无法从“keras.layers”导入名称“Merge”

Question

我尝试运行代码，但发现Keras 的合并层有问题。我正在使用 python 3 和 keras 2.2.4

这是代码的解码部分

import numpy as np
import pandas as pd
from keras.models import Sequential
from keras.layers import LSTM, Embedding, TimeDistributed, Dense, RepeatVector, Merge, Activation
from keras.preprocessing import image, sequence
import cPickle as pickle


    def create_model(self, ret_model = False):

        image_model = Sequential()
        image_model.add(Dense(EMBEDDING_DIM, input_dim = 4096, activation='relu'))
        image_model.add(RepeatVector(self.max_length))

        lang_model = Sequential()
        lang_model.add(Embedding(self.vocab_size, 256, input_length=self.max_length))
        lang_model.add(LSTM(256,return_sequences=True))
        lang_model.add(TimeDistributed(Dense(EMBEDDING_DIM)))

        model = Sequential()
        model.add(Merge([image_model, lang_model], mode='concat'))
        model.add(LSTM(1000,return_sequences=False))
        model.add(Dense(self.vocab_size))
        model.add(Activation('softmax'))

        print ("Model created!")

这是错误信息

from keras.layers import LSTM, Embedding, TimeDistributed, Dense, RepeatVector, Merge, Activation
ImportError: cannot import name 'Merge' from 'keras.layers'

Answer 1

Keras +2 不支持Merge。相反，您需要使用Concatenate 层：

merged = Concatenate()([x1, x2]) # NOTE: the layer is first constructed and then it's called on its input

或等效的功能接口concatenate（以小写c开头）：

merged = concatenate([x1,x2]) # NOTE: the input of layer is passed as an argument, hence named *functional interface*

如果您对其他形式的合并感兴趣，例如加法、减法等，然后就可以使用相关的图层了。有关合并层的完整列表，请参阅documentation。