unsky/FPN

Issues

• need ResNet50.v2.caffemodel

  #110 opened 4 years ago by wangq0688
  1

• 预训练权重

  #111 opened 4 years ago by zhaoyanjoin
  0

• TypeError: slice indices must be integers or None or have an __index__ method

  #104 opened 6 years ago by zh583007354
  3

• make Error: lopencv lhdf5_serial

  #77 opened 7 years ago by elesun2018
  1

• 关于模型中p3__lateral和p4__lateral层的疑问

  #109 opened 5 years ago by Yangel-hide
  0

• ModuleNotFoundError: No module named 'generate_anchors'

  #108 opened 6 years ago by kylezlj
  0

• When I run demo.py :Failed to parce NetParameter file

  #107 opened 6 years ago by ifuleyou
  0

• wen

  #106 opened 6 years ago by ifuleyou
  0

• Failed to parce NetParameter file:

  #105 opened 6 years ago by ifuleyou
  0

• when I run demo.py. i met a problem, shapes (1000,8) (84,)

  #87 opened 6 years ago by mantou22
  1

• paper和code中的2个疑问

  #73 opened 7 years ago by foralliance
  1

• 训练时一些细节问题

  #80 opened 7 years ago by zq130320339
  1

• 使用VGG作为backbone实现FPN的问题

  #82 opened 7 years ago by xiaoxiangyeyuwangye
  1

• Changes to Caffe-FPN

  #103 opened 6 years ago by SecretGuardian
  1

• 训练自己的数据集出现了问题

  #102 opened 6 years ago by gllspeed
  0

• mAP询问

  #101 opened 6 years ago by yangxue0827
  0

• 怎样生成log文件

  #99 opened 6 years ago by jjprincess
  1

• 修改了anchor数量之后报错了

  #100 opened 6 years ago by FakerYFX
  0

• make -j16 all出错

  #70 opened 7 years ago by yin19961202
  6

• 增大ratios和scale之后，代码哪些地方需要修改？

  #98 opened 6 years ago by FakerYFX
  0

• 如何进行多尺度的训练？

  #97 opened 6 years ago by FakerYFX
  4

• 保存检测框的结果要怎么修改代码？

  #95 opened 6 years ago by FakerYFX
  0

• how to continue train?

  #96 opened 6 years ago by FakerYFX
  0

• 无法测试

  #94 opened 6 years ago by jjprincess
  2

• self._allowed_border = layer_params.get('allowed_border', 1000)

  #88 opened 6 years ago by unicoe
  2

• can't find FP_Net_end2end_merge_rcnn.sh?

  #92 opened 6 years ago by FakerYFX
  2

• 训练集为100*100左右图片， 目标大小为40*40左右，请问如何设置anchor的大小

  #89 opened 6 years ago by mantou22
  1

• 请问 为什么我的训练自己的模型 时间为什么怎么慢呢？

  #93 opened 6 years ago by mantou22
  1

• when I run the demo.py,there is a problem,could you help me ?

  #61 opened 7 years ago by sujinpeng
  5

• another problem

  #63 opened 7 years ago by sujinpeng
  6

• Does the code support training fpn with batch size 1 with one gpu?

  #83 opened 6 years ago by anorthman
  9

• 请问merged rcnn和shared rcnn在原paper里头有说明的吗，还是大神你自己的两种具体实现方式呀？

  #76 opened 7 years ago by 646677064
  1

• can't find FP_Net_end2end_merge_rcnn.sh?

  #91 opened 6 years ago by FakerYFX
  0

• 各类的结果AP全是0，为什么？

  #90 opened 6 years ago by chaobeiying
  3

• where is the pre-training model ？ how to get it ?

  #84 opened 6 years ago by mantou22
  3

• 如果想要使用hard negative mining应该如何修改？

  #67 opened 7 years ago by leonshek
  2

• 训练时，任意width、height的图片都不会报错么？

  #65 opened 7 years ago by shuishiwojiade
  2

• make -j16 all BUG.....

  #71 opened 7 years ago by jiachen0212
  3

• scale problem

  #81 opened 7 years ago by SineXue
  3

• 训练好的caffemodel容量1.2G

  #79 opened 7 years ago by SineXue
  4

• when I run the demo.py,there is a problem,could you help me??????

  #86 opened 6 years ago by mantou22
  0

• Error parsing text-format caffe.NetParameter: 3205:24: Message type "caffe.LayerParameter" has no field named "smooth_l1_loss_param".

  #85 opened 6 years ago by mantou22
  0

• How many iterations of training？

  #75 opened 7 years ago by NIEQINQIN
  3

• 关于Anchor的调整

  #78 opened 7 years ago by Asagamiyou
  0

• 训练自己的数据集出现的问题

  #74 opened 7 years ago by wolf345
  0

• fpn利用test.sh测试时生成一个.pkl文件

  #72 opened 7 years ago by xubo00
  0

• /usr/bin/ld: cannot find -lopencv_dep_cudart

  #69 opened 7 years ago by lori0726
  1

• about model size

  #68 opened 7 years ago by lyj0823
  0

• config.py中的selective_search换成gt有什么区别影响

  #66 opened 7 years ago by leonshek
  1

• How do you deal with empty rois

  #64 opened 7 years ago by twmht
  0