LSTM之后是平均池

问题描述

我正在使用Keras 1.0.我的问题与这一问题相同(如何在Keras中实现均值池层)，但答案似乎并非如此对我来说足够.

I'm using Keras 1.0. My problem is identical to this one (How to implement a Mean Pooling layer in Keras), but the answer there does not seem to be sufficient for me.

我想实现这个网络:

I want to implement this network:

以下代码不起作用:

sequence = Input(shape=(max_sent_len,), dtype='int32')
embedded = Embedding(vocab_size, word_embedding_size)(sequence)
lstm = LSTM(hidden_state_size, activation='sigmoid', inner_activation='hard_sigmoid', return_sequences=True)(embedded)
pool = AveragePooling1D()(lstm)
output = Dense(1, activation='sigmoid')(pool)

如果未设置return_sequences=True，则在调用AveragePooling1D()时会出现此错误:

If I don't set return_sequences=True, I get this error when I call AveragePooling1D():

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/PATH/keras/engine/topology.py", line 462, in __call__
    self.assert_input_compatibility(x)
  File "/PATH/keras/engine/topology.py", line 382, in assert_input_compatibility
    str(K.ndim(x)))
Exception: ('Input 0 is incompatible with layer averagepooling1d_6: expected ndim=3', ' found ndim=2')

否则，我在呼叫Dense()时收到此错误:

Otherwise, I get this error when I call Dense():

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/PATH/keras/engine/topology.py", line 456, in __call__
    self.build(input_shapes[0])
  File "/fs/clip-arqat/mossaab/trec/liveqa/cmu/venv/lib/python2.7/site-packages/keras/layers/core.py", line 512, in build
    assert len(input_shape) == 2
AssertionError

推荐答案

添加TimeDistributed(Dense(1))帮助:

sequence = Input(shape=(max_sent_len,), dtype='int32')
embedded = Embedding(vocab_size, word_embedding_size)(sequence)
lstm = LSTM(hidden_state_size, activation='sigmoid', inner_activation='hard_sigmoid', return_sequences=True)(embedded)
distributed = TimeDistributed(Dense(1))(lstm)
pool = AveragePooling1D()(distributed)
output = Dense(1, activation='sigmoid')(pool)

这篇关于LSTM之后是平均池的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！