31. 文本预处理

文本预处理

• 将文本作为字符串加载到内存中
• 将字符串拆分为词元（如单词和字符）
• 建立一个词汇表，将拆分的词元映射到数字索引
• 将文本转换为数字索引序列，方便模型操作

1.读取数据集

import collections
import re
from d2l import torch as d2l

d2l.DATA_HUB['time_machine'] = (d2l.DATA_URL + 'timemachine.txt',
                                '090b5e7e70c295757f55df93cb0a180b9691891a')

def read_time_machine():
    """将时间机器数据集加载到文本行的列表中"""
    with open(d2l.download('time_machine'), 'r') as f:
        lines = f.readlines()
    # 把以“非字母”的符号替换为“ ”,删除首尾空格，将所有大写字母改为小写
    return [re.sub('[^A-Za-z]+',' ',line).strip().lower() for line in lines]
lines = read_time_machine()
print(f'# 文本总行数: {len(lines)}')
print(lines[0])
print(lines[10])

# 文本总行数: 3221
the time machine by h g wells
twinkled and his usually pale face was flushed and animated the

2.词元化

#文本序列又被拆分成一个词元列表，词元（token）是文本的基本单位, 最后，返回一个由词元列表组成的列表
def tokenize(lines,token='word'):
    if token == "word":
        #以空格作为间隔符分词
        return [line.split() for line in lines]
    elif token == 'char':
        return [list(line) for line in lines]
    else:
        print('错误：未知词元类型：' + token)
tokens = tokenize(lines)
for i in range(11):
    print(tokens[i])

['the', 'time', 'machine', 'by', 'h', 'g', 'wells']
[]
[]
[]
[]
['i']
[]
[]
['the', 'time', 'traveller', 'for', 'so', 'it', 'will', 'be', 'convenient', 'to', 'speak', 'of', 'him']
['was', 'expounding', 'a', 'recondite', 'matter', 'to', 'us', 'his', 'grey', 'eyes', 'shone', 'and']
['twinkled', 'and', 'his', 'usually', 'pale', 'face', 'was', 'flushed', 'and', 'animated', 'the']

3.词表化

class Vocab:
    """文本词表"""
    def __init__(self,tokens=None,min_frep=0,reserved_tokens=None):
        if tokens is None:
            tokens = []
        if reserved_tokens is None:
            reserved_tokens = []
        # 按出现频率排序
        counter = count_corpus(tokens)
        self._token_freqs = sorted(counter.items(),key = lambda x:x[1],reverse=True)
        # 未知词元的索引为0
        self.idx_to_token = ['<unk>'] + reserved_tokens#<unk>未知词元
        self.token_to_idx = {token: idx
                             for idx, token in enumerate(self.idx_to_token)}
        for token, freq in self._token_freqs:
            if freq < min_frep:
                break
            if token not in self.token_to_idx:
                self.idx_to_token.append(token)
                self.token_to_idx[token] = len(self.idx_to_token) - 1
    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]
    @property#禁止用户去改数据
    def unk(self):  # 未知词元的索引为0
        return 0

    @property
    def token_freqs(self):
        return self._token_freqs
def count_corpus(tokens):
    """统计词元的频率"""
    # 这里的tokens是1D列表或2D列表
    # 判断 tokens是否有数据，tokens【0】是否为列表形式
    if len(tokens) == 0 or isinstance(tokens[0], list):
        # 将词元列表展平成一个列表
        # line从tokens中取出数据
        tokens = [token for line in tokens for token in line]
         #可以支持方便、快速的计数，将元素数量统计，然后计数并返回一个字典，键为元素，值为元素个数。
        return collections.Counter(tokens)#计数器 

#tokens = [token for line in tokens for token in line]
flag=5

for line in tokens:
    print(line)
    flag-=1
    if flag < 0:
        break

['the', 'time', 'machine', 'by', 'h', 'g', 'wells']
[]
[]
[]
[]
['i']

#tokens = [token for line in tokens for token in line]
flag = 5
for line in tokens:
    for token in line:
        print(token)
    flag-=1
    if flag < 0:
        break

the
time
machine
by
h
g
wells
i

#打印前几个高频词元及其索引。
vocab = Vocab(tokens)
print(list(vocab.token_to_idx.items())[:10])

[('<unk>', 0), ('the', 1), ('i', 2), ('and', 3), ('of', 4), ('a', 5), ('to', 6), ('was', 7), ('in', 8), ('that', 9)]

vocab._token_freqs

[('the', 2261),
 ('i', 1267),
 ('and', 1245),
 ('of', 1155),
 ('a', 816),
 ('to', 695),
 ('was', 552),
 ('in', 541),
 ('that', 443),
 ('my', 440),
 ('it', 437),
 ('had', 354),
 ('me', 281),
 ('as', 270),
 ('at', 243),
 ('for', 221),
 ('with', 216),
 ('but', 204),
 ('time', 200),
 ('were', 158),
 ('this', 152),
 ('you', 137),

 ...]

vocab.idx_to_token

['<unk>',
 'the',
 'i',
 'and',
 'of',
 'a',
 'to',
 'was',
 'in',
 'that',
 'my',
 'it',
 'had',
 'me',
 'as',
 'at',
 'for',
 'with',
 'but',
 'time',
 'were',
 'this',
 'you',
 'on',
 'then',
 'his',
 'there',
 'he',
 'have',
 'they',
 'from',
 'one',
 'all',
 'not',
 'into',
 'upon',
 'little',
 'so',
 'is',
 'came',
 'by',
 'some',
 'be',
 'no',
 'could',
 'their',
 ...]

vocab.token_to_idx

{'<unk>': 0,
 'the': 1,
 'i': 2,
 'and': 3,
 'of': 4,
 'a': 5,
 'to': 6,
 'was': 7,
 'in': 8,
 'that': 9,
 'my': 10,
 'it': 11,
 'had': 12,
 'me': 13,
 'as': 14,
 'at': 15,
 'for': 16,
 'with': 17,
 'but': 18,
 'time': 19,
 'were': 20,
 'this': 21,
 ...}

for i in [0, 10]:
    print('文本:', tokens[i])
    print('索引:', vocab[tokens[i]])

文本: ['the', 'time', 'machine', 'by', 'h', 'g', 'wells']
索引: [1, 19, 50, 40, 2183, 2184, 400]
文本: ['twinkled', 'and', 'his', 'usually', 'pale', 'face', 'was', 'flushed', 'and', 'animated', 'the']
索引: [2186, 3, 25, 1044, 362, 113, 7, 1421, 3, 1045, 1]

4.整合所有功能

def load_corpus_time_machine(max_tokens=-1):  #@save
    """返回时光机器数据集的词元索引列表和词表"""
    lines = read_time_machine()
    tokens = tokenize(lines, 'char')
    vocab = Vocab(tokens)
    # 因为时光机器数据集中的每个文本行不一定是一个句子或一个段落，
    # 所以将所有文本行展平到一个列表中
    corpus = [vocab[token] for line in tokens for token in line]
    if max_tokens > 0:
        corpus = corpus[:max_tokens]
    return corpus, vocab

corpus, vocab = load_corpus_time_machine()
len(corpus), len(vocab)

(170580, 28)

5.小节

• 文本是序列数据的一种最常见的形式之一。

• 为了对文本进行预处理，我们通常将文本拆分为词元，构建词表将词元字符串映射为数字索引，并将文本数据转换为词元索引以供模型操作。