文本预处理
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
预处理通常包括四个步骤:
- 读入文本
英文小说:以英文小说为例展示文本预处理步骤
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] #每次处理文件的一行。strip()函数去掉前缀后缀的空白字符,lower()函数将大写转换为小写,re.sub是正则表达式的替换函数(将非英文字符的子串替换为空格)
return lines #lines是一个列表
lines = read_time_machine()
print('# sentences %d' % len(lines))
- 分词
def tokenize(sentences, token='word'): #参数sentences是一个列表,每个列表元素是一个字符串句子
#参数token是一个标志,用来表示我们要做哪一个级别的分词
"""Split sentences into word or char tokens"""
if token == 'word': #token='word'表示要做单词级别的分词
return [sentence.split(' ') for sentence in sentences]
#对sentence中的每个句子以空格作为分隔符来做分隔
elif token == 'char': #token='char'表示要做字符级别的分词
return [list(sentence) for sentence in sentences]
#对sentence中的每个句子直接将字符串转换为一个列表
else:
print('ERROR: unkown token type '+token)
tokens = tokenize(lines)
tokens[0:2]
#tokens函数返回的是一个二维列表,其中第一个维度是sentences当中的每个句子,第二个维度是每个句子分词之后得到的一个单词或者字符的序列
- 建立字典,将每个词映射到一个唯一的索引(index)
class Vocab(object): #Vocab(object)类希望把整个语料库上的每个词都唯一的映射到一个索引编号,当我们向Vocab查一个词的时候能够得到所查词的索引编号,查询索引可以得到索引对应的词
def __init__(self, tokens, min_freq=0, use_special_tokens=False): #构造函数
'''
参数tokens——函数tokenize()的返回结果,是一个二维列表,实际上是语料库中所有的词
参数min_freq——阈值,当构造字典时语料库中某些词出现的次数非常少,当出现次数小于这个阈值的词则忽略不计
参数use_special_tokens——一个标志,用来表示构建这个字典的时候是否使用一些特殊的token
'''
counter = count_corpus(tokens) #统计词频<key, value>:<词, 词频>
self.token_freqs = list(counter.items())
#此时将语料库的词进行了去重,且统计出了词频
self.idx_to_token = [] #self.idx_to_token列表用来记录语料库中最终需要维护的token
if use_special_tokens:
# padding, begin of sentence, end of sentence, unknown
'''
pad——padding,在短的句子中补上一些特殊token pad,使得其与其他句子长度一致
bos——begin of sentence,句子的开始处添加,用于表示句子的开始
eos——end of sentence,句子的结尾处添加,用于表示句子的结尾
unk——unknown,未登录词,从未在语料库中出现的词
'''
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 #处理完特殊token后,将语料库中的词添加到idx_to_token中
if freq >= min_freq and token not in self.idx_to_token] #如果词频>=阈值min_freq,且这个词未在idx_to_token中出现过,则将token添加到idx_to_token中
#idx_to_token是一个列表,所以天然就是token到索引的映射,因为每个词的下标可作为索引
self.token_to_idx = dict() #词到索引号的映射,定义为一个字典
for idx, token in enumerate(self.idx_to_token): #枚举idx_to_token中每个词和词下标
self.token_to_idx[token] = idx #将词和词下标添加到idx_to_token中
def __len__(self): #返回词典大小
return len(self.idx_to_token)
def __getitem__(self, tokens): #定义了Vocab类的索引,参数token可以是1.列表或元组,2.字符串。 从词到索引的映射
if not isinstance(tokens, (list, tuple)):
return self.token_to_idx.get(tokens, self.unk) #找到token则返回token,否则返回self.unk
return [self.__getitem__(token) for token in tokens]
def to_tokens(self, indices): #从索引到词的映射
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): #统计词频的函数,sentences是前面的tokens,是一个二维列表
tokens = [tk for st in sentences for tk in st] #将sentences展平得到一个一维列表
return collections.Counter(tokens) # 返回一个字典,记录每个词的出现次数
#例子:用Time Machine作为预料构建字典
vocab = Vocab(tokens)
print(list(vocab.token_to_idx.items())[0:10])
- 将文本从词的序列转换为索引的序列,方便输入模型
#将词转换为索引
for i in range(8, 10):
print('words:', tokens[i]) #tokens[i]表示的是第i行分词之后的单词序列
print('indices:', vocab[tokens[i]]) #用tokens[i]对vocab进行索引得到各个词的索引编号
#用现有工具进行分词的例子
text = "Mr. Chen doesn't agree with my suggestion."
