L_spa训练损失参数传递顺序？

[2972749844]

loss_spa = torch.mean(L_spa(enhanced_image, img_lowlight))
def forward(self, org , enhance )
感觉这个训练函数里L_spa参数的顺序和定义L_spa损失函数里的顺序不一致呢