Lambda会在这段代码中做什么python keras？

问题描述

def AdaIN(x):
    #normalize x[0] (image representation)
    mean = K.mean(x[0],axis = [1,2],keepdims = True)
    std = K.std(x[0],keepdims = True) + 1e-7
    y = (x[0] - mean) / std
    
    #Reshape scale and bias parameters
    pool_shape = [-1,1,y.shape[-1]]
    scale = K.reshape(x[1],pool_shape)
    bias = K.reshape(x[2],pool_shape)#Multiply by x[1] (GAMMA) and add x[2] (BETA)
    return y * scale + bias

    

def g_block(input_tensor,latent_vector,filters):
    gamma = Dense(filters,bias_initializer = 'ones')(latent_vector)
    beta = Dense(filters)(latent_vector)
    
    out = UpSampling2D()(input_tensor)
    out = Conv2D(filters,3,padding = 'same')(out)
    out = Lambda(AdaIN)([out,gamma,beta])
    out = Activation('relu')(out)
    
    return out

请参见上面的代码。我目前正在学习styleGAN。我正在尝试将此代码转换为pytorch，但我似乎无法理解Lambda在g_block中做什么。 AdaIN根据其声明仅需要一个输入，但是如何将gamma和beta用作输入？请告诉我Lambda在此代码中的作用。

非常感谢您。

解决方法

keras中的

Lambda层用于在模型内部调用自定义函数。在g_block中，Lambda调用AdaIN函数，并将out,gamma,beta作为参数传递到列表中。 AdaIN函数将这三个张量封装为x封装在一个列表中。而且，这些张量也可以通过索引列表AdaIN（x [0]，x [1]，x [2]）在x函数内部访问。

相当于pytorch：

import torch
import torch.nn as nn
import torch.nn.functional as F

class AdaIN(nn.Module):
    def forward(self,out,beta):
        bs,ch = out.size()[:2]
        mean   = out.reshape(bs,ch,-1).mean(dim=2).reshape(bs,1,1)
        std    = out.reshape(bs,-1).std(dim=2).reshape(bs,1) + 1e-7
        y      = (out - mean) / std
        bias   = beta.unsqueeze(-1).unsqueeze(-1).expand_as(out)
        scale  = gamma.unsqueeze(-1).unsqueeze(-1).expand_as(out)
        return y * scale + bias

           

class g_block(nn.Module):
    def __init__(self,filters,latent_vector_shape,input_tensor_channels):
        super().__init__()
        self.gamma = nn.Linear(in_features = latent_vector_shape,out_features = filters)
        # Initializes all bias to 1
        self.gamma.bias.data = torch.ones(filters)
        self.beta  = nn.Linear(in_features = latent_vector_shape,out_features = filters)
        # calculate appropriate padding 
        self.conv  = nn.Conv2d(input_tensor_channels,3,padding=1)# calc padding
        self.adain = AdaIN()

    def forward(self,input_tensor,latent_vector):
        gamma = self.gamma(latent_vector)
        beta  = self.beta(latent_vector)
        # check default interpolation mode in keras and replace mode below if different
        out   = F.interpolate(input_tensor,scale_factor=2,mode='nearest') 
        out   = self.conv(out)
        out   = self.adain(out,beta)
        out   = torch.relu(out)        
        return out

# Sample:
input_tensor  = torch.randn((1,10,10))
latent_vector = torch.randn((1,5))
g   = g_block(3,latent_vector.shape[1],input_tensor.shape[1])
out = g(input_tensor,latent_vector)
print(out)

注意：创建latent_vector时需要传递input_tensor和g_block形状。