzw820
  1

• 关于yolo+shufflenetv2在visdrone训练的结果

  #23 opened 4 years ago by gui-yuzhi
  2

• VoVNet

  #26 opened 4 years ago by jch-wang
  1

• where is cbam?

  #21 opened 4 years ago by xiaomahencai
  1

• “activation=linear”

  #25 opened 4 years ago by VegetablesBird6
  0

• cfg文件的问题

  #20 opened 4 years ago by joe660
  14

• 您好，作者，想问下哪个文件是vovnet的cfg文件

  #24 opened 4 years ago by oswNumber1
  0

• why don't you add code to forward in create_module?

  #22 opened 4 years ago by xiaomahencai
  1

• 请问有预训练权重吗

  #18 opened 4 years ago by dengfenglai321
  0

• Parameters for Ghostnet + YoloV3

  #17 opened 4 years ago
  2

• 请问作者有与训练权重吗

  #8 opened 4 years ago by g1115658192
  1

• 没有找到CBAM和ECA Block

  #11 opened 4 years ago by x2ss
  3

• 在‘create_modules’没有添加CA和SA？

  #9 opened 4 years ago by yangzhaojason
  1

• Error while training with a custom dataset

  #12 opened 4 years ago by alancarlosml
  3

• 量化问题

  #14 opened 4 years ago by lxn96
  4

• Could you make a tutorial in detail about how to train with custom datasets?

  #13 opened 4 years ago by BenoitKAO
  1

• 量化问题

  #10 opened 4 years ago by yolunghiu
  1

• Thank u for your great work， is it convenient for u to share the ShuffleNet and the attention mechanism module in github, I wanna to know how those mechanism work, thank u again, I have clicked star to your project : )

  #5 opened 4 years ago by a81940595
  1

• is it better way to construct models via pytorch api directly rather than parse from cfg?

  #4 opened 4 years ago by lucasjinreal
  1

• RuntimeError: Sizes of tensors must match except in dimension 2. Got 40 and 48

  #6 opened 4 years ago by GUZ-lkc2018
  1

• AssertionError: Unsupported fields ['stopbackward', 'kernel_size', 'max_delta']

  #7 opened 4 years ago by lccui
  2

• visdrone训练参数

  #1 opened 5 years ago by lilslsls
  1

• 问题请教

  #3 opened 5 years ago by 17868380981
  1

• 请问有比如加入se模块的预训练模型么？如果没有是否可以用传统结构的预训练模型，然后对se模块权重做随机初始化？

  #2 opened 5 years ago by clw5180
  3