用Keras实现LSTM的多层Seq2Seq模型

Question

我正在使用Keras制作seq2seq模型。我已经构建了单层编码器和解码器，它们工作得很好。但现在我想扩展为多层编码器和解码器。我使用Keras函数API进行构建。

训练：

编码器代码：

encoder_input=Input(shape=(None,vec_dimension))
encoder_lstm=LSTM(vec_dimension,return_state=True,return_sequences=True)(encoder_input)
encoder_lstm=LSTM(vec_dimension,return_state=True)(encoder_lstm)
encoder_output,encoder_h,encoder_c=encoder_lstm

代码解码器：

encoder_state=[encoder_h,encoder_c]
decoder_input=Input(shape=(None,vec_dimension))
decoder_lstm= LSTM(vec_dimension,return_state=True,return_sequences=True (decoder_input,initial_state=encoder_state)
decoder_lstm=LSTM(vec_dimension,return_state=True,return_sequences=True)(decoder_lstm)
decoder_output,_,_=decoder_lstm

用于测试：

encoder_model=Model(inputs=encoder_input,outputs=encoder_state)
decoder_state_input_h=Input(shape=(None,vec_dimension))
decoder_state_input_c=Input(shape=(None,vec_dimension))
decoder_states_input=[decoder_state_input_h,decoder_state_input_c]
decoder_output,decoder_state_h,decoder_state_c =decoder_lstm #(decoder_input,initial_state=decoder_states_input)
decoder_states=[decoder_state_h,decoder_state_c]
decoder_model=Model(inputs=[decoder_input]+decoder_states_input,outputs=[decoder_output]+decoder_states)

现在，当我尝试增加解码器中的层数进行训练时，训练可以正常进行，但在测试时就无法正常工作并出现错误。

实际上，问题出在我将initial_state从最后一层移到了中间一层，而在测试时调用它会导致出现错误。

RuntimeError: Graph disconnected: cannot obtain value for tensor Tensor("input_64:0", shape=(?, ?, 150), dtype=float32) at layer "input_64".The following previous layers were accessed without issue: []

我应该如何传递initial_state=decoder_states_input到输入层，以避免出现错误？我应该如何在最后一个层中传递initial_state=decoder_states_input到第一个输入层？在这段代码中，我尝试创建多层解码器LSTM，但是这导致错误。当使用单层时，正确的代码为： Encoder(Training)：

encoder_input=Input(shape=(None,vec_dimension))
encoder_lstm =LSTM(vec_dimension,return_state=True)(encoder_input)
encoder_output,encoder_h,encoder_c=encoder_lstm

解码器（训练）：-

encoder_state=[encoder_h,encoder_c]
decoder_input=Input(shape=(None,vec_dimension))
decoder_lstm= LSTM(vec_dimension, return_state=True, return_sequences=True)
decoder_output,_,_=decoder_lstm(decoder_input,initial_state=encoder_state)

解码器（测试）

decoder_output,decoder_state_h,decoder_state_c=decoder_lstm( decoder_input, initial_state=decoder_states_input)
decoder_states=[decoder_state_h,decoder_state_c]
decoder_output,decoder_state_h,decoder_state_c=decoder_lstm (decoder_input,initial_state=decoder_states_input)
decoder_model=Model(inputs=[decoder_input]+decoder_states_input,outputs=[decoder_output]+decoder_states)

- SAGAR

---

如果我理解正确，您提供的代码可以工作。能否请您添加不起作用的代码以澄清问题是什么？ - nisargjhaveri

其实那个代码是错的，因为我在那里添加了多层的解码器。我已经为单层添加了正确的代码。如何将该代码扩展到多层LSTM？ - SAGAR

你成功解决了这个问题吗？我也遇到了同样的问题。 - Amel Music

Answer 1

编辑 - 更新为在Keras中使用函数式API模型而不是RNN

from keras.models import Model
from keras.layers import Input, LSTM, Dense, RNN
layers = [256,128] # we loop LSTMCells then wrap them in an RNN layer

encoder_inputs = Input(shape=(None, num_encoder_tokens))

e_outputs, h1, c1 = LSTM(latent_dim, return_state=True, return_sequences=True)(encoder_inputs) 
_, h2, c2 = LSTM(latent_dim, return_state=True)(e_outputs) 
encoder_states = [h1, c1, h2, c2]

decoder_inputs = Input(shape=(None, num_decoder_tokens))

out_layer1 = LSTM(latent_dim, return_sequences=True, return_state=True)
d_outputs, dh1, dc1 = out_layer1(decoder_inputs,initial_state= [h1, c1])
out_layer2 = LSTM(latent_dim, return_sequences=True, return_state=True)
final, dh2, dc2 = out_layer2(d_outputs, initial_state= [h2, c2])
decoder_dense = Dense(num_decoder_tokens, activation='softmax')
decoder_outputs = decoder_dense(final)


model = Model([encoder_inputs, decoder_inputs], decoder_outputs)

model.summary()

