http://blog.topspeedsnail.com/archives/10542

主题 TensorFlow

RNN不像传统的神经网络-它们的输出输出是固定的，而RNN允许我们输入输出向量序列。RNN是为了对序列数据进行建模而产生的。

样本序列性：样本间存在顺序关系，每个样本和它之前的样本存在关联。比如说，在文本中，一个词和它前面的词是有关联的；在气象数据中，一天的气温和前几天的气温是有关联的。

例如本帖要使用RNN生成古诗，你给它输入一堆古诗词，它会学着生成和前面相关联的字词。如果你给它输入一堆姓名，它会学着生成姓名；给它输入一堆古典乐/歌词，它会学着生成古典乐/歌词，甚至可以给它输入源代码。

关于RNN：

• TensorFlow练习3: RNN, Recurrent Neural Networks
• http://karpathy.github.io/2015/05/21/rnn-effectiveness/

本帖代码移植自 char-rnn ，它是基于Torch的洋文模型，稍加修改即可应用于中文。char-rnn使用文本文件做为输入、训练RNN模型，然后使用它生成和训练数据类似的文本。

使用的数据集：全唐诗(43030首)：https://pan.baidu.com/s/1o7QlUhO

训练：

import os

import collections

import numpyas np

import tensorflowas tf

from tensorflow.python.opsimport rnn_cell

from tensorflow.python.opsimport seq2seq

import time

 

#-------------------------------数据预处理---------------------------#

 

poetry_file ='poetry.txt'

 

# 诗集

poetrys = []

with open(poetry_file, "r", encoding='utf-8',) as f:

 for linein f:

 try:

 title, content = line.strip().split(':')

 content = content.replace(' ','')

 if '_' in contentor '(' in contentor '（' in contentor '《' in contentor '[' in content:

 continue

 if len(content) < 5 or len(content) > 79:

 continue

 content = '[' + content + ']'

 poetrys.append(content)

 except Exception as e:

 pass

 

# 按诗的字数排序

poetrys = sorted(poetrys,key=lambda line: len(line))

print('唐诗总数: ', len(poetrys))

 

# 统计每个字出现次数

all_words = []

for poetryin poetrys:

 all_words += [wordfor wordin poetry]

counter = collections.Counter(all_words)

count_pairs = sorted(counter.items(), key=lambda x: -x[1])

words, _ = zip(*count_pairs)

 

# 取前多少个常用字

words = words[:len(words)] + (' ',)

# 每个字映射为一个数字ID

word_num_map = dict(zip(words, range(len(words))))

# 把诗转换为向量形式，参考TensorFlow练习1

to_num = lambda word: word_num_map.get(word, len(words))

poetrys_vector = [ list(map(to_num, poetry)) for poetryin poetrys]

#[[314, 3199, 367, 1556, 26, 179, 680, 0, 3199, 41, 506, 40, 151, 4, 98, 1],

#[339, 3, 133, 31, 302, 653, 512, 0, 37, 148, 294, 25, 54, 833, 3, 1, 965, 1315, 377, 1700, 562, 21, 37, 0, 2, 1253, 21, 36, 264, 877, 809, 1]

#....]

 

# 每次取64首诗进行训练

batch_size = 64

n_chunk = len(poetrys_vector) // batch_size

x_batches = []

y_batches = []

for i in range(n_chunk):

 start_index = i * batch_size

 end_index = start_index + batch_size

 

 batches = poetrys_vector[start_index:end_index]

 length = max(map(len,batches))

 xdata = np.full((batch_size,length), word_num_map[' '], np.int32)

 for rowin range(batch_size):

 xdata[row,:len(batches[row])] = batches[row]

 ydata = np.copy(xdata)

 ydata[:,:-1] = xdata[:,1:]

 """

xdata             ydata

[6,2,4,6,9]       [2,4,6,9,9]

[1,4,2,8,5]       [4,2,8,5,5]

"""

 x_batches.append(xdata)

 y_batches.append(ydata)

 

 

#---------------------------------------RNN--------------------------------------#

 

input_data = tf.placeholder(tf.int32, [batch_size, None])

output_targets = tf.placeholder(tf.int32, [batch_size, None])

# 定义RNN

def neural_network(model='lstm', rnn_size=128, num_layers=2):

 if model == 'rnn':

 cell_fun = rnn_cell.BasicRNNCell

 elif model == 'gru':

 cell_fun = rnn_cell.GRUCell

 elif model == 'lstm':

 cell_fun = rnn_cell.BasicLSTMCell

 

 cell = cell_fun(rnn_size, state_is_tuple=True)

 cell = rnn_cell.MultiRNNCell([cell] * num_layers, state_is_tuple=True)