#spaCy
import spacy
nlp = spacy.load('en_core_web_sm')
doc = nlp(text)
print([token.text for token in doc])
#NLTK
from nltk.tokenize import word_tokenize
from nltk import data
data.path.append('/home/kesci/input/nltk_data3784/nltk_data')
print(word_tokenize(text))
学习笔记
建立字典,设置阈值
去重筛选词,特殊需求token
1、count_corpus统计词频,得到counter
2、增删,利用空列表
pad:二维矩阵长度不一,短句子补token利用pad
bos:开始token
eos:结束token
unk:未登录词当作unk
3、词到索引号
建立词典:
词典的主要作用是将每一个词映射到一个唯一的索引号,主要构建了一个idx_to_token列表来存储所有的词,一个token_to_idx来存储所有词的索引。
在实现的的流程上是:
- 对语料进行分词,生成一个token列表,里面包含了语料的分词结果
- 对分好的词统计词频,然后根据词频来构建词典(统计好的词频完成了去重的操作,同时也保留了词的频率,方便后续的操作)
其中有一些名词的作用是视频里提出来的
1.pad的作用是在采用批量样本训练时,对于长度不同的样本(句子),对于短的样本采用pad进行填充,使得每个样本的长度是一致的 - bos( begin of sentence)和eos(end of sentence)是用来表示一句话的开始和结尾
3.unk(unknow)的作用是,处理遇到从未出现在预料库的词时都统一认为是unknow ,在代码中还可以将一些频率特别低的词也归为这一类
语言模型与数据集
语言模型数据集
#读取数据集
with open('/home/kesci/input/jaychou_lyrics4703/jaychou_lyrics.txt') as f:
corpus_chars = f.read()
print(len(corpus_chars))
print(corpus_chars[: 40])
corpus_chars = corpus_chars.replace('\n', ' ').replace('\r', ' ')
corpus_chars = corpus_chars[: 10000]
#建立字符索引
idx_to_char = list(set(corpus_chars)) # 去重,得到索引到字符的映射
char_to_idx = {char: i for i, char in enumerate(idx_to_char)} # 字符到索引的映射
vocab_size = len(char_to_idx)
print(vocab_size)
corpus_indices = [char_to_idx[char] for char in corpus_chars] # 将每个字符转化为索引,得到一个索引的序列
sample = corpus_indices[: 20]
print('chars:', ''.join([idx_to_char[idx] for idx in sample]))
print('indices:', sample)
def load_data_jay_lyrics():
with open('/home/kesci/input/jaychou_lyrics4703/jaychou_lyrics.txt') as f:
corpus_chars = f.read()
corpus_chars = corpus_chars.replace('\n', ' ').replace('\r', ' ')
corpus_chars = corpus_chars[0:10000]
idx_to_char = list(set(corpus_chars))
char_to_idx = dict([(char, i) for i, char in enumerate(idx_to_char)])
vocab_size = len(char_to_idx)
corpus_indices = [char_to_idx[char] for char in corpus_chars]
return corpus_indices, char_to_idx, idx_to_char, vocab_size
时序数据的采样:1.随机采样,2.相邻采样
#随机采样
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)
#测试随机采样
my_seq = list(range(30))
for X, Y in data_iter_random(my_seq, batch_size=2, num_steps=6):
print('X: ', X, '\nY:', Y, '\n')
#相邻采样
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
#测试相邻采样
for X, Y in data_iter_consecutive(my_seq, batch_size=2, num_steps=6):
print('X: ', X, '\nY:', Y, '\n')
循环神经网络从零开始实现
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')
#one-hot向量
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)
#初始化模型参数
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)
#定义模型
#函数rnn用循环的方式依次完成循环神经网络每个时间步的计算。
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,)
#函数init_rnn_state初始化隐藏变量,这里的返回值是一个元组。
def init_rnn_state(batch_size, num_hiddens, device):
return (torch.zeros((batch_size, num_hiddens), device=device), )
#简单的测试
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)
#剪裁梯度
def grad_clipping(params, theta, device):
norm = torch.tensor([0.0], device=device)
for param in params:
norm += (param.grad.data ** 2).sum()
norm = norm.sqrt().item()
if norm > theta:
for param in params:
param.grad.data *= (theta / norm)
#定义预测函数
def predict_rnn(prefix, num_chars, rnn, params, init_rnn_state,
num_hiddens, vocab_size, device, idx_to_char, char_to_idx):
state = init_rnn_state(1, num_hiddens, device)
output = [char_to_idx[prefix[0]]] # output记录prefix加上预测的num_chars个字符
for t in range(num_chars + len(prefix) - 1):
# 将上一时间步的输出作为当前时间步的输入
X = to_onehot(torch.tensor([[output[-1]]], device=device), vocab_size)
