文本是一类序列数据,一篇文章可以看作是字符或单词的序列,本节将介绍文本数据的常见预处理步骤,预处理通常包括四个步骤:
读入文本 分词 建立字典,将每个词映射到一个唯一的索引(index) 将文本从词的序列转换为索引的序列,方便输入模型以下代码是读取文章的demo
import collections
import re
def read_time_machine():
with open('/home/kesci/input/timemachine7163/timemachine.txt', 'r') as f:
lines = [re.sub('[^a-z]+', ' ', line.strip().lower()) for line in f]
return lines
lines = read_time_machine()
print('# sentences %d' % len(lines))
分词
def tokenize(sentences, token='word'):
"""Split sentences into word or char tokens"""
if token == 'word':
return [sentence.split(' ') for sentence in sentences]
elif token == 'char':
return [list(sentence) for sentence in sentences]
else:
print('ERROR: unkown token type '+token)
tokens = tokenize(lines)
tokens[0:2]
建立字典并映射唯一引索
class Vocab(object):
def __init__(self, tokens, min_freq=0, use_special_tokens=False):
counter = count_corpus(tokens) # :
self.token_freqs = list(counter.items())
self.idx_to_token = []
if use_special_tokens:
# padding, begin of sentence, end of sentence, unknown
self.pad, self.bos, self.eos, self.unk = (0, 1, 2, 3)
self.idx_to_token += ['', '', '', '']
else:
self.unk = 0
self.idx_to_token += ['']
self.idx_to_token += [token for token, freq in self.token_freqs
if freq >= min_freq and token not in self.idx_to_token]
self.token_to_idx = dict()
for idx, token in enumerate(self.idx_to_token):
self.token_to_idx[token] = idx
def __len__(self):
return len(self.idx_to_token)
def __getitem__(self, tokens):
if not isinstance(tokens, (list, tuple)):
return self.token_to_idx.get(tokens, self.unk)
return [self.__getitem__(token) for token in tokens]
def to_tokens(self, indices):
if not isinstance(indices, (list, tuple)):
return self.idx_to_token[indices]
return [self.idx_to_token[index] for index in indices]
def count_corpus(sentences):
tokens = [tk for st in sentences for tk in st]
return collections.Counter(tokens) # 返回一个字典,记录每个词的出现次数
将词转化为引索
for i in range(8, 10):
print('words:', tokens[i])
print('indices:', vocab[tokens[i]])
用NLTK做分词:
from nltk.tokenize import word_tokenize
from nltk import data
data.path.append('/home/kesci/input/nltk_data3784/nltk_data')
print(word_tokenize(text))
总结:关于类的构建很有学习价值,可以多看看
语言模型假设序列w1,w2,…,wTw_1, w_2, \ldots, w_Tw1,w2,…,wT中的每个词是依次生成的,我们有
P(w1,w2,…,wT)=∏t=1TP(wt∣w1,…,wt−1)=P(w1)P(w2∣w1)⋯P(wT∣w1w2⋯wT−1)
P(w_1, w_2, \ldots, w_T)= \prod_{t=1}^T P(w_t \mid w_1, \ldots, w_{t-1})\\ = P(w_1)P(w_2 \mid w_1) \cdots P(w_T \mid w_1w_2\cdots w_{T-1})
P(w1,w2,…,wT)=t=1∏TP(wt∣w1,…,wt−1)=P(w1)P(w2∣w1)⋯P(wT∣w1w2⋯wT−1)
例如,一段含有4个词的文本序列的概率
P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w1,w2,w3).
P(w_1, w_2, w_3, w_4) = P(w_1) P(w_2 \mid w_1) P(w_3 \mid w_1, w_2) P(w_4 \mid w_1, w_2, w_3).
P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w1,w2,w3).
