ujsyehao/yolov3-multigpu

Issues

• RuntimeError: cannot perform reduction function max on tensor with no elements because the operation does not have an identity

  #15 opened 3 years ago by Andyen512
  0

• Evaluation is very very slow.

  #12 opened 4 years ago by H19012
  2

• Failed training with 8 GPUs

  #11 opened 4 years ago by wildcat5566
  1

• best mAP=0.252

  #14 opened 3 years ago by hhhhhpy
  1

• 8张显卡训练有问题

  #10 opened 4 years ago by Miki-lin
  4

• RuntimeError: output with device cuda:2 doesn't match the desired device cpu

  #13 opened 3 years ago by hhhhhpy
  0

• 多GPU训练报错

  #9 opened 4 years ago by felix-wt
  2

• get error

  #2 opened 4 years ago by cv-small-snails
  1

• RuntimeError: all tensors must be on devices[0]

  #3 opened 4 years ago by guoyanzi
  1

• noobj_mask[b[i], anchor_ious > ignore_thres, gj[i], gi[i]] = 0 # i is index RuntimeError: CUDA error: device-side assert triggered

  #4 opened 4 years ago by Shunli-Wang
  5

• module must have its parameters and buffers on device cuda:0 (device_ids[0]) but found one of them on device: cuda:1

  #5 opened 4 years ago by AnhPC03
  1

• Did anyone successfully reproduce the result from scratch？

  #6 opened 4 years ago by gaopinghai
  3

• why is it needed to get the batch index in build_targets

  #8 opened 4 years ago by mailvijayasingh
  1

• 请问giou loss怎么实现？我自己写的貌似有问题0_0。可以请教您一下吗？

  #7 opened 4 years ago by xiaoyuzhu666
  1

• 请问作者，可以使用多gpu训练么？

  #1 opened 5 years ago by sditeng
  1