Is there a way to use the output of a given (middle) layer in Keras as the input of another network?

Question

Is there a way to use the output of a given (middle) layer in Keras as the input of another network?

For example use the last dense layer of the encoder here as the input for another network?

#encoder
encoder = keras.models.Sequential()
encoder.add(L.InputLayer(img_shape))
encoder.add(L.Conv2D(32, kernel_size=(3, 3),strides=1, padding='same', activation='elu'))
encoder.add(L.MaxPool2D(pool_size=(2, 2)))
encoder.add(L.Conv2D(64, kernel_size=(3, 3),strides=1, padding='same', activation='elu'))
encoder.add(L.MaxPool2D(pool_size=(2, 2)))
encoder.add(L.Conv2D(128, kernel_size=(3, 3),strides=1, padding='same', activation='elu'))
encoder.add(L.MaxPool2D(pool_size=(2, 2)))
encoder.add(L.Conv2D(256, kernel_size=(3, 3),strides=1, padding='same', activation='elu'))
encoder.add(L.MaxPool2D(pool_size=(2, 2)))
encoder.add(L.Flatten())
encoder.add(L.Dense(code_size))


    # decoder
decoder = keras.models.Sequential()
decoder.add(L.InputLayer((code_size,)))
decoder.add(L.Dense(147456))
decoder.add(L.Reshape((24, 24, 256)))
decoder.add(L.Conv2DTranspose(filters=128, kernel_size=(3, 3), strides=2, activation='elu', padding='same'))
decoder.add(L.Conv2DTranspose(filters=64, kernel_size=(3, 3), strides=2, activation='elu', padding='same'))
decoder.add(L.Conv2DTranspose(filters=32, kernel_size=(3, 3), strides=2, activation='elu', padding='same'))
decoder.add(L.Conv2DTranspose(filters=1, kernel_size=(3, 3), strides=2, activation=None, padding='same'))

Answer 1

如果您只想将网络的编码层用作输入（即，当训练模型的其余部分时，您不想通过反向传播来更改编码器网络的权重） ，则只需获取最后一层的输出即可对于您的所有示例，请在keras模型上使用predict_generator方法对解码器网络进行分析，并将其用作预测器网络的输入数据。

Answer 2

你试过用这个吗

decoder = keras.models.Sequential()(encoder)