guxinqian / AP3D

Pytorch implementation of "Appearance-Preserving 3D Convolution for Video-based Person Re-identification"

Apache License 2.0

97 stars 24 forks source link

Closed kanagi2587 closed 3 years ago

kanagi2587 commented 3 years ago

如题，下载了您的源代码和MARS数据集，进行源代码的运行时

for batch_idx, (vids, pids, _) in enumerate(trainloader):

中总是满足 if (pids-pids[0]).sum() == 0:

can't compute triplet loss

        continue

即pids总是相等的一组值。这是为什么

guxinqian commented 3 years ago

是不是trainloader没有shuffle？正常情况下batchsize设为32基本不会出现这种情况。设成16偶尔会这样