语言模型的参数就是词的概率以及给定前几个词情况下的条件概率。设训练数据集为一个大型文本语料库,如维基百科的所有条目,词的概率可以通过该词在训练数据集中的相对词频来计算,例如,w1w_1w1的概率可以计算为:
P^(w1)=n(w1)n
\hat P(w_1) = \frac{n(w_1)}{n}
P^(w1)=nn(w1)
其中n(w1)n(w_1)n(w1)为语料库中以w1w_1w1作为第一个词的文本的数量,nnn为语料库中文本的总数量。
类似的,给定w1w_1w1情况下,w2w_2w2的条件概率可以计算为:
P^(w2∣w1)=n(w1,w2)n(w1)
\hat P(w_2 \mid w_1) = \frac{n(w_1, w_2)}{n(w_1)}
P^(w2∣w1)=n(w1)n(w1,w2)
其中n(w1,w2)n(w_1, w_2)n(w1,w2)为语料库中以w1w_1w1作为第一个词,w2w_2w2作为第二个词的文本的数量。
序列长度增加,计算和存储多个词共同出现的概率的复杂度会呈指数级增加。nnn元语法通过马尔可夫假设简化模型,马尔科夫假设是指一个词的出现只与前面nnn个词相关,即nnn阶马尔可夫链(Markov chain of order nnn),如果n=1n=1n=1,那么有P(w3∣w1,w2)=P(w3∣w2)P(w_3 \mid w_1, w_2) = P(w_3 \mid w_2)P(w3∣w1,w2)=P(w3∣w2)。基于n−1n-1n−1阶马尔可夫链,我们可以将语言模型改写为
P(w1,w2,…,wT)=∏t=1TP(wt∣wt−(n−1),…,wt−1). P(w_1, w_2, \ldots, w_T) = \prod_{t=1}^T P(w_t \mid w_{t-(n-1)}, \ldots, w_{t-1}) . P(w1,w2,…,wT)=t=1∏TP(wt∣wt−(n−1),…,wt−1).
以上也叫nnn元语法(nnn-grams),它是基于n−1n - 1n−1阶马尔可夫链的概率语言模型。例如,当n=2n=2n=2时,含有4个词的文本序列的概率就可以改写为:
P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w1,w2,w3)=P(w1)P(w2∣w1)P(w3∣w2)P(w4∣w3) P(w_1, w_2, w_3, w_4) = P(w_1) P(w_2 \mid w_1) P(w_3 \mid w_1, w_2) P(w_4 \mid w_1, w_2, w_3)\\ = P(w_1) P(w_2 \mid w_1) P(w_3 \mid w_2) P(w_4 \mid w_3) P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w1,w2,w3)=P(w1)P(w2∣w1)P(w3∣w2)P(w4∣w3)
当nnn分别为1、2和3时,我们将其分别称作一元语法(unigram)、二元语法(bigram)和三元语法(trigram)。例如,长度为4的序列w1,w2,w3,w4w_1, w_2, w_3, w_4w1,w2,w3,w4在一元语法、二元语法和三元语法中的概率分别为
P(w1,w2,w3,w4)=P(w1)P(w2)P(w3)P(w4),P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w2)P(w4∣w3),P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w2,w3). P(w_1, w_2, w_3, w_4) = P(w_1) P(w_2) P(w_3) P(w_4) ,\\ P(w_1, w_2, w_3, w_4) = P(w_1) P(w_2 \mid w_1) P(w_3 \mid w_2) P(w_4 \mid w_3) ,\\ P(w_1, w_2, w_3, w_4) = P(w_1) P(w_2 \mid w_1) P(w_3 \mid w_1, w_2) P(w_4 \mid w_2, w_3) . P(w1,w2,w3,w4)=P(w1)P(w2)P(w3)P(w4),P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w2)P(w4∣w3),P(w1,w2,w3,w4)=P(w1)P(w2∣w1)P(w3∣w1,w2)P(w4∣w2,w3).
