python - 操纵神经网络的输出

Question

我有一个神经网络，输入为 (m, 2, 3, 96, 96) ，输出为 (m, 2, 128)。我试图通过减去输出 [m][0][0] - 输出 [m][0][1] 将该输出转换为 (m, 1, 128)，然后通过输入 (m, 1) 转换为 (m, 1) 1x128 输出到密集层

我尝试过网络和包装器中的 Lambda 和 keras.backend.Subtract 层

def faceRecoModel(input_shape):
    """
    Implementation of the Inception model used for FaceNet

    Arguments:
    input_shape -- shape of the images of the dataset

    Returns:
    model -- a Model() instance in Keras
    """

    # Define the input as a tensor with shape input_shape
    X_input = Input(input_shape)

    # Zero-Padding
    X = ZeroPadding2D((3, 3))(X_input)

    # First Block
    X = Conv2D(64, (7, 7), strides=(2, 2), name='conv1')(X)
    X = BatchNormalization(axis=1, name='bn1')(X)
    X = Activation('relu')(X)

    # Zero-Padding + MAXPOOL
    X = ZeroPadding2D((1, 1))(X)
    X = MaxPooling2D((3, 3), strides=2)(X)

    # Second Block
    X = Conv2D(64, (1, 1), strides=(1, 1), name='conv2')(X)
    X = BatchNormalization(axis=1, epsilon=0.00001, name='bn2')(X)
    X = Activation('relu')(X)

    # Zero-Padding + MAXPOOL
    X = ZeroPadding2D((1, 1))(X)

    # Second Block
    X = Conv2D(192, (3, 3), strides=(1, 1), name='conv3')(X)
    X = BatchNormalization(axis=1, epsilon=0.00001, name='bn3')(X)
    X = Activation('relu')(X)

    # Zero-Padding + MAXPOOL
    X = ZeroPadding2D((1, 1))(X)
    X = MaxPooling2D(pool_size=3, strides=2)(X)

    # Inception 1: a/b/c
    X = inception_block_1a(X)
    X = inception_block_1b(X)
    X = inception_block_1c(X)

    # Inception 2: a/b
    X = inception_block_2a(X)
    X = inception_block_2b(X)

    # Inception 3: a/b
    X = inception_block_3a(X)
    X = inception_block_3b(X)

    # Top layer
    X = AveragePooling2D(pool_size=(3, 3), strides=(1, 1), data_format='channels_first')(X)
    X = Flatten()(X)
    X = Dense(128, name='dense_layer')(X)

    # L2 normalization
    X = Lambda(lambda x: K.l2_normalize(x, axis=1))(X)

    # Create model instance
    model = Model(inputs=X_input, outputs=X, name='FaceRecoModel')

    return model



# now this is the wrapper I mentioned
model = faceRecoModel((3, 96, 96))
i = Input((2, 3, 96, 96))
o = TimeDistributed(model)(i)
model = Model(i, o)
model.compile(optimizer='adam', loss=pair_loss)

Answer 1

X = Lambda(lambda x: return x[:,0] - x[:,1])(X)
X = Dense(...)(X)