zzzxxxttt/pytorch_simple_CenterNet_45

Issues

• 加载预训练模型出错

  #36 opened 3 years ago by sublong
  2

• 想要指定4,5,6,7这几张GPU进行训练该怎么设置

  #23 opened 4 years ago by philiperss
  4

• No such file or directory: './data/voc/VOCdevkit/VOC2007/annotations_cache'

  #35 opened 3 years ago by 18342960922
  1

• 替换backbone

  #4 opened 4 years ago by kp-1
  3

• 楼主您好，训练好的 hourglass-104 39.9mAP的模型能共享一下吗？

  #41 opened 2 years ago by dashun6
  1

• RuntimeError: cublas runtime error : the GPU program failed to execute at /opt/conda/conda-bld/pytorch_1556653215914/work/aten/src/THC/THCBlas.cu:425

  #40 opened 3 years ago by BoXingJinQiu
  0

• v1.7的Pytorch如何实现DistributedDataParallel training？

  #39 opened 3 years ago by Jianxiang-Wang
  1

• FPS

  #37 opened 3 years ago by zjp99
  0

• Result Extraction

  #28 opened 4 years ago by JiaLim98
  9

• 大佬，我在用运行demo.py文件时，跑不通，训练跟测试都跑通了，

  #32 opened 4 years ago by xinwenbin
  0

• 验证时，输出的结果都是-1

  #25 opened 4 years ago by philiperss
  5

• CornerNet pretrained weights

  #27 opened 4 years ago by keravraj20
  5

• How can't I get the Theoretical performance

  #34 opened 3 years ago by ianwang35
  0

• Why I can't achieve the given mAP after retraining by following the steps？

  #33 opened 3 years ago by ZxSnow
  0

• 为什么我使用代码训练的时候得不到预期结果....

  #30 opened 4 years ago by linmo2333
  5

• 为什么test阶段需要nms？

  #31 opened 4 years ago by NeuZhangQiang
  2

• IndexError: index 100 is out of bounds for axis 0 with size 100

  #26 opened 4 years ago by Shank2358
  2

• Train centernet with custom dataset

  #17 opened 4 years ago by ashley-mv
  3

• dla_dcn没有实现吗?

  #29 opened 4 years ago by Jiachenyin1
  0

• cuda error

  #24 opened 4 years ago by dyl25984
  0

• IndexError: index 14 is out of bounds for axis 0 with size 2

  #22 opened 4 years ago by dyl25984
  4

• Gaussian radius implementation is wrong

  #5 opened 4 years ago by kikirizki
  3

• Illegal Instruction

  #21 opened 4 years ago by JiaLim98
  0

• AttributeError: 'list' object has no attribute 'to'

  #20 opened 4 years ago by JiaLim98
  7

• RuntimeError: The size of tensor a (152) must match the size of tensor b (150) at non-singleton dimension 3

  #19 opened 4 years ago by Kongsea
  12

• 能详细解释下get_affine_transform的意义吗？

  #18 opened 4 years ago by GYxiaOH
  1

• Can you share your pre-trained model weights?

  #15 opened 4 years ago by TomHeaven
  3

• Inference

  #16 opened 4 years ago by PengZH99
  4

• train custom data hmap_loss problem

  #14 opened 4 years ago by ltss1988
  2

• 请问一下，这个项目里面有用到DCN吗？

  #8 opened 4 years ago by songyuc
  2

• It seems the affine transformation in the coco.py force the image into the size of 512x512

  #12 opened 4 years ago by songyuc
  5

• There is no nms folder in the repo

  #13 opened 4 years ago by songyuc
  4

• Are the "labels" referred as the indexes matching by name of the list "COCO_NAMES"?

  #11 opened 4 years ago by songyuc
  2

• What is the meaning of "self.max_objs = 128"

  #10 opened 4 years ago by songyuc
  2

• 请问一下，可以把计算数据集均值、方差、特征值和特征向量的代码发给我学习一下吗？

  #9 opened 4 years ago by songyuc
  2

• How can we train on our own custom dataset?

  #7 opened 4 years ago by davidhuangal
  2

• 可以提供一个demo么？我自己写了半天，结果不对

  #6 opened 4 years ago by hyaihjq
  2

• coco 140个epochs??

  #3 opened 4 years ago by SSJIACV
  1

• centernet训练自己的数据遇到的错误

  #2 opened 4 years ago by longchao1
  1

• Hi, why do you setup the torch.backends.cudnn.enabled False?

  #1 opened 4 years ago by wuyxiquanquan
  6