当nnn较小时,nnn元语法往往并不准确。例如,在一元语法中,由三个词组成的句子“你走先”和“你先走”的概率是一样的。然而,当nnn较大时,nnn元语法需要计算并存储大量的词频和多词相邻频率。
时序数据的采样在训练中我们需要每次随机读取小批量样本和标签。与之前章节的实验数据不同的是,时序数据的一个样本通常包含连续的字符。假设时间步数为5,样本序列为5个字符,即“想”“要”“有”“直”“升”。该样本的标签序列为这些字符分别在训练集中的下一个字符,即“要”“有”“直”“升”“机”,即XXX=“想要有直升”,YYY=“要有直升机”。
现在我们考虑序列“想要有直升机,想要和你飞到宇宙去”,如果时间步数为5,有以下可能的样本和标签:
XXX:“想要有直升”,YYY:“要有直升机” XXX:“要有直升机”,YYY:“有直升机,” XXX:“有直升机,”,YYY:“直升机,想” … XXX:“要和你飞到”,YYY:“和你飞到宇” XXX:“和你飞到宇”,YYY:“你飞到宇宙” XXX:“你飞到宇宙”,YYY:“飞到宇宙去”可以看到,如果序列的长度为TTT,时间步数为nnn,那么一共有T−nT-nT−n个合法的样本,但是这些样本有大量的重合,我们通常采用更加高效的采样方式。我们有两种方式对时序数据进行采样,分别是随机采样和相邻采样。
随机采样与相邻采样demobatch_size
是每个小批量的样本数,num_steps
是每个样本所包含的时间步数。import torch
import random
def data_iter_random(corpus_indices, batch_size, num_steps, device=None):
# 减1是因为对于长度为n的序列,X最多只有包含其中的前n - 1个字符
num_examples = (len(corpus_indices) - 1) // num_steps # 下取整,得到不重叠情况下的样本个数
example_indices = [i * num_steps for i in range(num_examples)] # 每个样本的第一个字符在corpus_indices中的下标
random.shuffle(example_indices)
def _data(i):
# 返回从i开始的长为num_steps的序列
return corpus_indices[i: i + num_steps]
if device is None:
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
for i in range(0, num_examples, batch_size):
# 每次选出batch_size个随机样本
batch_indices = example_indices[i: i + batch_size] # 当前batch的各个样本的首字符的下标
X = [_data(j) for j in batch_indices]
Y = [_data(j + 1) for j in batch_indices]
yield torch.tensor(X, device=device), torch.tensor(Y, device=device)
def data_iter_consecutive(corpus_indices, batch_size, num_steps, device=None):
if device is None:
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
corpus_len = len(corpus_indices) // batch_size * batch_size # 保留下来的序列的长度
corpus_indices = corpus_indices[: corpus_len] # 仅保留前corpus_len个字符
indices = torch.tensor(corpus_indices, device=device)
indices = indices.view(batch_size, -1) # resize成(batch_size, )
batch_num = (indices.shape[1] - 1) // num_steps
for i in range(batch_num):
i = i * num_steps
X = indices[:, i: i + num_steps]
Y = indices[:, i + 1: i + num_steps + 1]
yield X, Y
注解:相邻采样是指把数据集线分为batchsize的大小的分数,每个batch取1个作为样本。
循环神经网络本节介绍循环神经网络,下图展示了如何基于循环神经网络实现语言模型。我们的目的是基于当前的输入与过去的输入序列,预测序列的下一个字符。循环神经网络引入一个隐藏变量HHH,用HtH_{t}Ht表示HHH在时间步ttt的值。HtH_{t}Ht的计算基于XtX_{t}Xt和Ht−1H_{t-1}Ht−1,可以认为HtH_{t}Ht记录了到当前字符为止的序列信息,利用HtH_{t}Ht对序列的下一个字符进行预测。
我们先尝试从零开始实现一个基于字符级循环神经网络的语言模型,这里我们使用周杰伦的歌词作为语料,首先我们读入数据:
import torch
import torch.nn as nn
import time
import math
import sys
sys.path.append("/home/kesci/input")
import d2l_jay9460 as d2l
(corpus_indices, char_to_idx, idx_to_char, vocab_size) = d2l.load_data_jay_lyrics()
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
def one_hot(x, n_class, dtype=torch.float32):
result = torch.zeros(x.shape[0], n_class, dtype=dtype, device=x.device) # shape: (n, n_class)
result.scatter_(1, x.long().view(-1, 1), 1) # result[i, x[i, 0]] = 1
return result
x = torch.tensor([0, 2])
x_one_hot = one_hot(x, vocab_size)
print(x_one_hot)
print(x_one_hot.shape)
print(x_one_hot.sum(axis=1))
def to_onehot(X, n_class):
return [one_hot(X[:, i], n_class) for i in range(X.shape[1])]
X = torch.arange(10).view(2, 5)
inputs = to_onehot(X, vocab_size)
print(len(inputs), inputs[0].shape)
num_inputs, num_hiddens, num_outputs = vocab_size, 256, vocab_size
# num_inputs: d
# num_hiddens: h, 隐藏单元的个数是超参数
# num_outputs: q
def get_params():
def _one(shape):
param = torch.zeros(shape, device=device, dtype=torch.float32)
