为什么训练的时候会出现min value和max value的打印

[zxj329]

max value is tensor(1.4666, device='cuda:3', grad_fn=<MaxBackward1>) max value is tensor(1.0702, device='cuda:6', grad_fn=<MaxBackward1>) min value is tensor(-1.0317, device='cuda:0', grad_fn=<MinBackward1>) max value is tensor(1.2846, device='cuda:0', grad_fn=<MaxBackward1>) min value is tensor(-1.0303, device='cuda:4', grad_fn=<MinBackward1>) min value is tensor(-1.1144, device='cuda:1', grad_fn=<MinBackward1>) max value is tensor(1.1782, device='cuda:1', grad_fn=<MaxBackward1>) min value is tensor(-1.1766, device='cuda:5', grad_fn=<MinBackward1>) max value is tensor(1.2042, device='cuda:5', grad_fn=<MaxBackward1>) max value is tensor(1.1102, device='cuda:0', grad_fn=<MaxBackward1>) min value is tensor(-1.1077, device='cuda:5', grad_fn=<MinBackward1>) max value is tensor(1.2457, device='cuda:5', grad_fn=<MaxBackward1>) max value is tensor(1.0143, device='cuda:2', grad_fn=<MaxBackward1>)

是因为生成器还不稳定吗？