 

 initial_state = cell.zero_state(batch_size, tf.float32)

 

 with tf.variable_scope('rnnlm'):

 softmax_w = tf.get_variable("softmax_w", [rnn_size, len(words)+1])

 softmax_b = tf.get_variable("softmax_b", [len(words)+1])

 with tf.device("/cpu:0"):

 embedding = tf.get_variable("embedding", [len(words)+1, rnn_size])

 inputs = tf.nn.embedding_lookup(embedding, input_data)

 

 outputs, last_state = tf.nn.dynamic_rnn(cell, inputs, initial_state=initial_state, scope='rnnlm')

 output = tf.reshape(outputs,[-1, rnn_size])

 

 logits = tf.matmul(output, softmax_w) + softmax_b

 probs = tf.nn.softmax(logits)

 return logits, last_state, probs, cell, initial_state

#训练

def train_neural_network():

 logits, last_state, _, _, _ = neural_network()

 targets = tf.reshape(output_targets, [-1])

 loss = seq2seq.sequence_loss_by_example([logits], [targets], [tf.ones_like(targets, dtype=tf.float32)], len(words))

 cost = tf.reduce_mean(loss)

 learning_rate = tf.Variable(0.0, trainable=False)

 tvars = tf.trainable_variables()

 grads, _ = tf.clip_by_global_norm(tf.gradients(cost, tvars), 5)

 optimizer = tf.train.AdamOptimizer(learning_rate)

 train_op = optimizer.apply_gradients(zip(grads, tvars))

 

 with tf.Session() as sess:

 sess.run(tf.initialize_all_variables())

 

 saver = tf.train.Saver(tf.all_variables())

 

 for epochin range(50):

 sess.run(tf.assign(learning_rate, 0.002 * (0.97 ** epoch)))

 n = 0

 for batchein range(n_chunk):

 train_loss, _ , _ = sess.run([cost, last_state, train_op], feed_dict={input_data: x_batches[n], output_targets: y_batches[n]})

 n += 1

 print(epoch, batche, train_loss)

 if epoch % 7 == 0:

 saver.save(sess, 'poetry.module', global_step=epoch)

 

train_neural_network()

使用训练好的模型生成古诗：

import os

import collections

import numpyas np

import tensorflowas tf

from tensorflow.python.opsimport rnn_cell

from tensorflow.python.opsimport seq2seq

import time

 

#-------------------------------数据预处理---------------------------#

 

poetry_file ='poetry.txt'

 

# 诗集

poetrys = []

with open(poetry_file, "r", encoding='utf-8',) as f:

 for linein f:

 try:

 title, content = line.strip().split(':')

 content = content.replace(' ','')

 if '_' in contentor '(' in contentor '（' in contentor '《' in contentor '[' in content:

 continue

 if len(content) < 5 or len(content) > 79:

 continue

 content = '[' + content + ']'

 poetrys.append(content)

 except Exception as e:

 pass

 

# 按诗的字数排序

poetrys = sorted(poetrys,key=lambda line: len(line))

print('唐诗总数: ', len(poetrys))

 

# 统计每个字出现次数

all_words = []

for poetryin poetrys:

 all_words += [wordfor wordin poetry]

counter = collections.Counter(all_words)

count_pairs = sorted(counter.items(), key=lambda x: -x[1])

words, _ = zip(*count_pairs)

 

# 取前多少个常用字

words = words[:len(words)] + (' ',)

# 每个字映射为一个数字ID

word_num_map = dict(zip(words, range(len(words))))

# 把诗转换为向量形式，参考TensorFlow练习1

to_num = lambda word: word_num_map.get(word, len(words))

poetrys_vector = [ list(map(to_num, poetry)) for poetryin poetrys]

#[[314, 3199, 367, 1556, 26, 179, 680, 0, 3199, 41, 506, 40, 151, 4, 98, 1],

#[339, 3, 133, 31, 302, 653, 512, 0, 37, 148, 294, 25, 54, 833, 3, 1, 965, 1315, 377, 1700, 562, 21, 37, 0, 2, 1253, 21, 36, 264, 877, 809, 1]

#....]

