jwyang/fpn.pytorch

Issues

• ImportError:/fpn.pytorch-master/lib/model/nms/_ext/nms/_nms.so: undefined symbol: __cudaPopCallConfiguration

  #56 opened 4 years ago by AndrewNest
  1

• Does it support PyTorch1.0 or higher?

  #54 opened 5 years ago by yudawen
  1

• sh make.sh error

  #34 opened 6 years ago by he010103
  2

• sh make.sh 编译不通过

  #55 opened 4 years ago by lllllmaster-lulllll
  0

• Resnet output doesn't match the FPN_FEAT_STRIDES

  #12 opened 4 years ago by destinyzs
  2

• change anchor size to detect extremely small objects

  #26 opened 6 years ago by KevinQian97
  2

• train on my dataset

  #48 opened 5 years ago by ttianwang
  1

• Training suddenly terminate after first epoch. Looking for help, plz

  #25 opened 6 years ago by KevinQian97
  5

• rcnn_box loss equals to 0 when training

  #17 opened 7 years ago by maozezhong
  5

• Why is the result not as good as faster-rcnn? Have you modified the code?

  #53 opened 5 years ago by puyiwen
  0

• train with multi gpus can not work

  #52 opened 5 years ago by mynameischaos
  0

• COCO Results

  #29 opened 6 years ago by haorui-ji
  5

• from model.utils.cython_bbox import bbox_overlaps

  #47 opened 5 years ago by brooklyn1900
  1

• No such file or directory：log

  #24 opened 7 years ago by KevinQian97
  3

• torch.log(torch.sqrt(h * w) / 224.0) in fpn.py

  #49 opened 5 years ago by fafancier
  1

• Conversion to TensorRT

  #51 opened 5 years ago by ma-siddiqui
  0

• roi pooling

  #50 opened 5 years ago by Youggie
  0

• RCNN_roi_align ERROR when training

  #21 opened 7 years ago by xiaomengyc
  7

• ImportError: No module named faster_rcnn.faster_rcnn_cascade

  #9 opened 7 years ago by chj1933
  5

• about rpn_loss_box is zero

  #46 opened 5 years ago by ww451575464
  0

• There was a bug when I continued to train the model. RuntimeError: Expected object of type torch.FloatTensor but found type torch.cuda.FloatTensor for argument #4 'other'

  #45 opened 6 years ago by smileSJin
  1

• bbox_outside_weights normalized incorrectly (anchor_target_layer_fpn.py:136)

  #44 opened 6 years ago by rschwieterman
  0

• when i porting fpn pytorch version 0.4 to 1.0, i face with error

  #36 opened 6 years ago by EzoBear
  1

• inconsistant anchor reference (may be the cause of the lower accuracy than faster-rcnn)

  #6 opened 7 years ago by doronpor
  22

• Where I can crop the image then send it into backbone

  #43 opened 6 years ago by JackJWWong
  0

• TypeError: load() got an unexpected keyword argument 'encoding'

  #42 opened 6 years ago by wddwzwhhxx
  0

• Have somebody used ROI pooling successfully?I have some errors

  #41 opened 6 years ago by yin19961202
  0

• irq/133-nvidia error

  #40 opened 6 years ago by mathpopo
  0

• longtensor and reciprocal

  #31 opened 6 years ago by planckztd
  2

• Nan

  #39 opened 6 years ago by SmallHedgehog
  0

• The version of pytorch

  #27 opened 6 years ago by bnu-wangxun
  1

• How to load the Images into the network?

  #38 opened 6 years ago by JackJWWong
  0

• TypeError: 'list' object is not callable

  #37 opened 6 years ago by XingLiuJia
  0

• COMPILEERROR（I already change the sm_61 but error still happened）

  #32 opened 6 years ago by yuyijie1995
  6

• demo.py is not updated for fpn

  #15 opened 7 years ago by Karthik-Suresh93
  3

• Cuda Error: invalid argument nms_cuda_kernel.cu

  #18 opened 7 years ago by ljtruong
  2

• When I set the pooling mode as 'pool', an error arise

  #33 opened 6 years ago by duwangzhe
  1

• dimension out of range (expected to be in range of [-1, 0], but got 1)

  #30 opened 6 years ago by CXY573
  2

• no file or directory: RCNN_rpn

  #10 opened 7 years ago by wyzjack
  4

• Specify dim in softmax function

  #22 opened 7 years ago by tianlu-wang
  3

• Why is FPN worse than Faster-rcnn?

  #5 opened 7 years ago by laidy3
  3

• I want to join you do some code task for free.

  #28 opened 6 years ago by ImMrMa
  0

• Training suddenly terminates with run time error, please help

  #13 opened 7 years ago by Karthik-Suresh93
  3

• Do you figured out why the performance is worse than faster rcnn?

  #16 opened 7 years ago by ruoqianguo
  1

• collapses after the trainval_net.py finishing its first training iter

  #11 opened 7 years ago by destinyzs
  3

• GPU Memory issues about the ROI Pooling lib

  #23 opened 7 years ago by lyxok1
  1

• Got 80.5 mAP on Pascal VOC 2007

  #20 opened 7 years ago by ruoqianguo
  0

• what is resnet101_caffe.pth

  #14 opened 7 years ago by yuanyuan1993
  1

• Definition of ResNet

  #8 opened 7 years ago by qiuzheFD
  0

• confused about the BN layer

  #7 opened 7 years ago by iSmida
  0