# 计算输出和更新隐藏状态
(Y, state) = rnn(X, state, params)
# 下一个时间步的输入是prefix里的字符或者当前的最佳预测字符
if t < len(prefix) - 1:
output.append(char_to_idx[prefix[t + 1]])
else:
output.append(Y[0].argmax(dim=1).item())
return ''.join([idx_to_char[i] for i in output])
#测试预测函数
predict_rnn('分开', 10, rnn, params, init_rnn_state, num_hiddens, vocab_size,
device, idx_to_char, char_to_idx)
#定义模型训练函数
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))
#训练模型并创作歌词
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']
#采用随机采样训练模型并创作歌词。
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)
#采用相邻采样训练模型并创作歌词。
train_and_predict_rnn(rnn, get_params, init_rnn_state, num_hiddens,
vocab_size, device, corpus_indices, idx_to_char,
char_to_idx, False, num_epochs, num_steps, lr,
clipping_theta, batch_size, pred_period, pred_len,
prefixes)
循环神经网络的简洁实现
#定义模型
class RNNModel(nn.Module):
def __init__(self, rnn_layer, vocab_size):
super(RNNModel, self).__init__()
self.rnn = rnn_layer
self.hidden_size = rnn_layer.hidden_size * (2 if rnn_layer.bidirectional else 1)
self.vocab_size = vocab_size
self.dense = nn.Linear(self.hidden_size, vocab_size)
def forward(self, inputs, state):
# inputs.shape: (batch_size, num_steps)
X = to_onehot(inputs, vocab_size)
X = torch.stack(X) # X.shape: (num_steps, batch_size, vocab_size)
hiddens, state = self.rnn(X, state)
hiddens = hiddens.view(-1, hiddens.shape[-1]) # hiddens.shape: (num_steps * batch_size, hidden_size)
output = self.dense(hiddens)
return output, state
#定义预测函数
def predict_rnn_pytorch(prefix, num_chars, model, vocab_size, device, idx_to_char,
char_to_idx):
state = None
output = [char_to_idx[prefix[0]]] # output记录prefix加上预测的num_chars个字符
for t in range(num_chars + len(prefix) - 1):
X = torch.tensor([output[-1]], device=device).view(1, 1)
(Y, state) = model(X, state) # 前向计算不需要传入模型参数
if t < len(prefix) - 1:
output.append(char_to_idx[prefix[t + 1]])
else:
output.append(Y.argmax(dim=1).item())
return ''.join([idx_to_char[i] for i in output])
#使用权重为随机值的模型来预测一次。
model = RNNModel(rnn_layer, vocab_size).to(device)
predict_rnn_pytorch('分开', 10, model, vocab_size, device, idx_to_char, char_to_idx)
#使用相邻采样实现训练函数
def train_and_predict_rnn_pytorch(model, num_hiddens, vocab_size, device,
corpus_indices, idx_to_char, char_to_idx,
num_epochs, num_steps, lr, clipping_theta,
batch_size, pred_period, pred_len, prefixes):
loss = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=lr)
model.to(device)
for epoch in range(num_epochs):
l_sum, n, start = 0.0, 0, time.time()
data_iter = d2l.data_iter_consecutive(corpus_indices, batch_size, num_steps, device) # 相邻采样
state = None
for X, Y in data_iter:
if state is not None:
# 使用detach函数从计算图分离隐藏状态
if isinstance (state, tuple): # LSTM, state:(h, c)
state[0].detach_()
state[1].detach_()
else:
state.detach_()
(output, state) = model(X, state) # output.shape: (num_steps * batch_size, vocab_size)
y = torch.flatten(Y.T)
l = loss(output, y.long())
optimizer.zero_grad()
l.backward()
grad_clipping(model.parameters(), clipping_theta, device)
optimizer.step()
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_pytorch(
prefix, pred_len, model, vocab_size, device, idx_to_char,
char_to_idx))
#训练模型
num_epochs, batch_size, lr, clipping_theta = 250, 32, 1e-3, 1e-2
pred_period, pred_len, prefixes = 50, 50, ['分开', '不分开']
train_and_predict_rnn_pytorch(model, num_hiddens, vocab_size, device,
corpus_indices, idx_to_char, char_to_idx,
num_epochs, num_steps, lr, clipping_theta,
batch_size, pred_period, pred_len, prefixes)