keras官方seq2seq模型实践

主要参考：https://blog.csdn.net/weiwei9363/article/details/79464789
以及keras官方代码:https://github.com/keras-team/keras/blob/master/examples/lstm_seq2seq.py

from keras.models import Model
from keras.layers import Input, LSTM, Dense
import numpy as np

batch_size = 256  # 训练的batch_size.
epochs = 100  # 训练轮数epoch.
latent_dim = 256  # 编码层LSTM的单元个数.
num_samples = 10000  # 训练数据集数量.
# 数据集文件，具体格式.
'''
xxxxx(target)\txxxxx(input)\n
'''
data_path = ''

# 处理数据，\n分行，\t分输入序列以及目标序列.
input_texts = []
target_texts = []
input_characters = set()
target_characters = set()
with open(data_path, 'r', encoding='utf-8') as f:
    lines = f.read().split('\n')
for line in lines[: min(num_samples, len(lines) - 1)]:
    target_text, input_text = line.split('\t')
    # 用'tab'作为 一个序列的开始字符
    # 用 '\n' 作为 序列的结束字符
    target_text = '\t' + target_text + '\n'
    input_texts.append(input_text)
    target_texts.append(target_text)
    #将新的字符加入字典，方便one-hot
    for char in input_text:
        if char not in input_characters:
            input_characters.add(char)
    for char in target_text:
        if char not in target_characters:
            target_characters.add(char)

#按照出现次数排序
input_characters = sorted(list(input_characters))
target_characters = sorted(list(target_characters))
#获得输入序列与目标序列的单词种类数
num_encoder_tokens = len(input_characters)
num_decoder_tokens = len(target_characters)
#获得最大长度以便统一
max_encoder_seq_length = max([len(txt) for txt in input_texts])
max_decoder_seq_length = max([len(txt) for txt in target_texts])

print('Number of samples:', len(input_texts))
print('Number of unique input tokens:', num_encoder_tokens)
print('Number of unique output tokens:', num_decoder_tokens)
print('Max sequence length for inputs:', max_encoder_seq_length)
print('Max sequence length for outputs:', max_decoder_seq_length)

#构建字典
input_token_index = dict(
    [(char, i) for i, char in enumerate(input_characters)])
target_token_index = dict(
    [(char, i) for i, char in enumerate(target_characters)])

#创建数组，使用one-hot编码
encoder_input_data = np.zeros(
    (len(input_texts), max_encoder_seq_length, num_encoder_tokens),
    dtype='float32')
decoder_input_data = np.zeros(
    (len(input_texts), max_decoder_seq_length, num_decoder_tokens),
    dtype='float32')
decoder_target_data = np.zeros(
    (len(input_texts), max_decoder_seq_length, num_decoder_tokens),
    dtype='float32')

#进行one-hot编码
for i, (input_text, target_text) in enumerate(zip(input_texts, target_texts)):
    for t, char in enumerate(input_text):
        encoder_input_data[i, t, input_token_index[char]] = 1.
    encoder_input_data[i, t + 1:, input_token_index[' ']] = 1.
    for t, char in enumerate(target_text):
        # decoder_target_data 比encoder_input_data快一步.
        decoder_input_data[i, t, target_token_index[char]] = 1.
        if t > 0:
            # decoder_target_data 不包含开始字符，并且比decoder_input_data提前一步
            decoder_target_data[i, t - 1, target_token_index[char]] = 1.
    decoder_input_data[i, t + 1:, target_token_index[' ']] = 1.
    decoder_target_data[i, t:, target_token_index[' ']] = 1.
# 定义编码器的输入，None表示任意长度输入.
encoder_inputs = Input(shape=(None, num_encoder_tokens))
encoder = LSTM(latent_dim, return_state=True)
encoder_outputs, state_h, state_c = encoder(encoder_inputs)
# 丢弃encoder_outputs, 我们只需要编码器的状态.
encoder_states = [state_h, state_c]

# 定义解码器的输入.
decoder_inputs = Input(shape=(None, num_decoder_tokens))
# 接下来建立解码器，解码器将返回整个输出序列.
# 并且返回其中间状态，中间状态在训练阶段不会用到，但是在推理阶段将是有用的
decoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True)

# 将编码器输出的状态作为初始解码器的初始状态.
decoder_outputs, _, _ = decoder_lstm(decoder_inputs,
                                     initial_state=encoder_states)

# 添加全连接层
decoder_dense = Dense(num_decoder_tokens, activation='softmax')
decoder_outputs = decoder_dense(decoder_outputs)

# 定义整个模型,
# 将`encoder_input_data` & `decoder_input_data` 训练目标 `decoder_target_data`
model = Model([encoder_inputs, decoder_inputs], decoder_outputs)

# 开始训练
model.compile(optimizer='rmsprop', loss='categorical_crossentropy',
              metrics=['accuracy'])
model.fit([encoder_input_data, decoder_input_data], decoder_target_data,
          batch_size=batch_size,
          epochs=epochs,
          validation_split=0.2)
# 保存模型
model.save('s2s.h5')

# Next: 建立推断模型 (sampling).
# 步骤如下:
# 1) 将输入编码，得到解码器所需初始状态.
# 2) 结合初始状态，对一个size=1的序列(其中只包含开始字符)做模型推断，得到的输出作为下一个size=1序列的内容
# 3) Repeat with the current target token and current states

# Define sampling models
encoder_model = Model(encoder_inputs, encoder_states)

decoder_state_input_h = Input(shape=(latent_dim,))
decoder_state_input_c = Input(shape=(latent_dim,))
decoder_states_inputs = [decoder_state_input_h, decoder_state_input_c]
decoder_outputs, state_h, state_c = decoder_lstm(
    decoder_inputs, initial_state=decoder_states_inputs)
decoder_states = [state_h, state_c]
decoder_outputs = decoder_dense(decoder_outputs)
decoder_model = Model(
    [decoder_inputs] + decoder_states_inputs,
    [decoder_outputs] + decoder_states)

# 建立 数字->字符 的字典，用于恢复
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 = decoder_model.predict(
            [target_seq] + states_value)

        # 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]

    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('Decoded sentence:', decoded_sentence)

输入test进行预测，注意重复之前的数据处理步骤，test句子在test_data[len(input_texts)-1:len(input_texts)]处

test_data = np.zeros(
    (len(input_texts), max_encoder_seq_length, num_encoder_tokens),
    dtype='float32')
test = 'a alloy purple '
for t, char in enumerate(test):
        test_data[i, t, input_token_index[char]] = 1.0
test_data[i, t + 1:, input_token_index[' ']] = 1.0
print(test_data.shape)
sentence = decode_sequence(test_data[len(input_texts)-1:len(input_texts)])
print(sentence)```