 

# 每次取64首诗进行训练

batch_size = 1

n_chunk = len(poetrys_vector) // batch_size

x_batches = []

y_batches = []

for i in range(n_chunk):

 start_index = i * batch_size

 end_index = start_index + batch_size

 

 batches = poetrys_vector[start_index:end_index]

 length = max(map(len,batches))

 xdata = np.full((batch_size,length), word_num_map[' '], np.int32)

 for rowin range(batch_size):

 xdata[row,:len(batches[row])] = batches[row]

 ydata = np.copy(xdata)

 ydata[:,:-1] = xdata[:,1:]

 """

xdata             ydata

[6,2,4,6,9]       [2,4,6,9,9]

[1,4,2,8,5]       [4,2,8,5,5]

"""

 x_batches.append(xdata)

 y_batches.append(ydata)

 

 

#---------------------------------------RNN--------------------------------------#

 

input_data = tf.placeholder(tf.int32, [batch_size, None])

output_targets = tf.placeholder(tf.int32, [batch_size, None])

# 定义RNN

def neural_network(model='lstm', rnn_size=128, num_layers=2):

 if model == 'rnn':

 cell_fun = rnn_cell.BasicRNNCell

 elif model == 'gru':

 cell_fun = rnn_cell.GRUCell

 elif model == 'lstm':

 cell_fun = rnn_cell.BasicLSTMCell

 

 cell = cell_fun(rnn_size, state_is_tuple=True)

 cell = rnn_cell.MultiRNNCell([cell] * num_layers, state_is_tuple=True)

 

 initial_state = cell.zero_state(batch_size, tf.float32)

 

 with tf.variable_scope('rnnlm'):

 softmax_w = tf.get_variable("softmax_w", [rnn_size, len(words)+1])

 softmax_b = tf.get_variable("softmax_b", [len(words)+1])

 with tf.device("/cpu:0"):

 embedding = tf.get_variable("embedding", [len(words)+1, rnn_size])

 inputs = tf.nn.embedding_lookup(embedding, input_data)

 

 outputs, last_state = tf.nn.dynamic_rnn(cell, inputs, initial_state=initial_state, scope='rnnlm')

 output = tf.reshape(outputs,[-1, rnn_size])

 

 logits = tf.matmul(output, softmax_w) + softmax_b

 probs = tf.nn.softmax(logits)

 return logits, last_state, probs, cell, initial_state

 

#-------------------------------生成古诗---------------------------------#

# 使用训练完成的模型

 

def gen_poetry():

 def to_word(weights):

 t = np.cumsum(weights)

 s = np.sum(weights)

 sample = int(np.searchsorted(t, np.random.rand(1)*s))

 return words[sample]

 

 _, last_state, probs, cell, initial_state = neural_network()

 

 with tf.Session() as sess:

 sess.run(tf.initialize_all_variables())

 

 saver = tf.train.Saver(tf.all_variables())

 saver.restore(sess, 'poetry.module-49')

 

 state_ = sess.run(cell.zero_state(1, tf.float32))

 

 x = np.array([list(map(word_num_map.get, '['))])

 [probs_, state_] = sess.run([probs, last_state], feed_dict={input_data: x, initial_state: state_})

 word = to_word(probs_)

 #word = words[np.argmax(probs_)]

 poem = ''

 while word != ']':

 poem += word

 x = np.zeros((1,1))

 x[0,0] = word_num_map[word]

 [probs_, state_] = sess.run([probs, last_state], feed_dict={input_data: x, initial_state: state_})

 word = to_word(probs_)

 #word = words[np.argmax(probs_)]

 return poem

 

print(gen_poetry())

生成的“诗词”（至少格式对上了）：

新犬随风起，一璃迹阵悲。

浅昏罢庄哉，清插去园空。

双叶坐成鉴，王妓水正苑。

鸟声不成影，胙滩朱瓮声。

无斑红芜踏，那期日正闲。

吾燕登无士，无处得赵名。

生成藏头诗：

def gen_poetry_with_head(head):

 def to_word(weights):

 t = np.cumsum(weights)

 s = np.sum(weights)

 sample = int(np.searchsorted(t, np.random.rand(1)*s))

 return words[sample]

 

 _, last_state, probs, cell, initial_state = neural_network()

 

 with tf.Session() as sess:

 sess.run(tf.initialize_all_variables())

 

 saver = tf.train.Saver(tf.all_variables())

 saver.restore(sess, 'poetry.module-7')

 

 state_ = sess.run(cell.zero_state(1, tf.float32))

 poem = ''

 i = 0

 for wordin head:

 while word != '，' and word != '。':

 poem += word

 x = np.array([list(map(word_num_map.get, word))])

 [probs_, state_] = sess.run([probs, last_state], feed_dict={input_data: x, initial_state: state_})

 word = to_word(probs_)

 time.sleep(1)

 if i % 2 == 0:

 poem += '，'

 else:

 poem += '。'

 i += 1

 return poem

 

print(gen_poetry_with_head('一二三四'))