多输出 Keras 模型中的消失维度

Question

当我尝试训练下面描述的自动编码器时，我收到一个错误 ' 形状为 (256, 28, 28, 1) 的目标数组已传递给形状 (None, 0, 28, 1）同时用作损失`binary_crossentropy。这种损失期望目标具有与输出相同的形状。' 输入和输出维度都应该是 (28,28,1)，其中 256 是批量大小。运行 .summary() 确认解码器模型的输出是正确的 (28,28,1)，但是当编码器和解码器一起编译时，这似乎会发生变化。知道这里发生了什么吗？这三个函数在网络生成时依次调用。

def buildEncoder():
    input1 = Input(shape=(28,28,1))
    input2 = Input(shape=(28,28,1))
    merge = concatenate([input1,input2])
    convEncode1 = Conv2D(16, (3,3), activation = 'relu', padding = 'same')(merge)
    maxPoolEncode1 = MaxPooling2D(pool_size=(2, 1))(convEncode1)
    convEncode2 = Conv2D(16, (3,3), activation = 'sigmoid', padding = 'same')(maxPoolEncode1)
    convEncode3 = Conv2D(1, (3,3), activation = 'sigmoid', padding = 'same')(convEncode2)
    model = Model(inputs = [input1,input2], outputs = convEncode3)
    model.compile(loss='binary_crossentropy', optimizer=adam)
    return model

def buildDecoder():
    input1 = Input(shape=(28,28,1))
    upsample1 = UpSampling2D((2,1))(input1)
    convDecode1 = Conv2D(16, (3,3), activation = 'relu', padding = 'same')(upsample1)
    crop1 = Cropping2D(cropping = ((0,28),(0,0)))(convDecode1)
    crop2 = Cropping2D(cropping = ((28,0),(0,0)))(convDecode1)
    convDecode2_1 = Conv2D(16, (3,3), activation = 'relu', padding = 'same')(crop1)
    convDecode3_1 = Conv2D(16, (3,3), activation = 'relu', padding = 'same')(crop2)
    convDecode2_2 = Conv2D(1, (3,3), activation = 'sigmoid', padding = 'same')(convDecode2_1)
    convDecode3_2 = Conv2D(1, (3,3), activation = 'sigmoid', padding = 'same')(convDecode3_1)
    model = Model(inputs=input1, outputs=[convDecode2_2,convDecode3_2])
    model.compile(loss='binary_crossentropy', optimizer=adam)
    return model

def buildAutoencoder():   
    autoInput1 = Input(shape=(28,28,1))
    autoInput2 = Input(shape=(28,28,1))
    encode = encoder([autoInput1,autoInput2])
    decode = decoder(encode)
    model = Model(inputs=[autoInput1,autoInput2], outputs=[decode[0],decode[1]])
    model.compile(loss='binary_crossentropy', optimizer=adam)
    return model

运行 model.summary() 函数确认 this 的最终输出维度

Answer 1

您的编码器中的形状计算似乎有误。您假设解码器会得到 (None, 28, 28, 1)，但您的编码器实际上会输出 (None, 14, 28, 28, 1)。

print(encoder) # Tensor("model_1/conv2d_3/Sigmoid:0", shape=(?, 14, 28, 1), dtype=float32)

现在在您的解码器中，您正在裁剪等，假设您有 (28, 28, 1) 可能会将其削减为 0。模型自行工作，当您连接它们时会发生不匹配。