以下是推理设置：

encoder_model = Model(encoder_inputs, encoder_states)

decoder_state_input_h = Input(shape=(latent_dim,))
decoder_state_input_c = Input(shape=(latent_dim,))
decoder_state_input_h1 = Input(shape=(latent_dim,))
decoder_state_input_c1 = Input(shape=(latent_dim,))
decoder_states_inputs = [decoder_state_input_h, decoder_state_input_c, 
                         decoder_state_input_h1, decoder_state_input_c1]
d_o, state_h, state_c = out_layer1(
    decoder_inputs, initial_state=decoder_states_inputs[:2])
d_o, state_h1, state_c1 = out_layer2(
    d_o, initial_state=decoder_states_inputs[-2:])
decoder_states = [state_h, state_c, state_h1, state_c1]
decoder_outputs = decoder_dense(d_o)
decoder_model = Model(
    [decoder_inputs] + decoder_states_inputs,
    [decoder_outputs] + decoder_states)

decoder_model.summary()

最后，如果您在遵循Keras seq2seq示例，则需要更改预测脚本，因为有多个隐藏状态需要管理，而不仅仅是单层示例中的两个。将会有2倍于层数的隐藏状态。

# Reverse-lookup token index to decode sequences back to
# something readable.
reverse_input_char_index = dict(
    (i, char) for char, i in input_token_index.items())
reverse_target_char_index = dict(
    (i, char) for char, i in target_token_index.items())

def decode_sequence(input_seq):
    # Encode the input as state vectors.
    states_value = encoder_model.predict(input_seq)

    # Generate empty target sequence of length 1.
    target_seq = np.zeros((1, 1, num_decoder_tokens))
    # Populate the first character of target sequence with the start character.
    target_seq[0, 0, target_token_index['\t']] = 1.

    # Sampling loop for a batch of sequences
    # (to simplify, here we assume a batch of size 1).
    stop_condition = False
    decoded_sentence = ''
    while not stop_condition:
        output_tokens, h, c, h1, c1 = decoder_model.predict(
            [target_seq] + states_value) #######NOTICE THE ADDITIONAL HIDDEN STATES

        # Sample a token
        sampled_token_index = np.argmax(output_tokens[0, -1, :])
        sampled_char = reverse_target_char_index[sampled_token_index]
        decoded_sentence += sampled_char

        # Exit condition: either hit max length
        # or find stop character.
        if (sampled_char == '\n' or
           len(decoded_sentence) > max_decoder_seq_length):
            stop_condition = True

        # Update the target sequence (of length 1).
        target_seq = np.zeros((1, 1, num_decoder_tokens))
        target_seq[0, 0, sampled_token_index] = 1.

        # Update states
        states_value = [h, c, h1, c1]#######NOTICE THE ADDITIONAL HIDDEN STATES

    return decoded_sentence


for seq_index in range(100):
    # Take one sequence (part of the training set)
    # for trying out decoding.
    input_seq = encoder_input_data[seq_index: seq_index + 1]
    decoded_sentence = decode_sequence(input_seq)
    print('-')
    print('Input sentence:', input_texts[seq_index])
    print('Target sentence:', target_texts[seq_index])
    print('Decoded sentence:', decoded_sentence)

- JW_

---

谢谢您提供详细的回答。能否请您确认一下，按照您的方法创建两个LSTMCells并将它们包装在一个RNN层中，在功能上是否与创建两个LSTM层（第一层馈入第二层）相同？ - Aesir

如果您希望使用CuDNN层，则此方法不起作用，因为在keras中它们没有提供等效的Cell。 - Aesir

我更新了答案，使用了具有多个LSTM层的函数API。测试过了，翻译效果良好。 - JW_

谢谢@JeremyWortz。我已经为此苦苦挣扎了一个月，终于解决了。看来我的主要失误是没有考虑到推理中的额外层状态。 - jwanga

这非常有帮助！我已经修改了您的代码，使其适用于深度n而不是固定的2，因此它会循环遍历一个latent_dims数组，其长度定义了堆叠LSTM层数。我将在明天结束时使用它，我会确保清理并在之后发布（明天）！PS：您的“layers”数组实际上没有被使用，并且两个层都使用相同的latent_dim变量。 - Sam Ragusa

