luuuyi/RefineDet.PyTorch

Issues

• How to do training on the SAR dataset?

  #50 opened 4 months ago by Leelibrary
  0

• 运行eval_refinedet.py 报错：AttributeError: 'FunctionBackward' object has no attribute 'objectness_thre'

  #49 opened 4 years ago by qinghua-gui
  0

• 只有一类目标的时候报错，训练不起来

  #45 opened 4 years ago by wuzuowuyou
  4

• RuntimeError: copy_if failed to synchronize: device-side assert triggere

  #43 opened 5 years ago by choubin
  3

• RuntimeError: copy_if failed to synchronize: device-side assert triggered

  #38 opened 5 years ago by Crystal950504
  4

• Question about "Hard Negative Mining" code

  #48 opened 4 years ago by JiXuKong
  0

• eval_refinedet.py RuntimeError: Legacy autograd function with non-static forward method is deprecated. Please use new-style autograd function with static forward method.

  #47 opened 4 years ago by sdwangyifan001
  2

• how to train model in 2014*2014 size

  #46 opened 4 years ago by git-hub-tig
  0

• 第一次forward的时候显存暴涨vgg

  #44 opened 4 years ago by wuzuowuyou
  0

• COCO training and evaluation

  #33 opened 5 years ago by andyhahaha
  1

• Hi, thanks for your share! I trained RefineDet320, but only get the result :0.78, not the reported result 79.81%

  #30 opened 5 years ago by espectre
  1

• Memory Requirement

  #42 opened 5 years ago by 0fine0
  2

• train_refinedet.py RunError

  #39 opened 5 years ago by RNG-cz
  1

• can‘t download the file https://s3.amazonaws.com/amdegroot-datasets/instances_trainval35k.json.zip

  #37 opened 5 years ago by zhangting1994
  2

• train on my datasets

  #29 opened 5 years ago by yanelezk
  1

• RuntimeError: cuda runtime error (59) : device-side assert triggered

  #3 opened 6 years ago by wnzhyee
  6

• 您好！请问一下RefineDet.PyTorch工程训练完模型后，进行图像测试时，demo里面的测试程序跟RefineDet.PyTorch对不上的。demo里面的应该是测试SSD网络的吧？

  #7 opened 6 years ago by LYJvincent
  2

• train error

  #41 opened 5 years ago by tuyaxin
  0

• Wrong with TCB module

  #40 opened 5 years ago by pengfeidip
  1

• Something wrong in training own data

  #10 opened 6 years ago by RichardHWD
  4

• when i run eval_refinedet.py

  #36 opened 5 years ago by jaychou790118
  0

• No such file or directory.but i put data right.

  #35 opened 5 years ago by jaychou790118
  0

• Can I get a pretrained weight file ?

  #34 opened 5 years ago by MAhmad-nn
  0

• coco

  #31 opened 5 years ago by luyaozhi
  2

• find negative numbers while testing

  #32 opened 5 years ago by wrs777
  0

• Train loss might be nan

  #18 opened 6 years ago by Fighting-JJ
  12

• Question about the extra layers

  #17 opened 6 years ago by Fighting-JJ
  4

• What if I implement a focal loss?

  #26 opened 5 years ago by DonghoonPark12
  4

• when i test the code i have meet problems

  #25 opened 5 years ago by YYZ-rose
  3

• train on my datasets

  #28 opened 5 years ago by yanelezk
  1

• 关于COCO数据集的测试

  #27 opened 5 years ago by Rodecin
  1

• when i test the code i have meet problems

  #24 opened 5 years ago by YYZ-rose
  0

• when i test the code i have meet problems

  #23 opened 5 years ago by YYZ-rose
  0

• python eval_refinedet.py 时出现一下问题不知道怎么回事请楼主给予解答

  #22 opened 5 years ago by zhangqiang12345678
  3

• problem or bug in refine_match

  #21 opened 5 years ago by pengfeidip
  2

• sorry,something wrong i don't know how to solve it

  #20 opened 5 years ago by rw1995
  4

• about trianing two class

  #12 opened 6 years ago by Amelia0911
  1

• train error

  #19 opened 6 years ago by wdxpython
  1

• test images meet many problems

  #16 opened 6 years ago by lixiandea
  1

• why the mAP on VOC2007 only 69.71%

  #2 opened 6 years ago by lzcdawang
  3

• when i sh train_refinedet320.sh,something is wrong

  #15 opened 6 years ago by rw1995
  2

• test

  #14 opened 6 years ago by wanghai111
  1

• RuntimeError: CUDA error: out of memory

  #13 opened 6 years ago by wanghai111
  1

• Where do you use confidence_threshold when test?

  #11 opened 6 years ago by RichardHWD
  2

• resnet-101 as backbone

  #9 opened 6 years ago by pengfeidip
  2

• Difference between caffe version and paper version VOC mAP

  #8 opened 6 years ago by LovPe
  2

• Pre-trained model

  #6 opened 6 years ago by dalmia
  1

• some confusions in decode module

  #5 opened 6 years ago by cs-heibao
  2

• About Upsampling Conv kernel size

  #4 opened 6 years ago by timho102003
  2

• when support for multi-scale testing?

  #1 opened 6 years ago by haochange
  1