nn.init.normal_(param, 0, 0.01)
return torch.nn.Parameter(param)
# 隐藏层参数
W_xh = _one((num_inputs, num_hiddens))
W_hh = _one((num_hiddens, num_hiddens))
b_h = torch.nn.Parameter(torch.zeros(num_hiddens, device=device))
# 输出层参数
W_hq = _one((num_hiddens, num_outputs))
b_q = torch.nn.Parameter(torch.zeros(num_outputs, device=device))
return (W_xh, W_hh, b_h, W_hq, b_q)
def rnn(inputs, state, params):
# inputs和outputs皆为num_steps个形状为(batch_size, vocab_size)的矩阵
W_xh, W_hh, b_h, W_hq, b_q = params
H, = state
outputs = []
for X in inputs:
H = torch.tanh(torch.matmul(X, W_xh) + torch.matmul(H, W_hh) + b_h)
Y = torch.matmul(H, W_hq) + b_q
outputs.append(Y)
return outputs, (H,)
def init_rnn_state(batch_size, num_hiddens, device):
return (torch.zeros((batch_size, num_hiddens), device=device), )
print(X.shape)
print(num_hiddens)
print(vocab_size)
state = init_rnn_state(X.shape[0], num_hiddens, device)
inputs = to_onehot(X.to(device), vocab_size)
params = get_params()
outputs, state_new = rnn(inputs, state, params)
print(len(inputs), inputs[0].shape)
print(len(outputs), outputs[0].shape)
print(len(state), state[0].shape)
print(len(state_new), state_new[0].shape)
def train_and_predict_rnn(rnn, get_params, init_rnn_state, num_hiddens,
vocab_size, device, corpus_indices, idx_to_char,
char_to_idx, is_random_iter, num_epochs, num_steps,
lr, clipping_theta, batch_size, pred_period,
pred_len, prefixes):
if is_random_iter:
data_iter_fn = d2l.data_iter_random
else:
data_iter_fn = d2l.data_iter_consecutive
params = get_params()
loss = nn.CrossEntropyLoss()
for epoch in range(num_epochs):
if not is_random_iter: # 如使用相邻采样,在epoch开始时初始化隐藏状态
state = init_rnn_state(batch_size, num_hiddens, device)
l_sum, n, start = 0.0, 0, time.time()
data_iter = data_iter_fn(corpus_indices, batch_size, num_steps, device)
for X, Y in data_iter:
if is_random_iter: # 如使用随机采样,在每个小批量更新前初始化隐藏状态
state = init_rnn_state(batch_size, num_hiddens, device)
else: # 否则需要使用detach函数从计算图分离隐藏状态
for s in state:
s.detach_()
# inputs是num_steps个形状为(batch_size, vocab_size)的矩阵
inputs = to_onehot(X, vocab_size)
# outputs有num_steps个形状为(batch_size, vocab_size)的矩阵
(outputs, state) = rnn(inputs, state, params)
# 拼接之后形状为(num_steps * batch_size, vocab_size)
outputs = torch.cat(outputs, dim=0)
# Y的形状是(batch_size, num_steps),转置后再变成形状为
# (num_steps * batch_size,)的向量,这样跟输出的行一一对应
y = torch.flatten(Y.T)
# 使用交叉熵损失计算平均分类误差
l = loss(outputs, y.long())
# 梯度清0
if params[0].grad is not None:
for param in params:
param.grad.data.zero_()
l.backward()
grad_clipping(params, clipping_theta, device) # 裁剪梯度
d2l.sgd(params, lr, 1) # 因为误差已经取过均值,梯度不用再做平均
l_sum += l.item() * y.shape[0]
n += y.shape[0]
if (epoch + 1) % pred_period == 0:
print('epoch %d, perplexity %f, time %.2f sec' % (
epoch + 1, math.exp(l_sum / n), time.time() - start))
for prefix in prefixes:
print(' -', predict_rnn(prefix, pred_len, rnn, params, init_rnn_state,
num_hiddens, vocab_size, device, idx_to_char, char_to_idx))
num_epochs, num_steps, batch_size, lr, clipping_theta = 250, 35, 32, 1e2, 1e-2
pred_period, pred_len, prefixes = 50, 50, ['分开', '不分开']
train_and_predict_rnn(rnn, get_params, init_rnn_state, num_hiddens,
vocab_size, device, corpus_indices, idx_to_char,
char_to_idx, True, num_epochs, num_steps, lr,
clipping_theta, batch_size, pred_period, pred_len,
prefixes)