open-mmlab/OpenUnReID

Issues

• Faiss assertion 'err__ == cudaSuccess' failed in void faiss::gpu::runL2Norm(

  #7 opened 4 years ago by 944284742
  6

• super strong baseline

  #56 opened a year ago by llf0223
  0

• 你好，我的faiss出了点问题；

  #24 opened 3 years ago by Reggieharden
  12

• Ask for help for solving faiss assertion error

  #54 opened 2 years ago by cd70zyx
  1

• severe overfitting with a single GPU

  #53 opened 2 years ago by hartvon
  2

• Faiss assertion 'err__ == cudaSuccess' failed in void faiss::gpu::runL2Norm(faiss::gpu::Tensor<T, 2,

  #44 opened 3 years ago by July-zh
  1

• batchsize

  #55 opened 2 years ago by llf0223
  0

• Loss becomes nan when I try to train MMT for 100 epochs

  #52 opened 2 years ago by TheBobbyliu
  0

• strong_basline训练到后期acc突然降到0 loss变为nan

  #43 opened 3 years ago by Tidewww
  3

• 这是什么问题？

  #50 opened 2 years ago by Jacobxz
  0

• K-means for SpCL

  #51 opened 2 years ago by yjh576
  0

• 计算三元组损失时，特征需要归一化吗，代码里没有做归一化

  #49 opened 3 years ago by chongyangwang-song
  0

• Validation freq

  #46 opened 3 years ago by RAMIhamza
  3

• About Leaderboard.

  #42 opened 3 years ago by clarkhedi
  1

• 关于num_parts的问题

  #45 opened 3 years ago by chongyangwang-song
  3

• Question about 'jaccard distance'

  #48 opened 3 years ago by chongyangwang-song
  0

• Question about train loader

  #47 opened 3 years ago by chongyangwang-song
  0

• Question on performance

  #4 opened 4 years ago by yjh576
  21

• spcl+在UDA上不能复现结果

  #41 opened 3 years ago by keenJMS
  0

• Implement of SpCL

  #40 opened 3 years ago by yjh576
  1

• Something about the Sampler

  #39 opened 3 years ago by Frost-Yang-99
  1

• Difference between OpenUnReID and the original MMT's implementation

  #33 opened 3 years ago by RAMIhamza
  2

• AttributeError: module 'faiss._swigfaiss' has no attribute 'delete_SwigPyIterator'

  #31 opened 3 years ago by zhoufengfan
  1

• 请问如何扩大验证集的规模

  #30 opened 3 years ago by HWNHJJ
  0

• The difference between strong_baseline and MMT proposed in unsupervised domain adaptation

  #22 opened 4 years ago by GJTNB
  3

• Market1501->Duck,单GPU可以汇报下么？

  #29 opened 3 years ago by muzishen
  2

• AttributeError: Can't pickle local object 'SubPolicy.__init__.<locals>.<lambda>'

  #1 opened 4 years ago by PangJian123
  2

• spcl 无监督域自适应

  #23 opened 3 years ago by HWNHJJ
  6

• question about "samples_per_bn"

  #13 opened 3 years ago by valencebond
  3

• why source_pretrain wrong?

  #21 opened 4 years ago by 969191832
  2

• about "select_cams"

  #20 opened 4 years ago by SunskyF
  16

• visualize test results

  #19 opened 4 years ago by hjwang-824
  1

• Question about Reliable Clusters

  #18 opened 4 years ago by Youskrpig
  2

• 关于CUDA out of memory. Tried to allocate 18.00 MiB (GPU 0; 5.80 GiB total capacity; 4.98 GiB already allocated; 4.25 MiB free; 9.66 MiB cached)

  #17 opened 4 years ago by GJTNB
  2

• 请问测试多个数据集时，如何在optional arguments中添加参数？

  #16 opened 4 years ago by PangJian123
  4

• 关于分chunk的问题

  #15 opened 4 years ago by huang1997214
  3

• 距离选择的问题。

  #12 opened 4 years ago by huang1997214
  5

• SpCL Train_step的问题

  #14 opened 4 years ago by huang1997214
  1

• Will you share trained models for vehicle re-ID?

  #11 opened 4 years ago by j-adamczyk
  3

• 训练时候的小问题?

  #10 opened 4 years ago by huang1997214
  5

• 关于大规模数据的问题

  #9 opened 4 years ago by huang1997214
  8

• 您好，有两个小问题。

  #8 opened 4 years ago by huang1997214
  4

• When will the trained model of SSG be Shared？

  #6 opened 4 years ago by YuanXinCherry
  2

• Questions about copy_state_dict(state_dict[key], model) in test_reid.py

  #5 opened 4 years ago by YuanXinCherry
  13

• Question on SPCL+

  #3 opened 4 years ago by Terminator8758
  2

• 关于ScPL的问题

  #2 opened 4 years ago by zengwb-lx
  6