这很有帮助，不过第二/更高的编码器层状态不应该成为第一/更高解码器层的状态输入吗？第二个解码器层是较低的层，提供输入到密集输出层，所以我期望它从同一级别的编码器层接收状态 - 离输入/离输出相隔一个。也许我误解了这种方法的某些方面。 - Todd Gillette

Answer 2

我将Jeremy Wortz的精彩回答推广到了从列表"latent_dims"创建模型，该模型的深度为"len(latent_dims)"，而不是固定的2。

从"latent_dims"声明开始：

# latent_dims is an array which defines the depth of the encoder/decoder, as well as how large
# the layers should be.   So an array of sizes [a,b,c]  would produce a depth-3 encoder and decoder
# with layer sizes equal to [a,b,c] and [c,b,a] respectively.
latent_dims = [1024, 512,  256]

创建训练模型：

# Define an input sequence and process it by going through a len(latent_dims)-layer deep encoder
encoder_inputs = Input(shape=(None, num_encoder_tokens))

outputs = encoder_inputs
encoder_states = []
for j in range(len(latent_dims))[::-1]:
    outputs, h, c = LSTM(latent_dims[j], return_state=True, return_sequences=bool(j))(outputs)
    encoder_states += [h, c]

# Set up the decoder, setting the initial state of each layer to the state of the layer in the encoder
# which is it's mirror (so for encoder: a->b->c, you'd have decoder initial states: c->b->a).
decoder_inputs = Input(shape=(None, num_decoder_tokens))

outputs = decoder_inputs
output_layers = []
for j in range(len(latent_dims)):
    output_layers.append(
        LSTM(latent_dims[len(latent_dims) - j - 1], return_sequences=True, return_state=True)
    )
    outputs, dh, dc = output_layers[-1](outputs, initial_state=encoder_states[2*j:2*(j+1)])


decoder_dense = Dense(num_decoder_tokens, activation='softmax')
decoder_outputs = decoder_dense(outputs)

# Define the model that will turn
# `encoder_input_data` & `decoder_input_data` into `decoder_target_data`
model = Model([encoder_inputs, decoder_inputs], decoder_outputs)

对于推理，它如下所示：

# Define sampling models (modified for n-layer deep network)
encoder_model = Model(encoder_inputs, encoder_states)


d_outputs = decoder_inputs
decoder_states_inputs = []
decoder_states = []
for j in range(len(latent_dims))[::-1]:
    current_state_inputs = [Input(shape=(latent_dims[j],)) for _ in range(2)]

    temp = output_layers[len(latent_dims)-j-1](d_outputs, initial_state=current_state_inputs)

    d_outputs, cur_states = temp[0], temp[1:]

    decoder_states += cur_states
    decoder_states_inputs += current_state_inputs

decoder_outputs = decoder_dense(d_outputs)
decoder_model = Model(
    [decoder_inputs] + decoder_states_inputs,
    [decoder_outputs] + decoder_states)

最后，我们对Jeremy Wortz的“decode_sequence”函数进行了一些修改，得到了以下结果：

def decode_sequence(input_seq, encoder_model, decoder_model):
    # Encode the input as state vectors.
    states_value = encoder_model.predict(input_seq)

    # Generate empty target sequence of length 1.
    target_seq = np.zeros((1, 1, num_decoder_tokens))
    # Populate the first character of target sequence with the start character.
    target_seq[0, 0, target_token_index['\t']] = 1.

    # Sampling loop for a batch of sequences
    # (to simplify, here we assume a batch of size 1).
    stop_condition = False
    decoded_sentence = []  #Creating a list then using "".join() is usually much faster for string creation
    while not stop_condition:
        to_split = decoder_model.predict([target_seq] + states_value)

        output_tokens, states_value = to_split[0], to_split[1:]

        # Sample a token
        sampled_token_index = np.argmax(output_tokens[0, 0])
        sampled_char = reverse_target_char_index[sampled_token_index]
        decoded_sentence.append(sampled_char)

        # Exit condition: either hit max length
        # or find stop character.
        if sampled_char == '\n' or len(decoded_sentence) > max_decoder_seq_length:
            stop_condition = True

        # Update the target sequence (of length 1).
        target_seq = np.zeros((1, 1, num_decoder_tokens))
        target_seq[0, 0, sampled_token_index] = 1.

    return "".join(decoded_sentence)

- Sam Ragusa

---

嗨，如果我们没有使用dh和dc，为什么在“LSTM（latent_dims [len（latent_dims） - j - 1]，return_sequences = True，return_state = True）”中会有“return_state”呢？ - Prox