分层softmax_分层注意网络中的输入层代表什么

embedding_layer=Embedding(len(word_index)+1,EMBEDDING_DIM,weights=[embedding_matrix],input_length=MAX_SENT_LENGTH,trainable=True)

sentence_input = Input(shape=(MAX_SENT_LENGTH,),dtype='int32',name='input1')

embedded_sequences = embedding_layer(sentence_input)

l_lstm = Bidirectional(LSTM(100))(embedded_sequences)

sentEncoder = Model(sentence_input,l_lstm)

review_input = Input(shape=(MAX_SENTS,MAX_SENT_LENGTH),name='input2')

review_encoder = TimeDistributed(sentEncoder)(review_input)

l_lstm_sent = Bidirectional(LSTM(100))(review_encoder)

preds = Dense(len(macronum),activation='softmax')(l_lstm_sent)

model = Model(review_input,preds)

我的问题是：这里的输入层代表什么？我猜测input1代表用嵌入层包装的句子,但是在这种情况下,input2是什么？它是sendEncoder的输出吗？在这种情况下,它应该是浮点数,或者如果它是嵌入单词的另一层,那么它也应该被嵌入层包裹.