漫谈四种神经网络序列解码模型以及示例代码

, 'twenty_one twenty_two twenty_three twenty_four twenty_five']

input_list = []

tar_list = []

for tmp_input in input_text:

input_list.append(tokenize(tmp_input))

for tmp_tar in tar_text:

tar_list.append(tokenize(tmp_tar))

vocab = sorted(reduce(lambda x, y: x | y, (set(tmp_list) for tmp_list in input_list tar_list)))

# Reserve 0 for masking via pad_sequences

vocab_size = len(vocab) 1 # keras进行embedding的时刻必须进行len(vocab) 1

input_maxlen = max(map(len, (x for x in input_list)))

tar_maxlen = max(map(len, (x for x in tar_list)))

output_dim = vocab_size

hidden_dim = 20

print(‘-’)

print(‘Vocab size:’, vocab_size, ’unique words’)

print(‘Input max length:’, input_maxlen, ’words’)

print(‘Target max length:’, tar_maxlen, ’words’)

print(‘Dimension of hidden vectors:’, hidden_dim)

print(‘Number of training stories:’, len(input_list))

print(‘Number of test stories:’, len(input_list))

print(‘-’)

print(‘Vectorizing the word sequences…’)

word_to_idx = dict((c, i 1) for i, c in enumerate(vocab)) # 编码时须要将字符映射成数字index

idx_to_word = dict((i 1, c)

　　推荐阅读

　　如何使用batch-import工具向neo4j中导入海量数据

【引自T_SevenS的博客】在开辟neo4j的过程中，经常会有同窗问若何向neo4j中导入大年夜量的汗青数据，而这些数据一般都邑存在于关系型数据库中，如今本人就根据本身的导入经历，把导入的过>>>详细阅读

地址：http://www.17bianji.com/lsqh/34942.html