另一种提取中间层输出的办法与风格迁移的改进

提取中间层输出

提到的使用hook函数获取中间层输出以外，还可以用更直接的办法，不使用hook函数

方法一：将model中的每层拆解出来依次在监控下让数据流过
可以借助isinstance(layer, nn.Conv2d)判断层类型

def extract_layers(layers,img,model):
    layer_index = 0
    res = []
    for layer in model:
        if layer_index == 0:
            out = layer(img)
        else:
            out = layer(out)
        if layer_index in layers:
            res.append(out)
        layer_index += 1
    return res

方法二：直接在定义模型的时候输出中间层数据

class model(nn.Module):
	def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(64,32)
        self.linear2 = nn.Linear(32,16)
    def forward(self,input):
        linear1_out = self.linear1(input)
        linear2_out = self.linear2(linear1_out)
        return linear1_out,linear2_out

对于噪点的改进-降噪

修改方程，向方程中加入总变差损失(total variation denoising)

$L_{total}(S,C,G) = \alpha L_{content}(C,G)+\beta L_{style} (S,G)+\gamma L_{noise}(C,G)$ $L_{noise}(C,G)=\Sigma_{i,j}|x_{i,j}-x_{i+1,j}|+|x_{i,j}-x_{i,j+1}|$

class TotalVariationLoss(nn.Module):
    def __init__(self,weight):
        super().__init__()
        self.weight = weight
    def forward(self,inputs):
        #矩阵计算，一步到位
        loss = 0.5*((inputs[:, :, 1:, :]-inputs[:, :, :-1, :]).abs().mean()+(inputs[:, :, :, 1:]-inputs[:, :, :, :-1]).abs().mean())
        return loss * self.weight

在loss function中对应作修改

#总变差损失计算
    noise_loss = TotalVariationLoss(noise_weight)
    noise_loss_this = noise_loss(input_param)
    
    loss = style_loss_this + content_loss_this + noise_loss_this