当把G修改为只剩一个res（个人电脑显存不够用，默认9个在回传误差时候就溢了）
print(self.netG_A)
DataParallel(
(module): ResnetGenerator(
(model): Sequential(
(0): ReflectionPad2d((3, 3, 3, 3))
(1): Conv2d(3, 64, kernel_size=(7, 7), stride=(1, 1))
(2): InstanceNorm2d(64, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(3): ReLU(inplace=True)
(4): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(5): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(6): ReLU(inplace=True)
(7): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(8): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(9): ReLU(inplace=True)
(10): ResnetBlock(
(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, track_running_stats=False)
(3): ReLU(inplace=True)
(4): ReflectionPad2d((1, 1, 1, 1))
(5): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1))
(6): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
)
)
(11): ConvTranspose2d(256, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), output_padding=(1, 1))
(12): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(13): ReLU(inplace=True)
(14): ConvTranspose2d(128, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), output_padding=(1, 1))
(15): InstanceNorm2d(64, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(16): ReLU(inplace=True)
(17): ReflectionPad2d((3, 3, 3, 3))
(18): Conv2d(64, 3, kernel_size=(7, 7), stride=(1, 1))
(19): Tanh()
)
)
)
print(self.netG_B)
DataParallel(
(module): ResnetGenerator(
(model): Sequential(
(0): ReflectionPad2d((3, 3, 3, 3))
(1): Conv2d(3, 64, kernel_size=(7, 7), stride=(1, 1))
(2): InstanceNorm2d(64, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(3): ReLU(inplace=True)
(4): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(5): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(6): ReLU(inplace=True)
(7): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(8): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(9): ReLU(inplace=True)
(10): ResnetBlock(
(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, track_running_stats=False)
(3): ReLU(inplace=True)
(4): ReflectionPad2d((1, 1, 1, 1))
(5): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1))
(6): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
)
)
(11): ConvTranspose2d(256, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), output_padding=(1, 1))
(12): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(13): ReLU(inplace=True)
(14): ConvTranspose2d(128, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), output_padding=(1, 1))
(15): InstanceNorm2d(64, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(16): ReLU(inplace=True)
(17): ReflectionPad2d((3, 3, 3, 3))
(18): Conv2d(64, 3, kernel_size=(7, 7), stride=(1, 1))
(19): Tanh()
)
)
)
然后是D的
print(self.netD_A)
DataParallel(
(module): NLayerDiscriminator(
(model): Sequential(
(0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(1): LeakyReLU(negative_slope=0.2, inplace=True)
(2): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(3): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(4): LeakyReLU(negative_slope=0.2, inplace=True)
(5): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(6): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(7): LeakyReLU(negative_slope=0.2, inplace=True)
(8): Conv2d(256, 512, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1))
(9): InstanceNorm2d(512, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(10): LeakyReLU(negative_slope=0.2, inplace=True)
(11): Conv2d(512, 1, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1))
)
)
)print(self.netD_B)
DataParallel(
(module): NLayerDiscriminator(
(model): Sequential(
(0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(1): LeakyReLU(negative_slope=0.2, inplace=True)
(2): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(3): InstanceNorm2d(128, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(4): LeakyReLU(negative_slope=0.2, inplace=True)
(5): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
(6): InstanceNorm2d(256, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(7): LeakyReLU(negative_slope=0.2, inplace=True)
(8): Conv2d(256, 512, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1))
(9): InstanceNorm2d(512, eps=1e-05, momentum=0.1, affine=False, track_running_stats=False)
(10): LeakyReLU(negative_slope=0.2, inplace=True)
(11): Conv2d(512, 1, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1))
)
)
)yield 是实现生成器方法之一，当函数使用yield方法，则该函数就成为了一个生成器。调用该函数，就等于创建了一个生成器对象。可用于迭代
yield 是一个类似 return的关键字，迭代一次遇到yield时就返回yield后面的值。重点是：下一次迭代时，从上一次迭代遇到的yield后面的代码开始执行。
send(msg)与next()的区别在于send可以传递参数给yield表达式，这时传递的参数会作为yield表达式的值，而yield的参数是返回给调用者的值。——换句话说，就是send可以强行修改上一个yield表达式值。比如函数中有一个yield赋值，a = yield 5，第一次迭代到这里会返回5，a还没有赋值。第二次迭代时，使用.send(10)，那么，就是强行修改yield 5表达式的值为10，本来是5的，那么a=10
send(msg)与next()都有返回值，它们的返回值是当前迭代遇到yield时，yield后面表达式的值，其实就是当前迭代中yield后面的参数。
第一次调用时必须先next()或send(None)，否则会报错，send后之所以为None是因为这时候没有上一个yield。可以认为，next()等同于send(None)。