Qiuxiaqing

train.py stopped in epoch[0],[49/63], when it training

1

Epoch: [ 0], [ 49/ 63] time: 115.1479, d_loss: 58.66755, g_loss: 20.59421, category_loss: 0.00000, cheat_loss: 0.00075, const_loss: 0.07337, l1_loss: 20.52009, tv_loss: 0.00000 when i was training the file, the program...

There are something wrong in nd_loss compute

2

The error is following: File "D:\PyCharm_workspace\KD\Knowledge-Distillation-via-ND-main\CIFAR\ReviewKD++\utils.py", line 62, in project_center loss += 1 - torch.dot(s, e_c) / max_norm RuntimeError: inconsistent tensor size, expected tensor [256] and src [64] to have...