这两天跑测试图时,发现用同样的model,同样的测试图,每次运行结果不同; 经过漫长的debug发现,在net architure中有dropout,如下(4): (conv_block): Sequential( (0): ReflectionPad2d((1, 1, 1, 1)) (1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1)) (2): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False) (3): ReLU(inplace)