Unicode 字符串

在 tensorFlow.google.cn 上查看 在 Google Colab 中运行 在 GitHub 上查看源代码 下载 notebook

简介

处理自然语言的模型通常使用不同的字符集来处理不同的语言。Unicode 是一种标准的编码系统,用于表示几乎所有语言的字符。每个字符使用 00x10FFFF 之间的唯一整数码位进行编码。Unicode 字符串是由零个或更多码位组成的序列。

本教程介绍了如何在 TensorFlow 中表示 Unicode 字符串,以及如何使用标准字符串运算的 Unicode 等效项对其进行操作。它会根据字符体系检测将 Unicode 字符串划分为不同词例。

import tensorflow as tf
2023-11-07 23:56:44.044088: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2023-11-07 23:56:44.044147: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2023-11-07 23:56:44.045783: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered

tf.string 数据类型

您可以使用基本的 TensorFlow tf.string dtype 构建字节字符串张量。Unicode 字符串默认使用 UTF-8 编码。

tf.constant(u"Thanks 😊")
<tf.Tensor: shape=(), dtype=string, numpy=b'Thanks \xf0\x9f\x98\x8a'>

tf.string 张量可以容纳不同长度的字节字符串,因为字节字符串会被视为原子单元。字符串长度不包括在张量维度中。

tf.constant([u"You're", u"welcome!"]).shape
TensorShape([2])

注:使用 Python 构造字符串时,v2 和 v3 对 Unicode 的处理方式有所不同。在 v2 中,Unicode 字符串用前缀“u”表示(如上所示)。在 v3 中,字符串默认使用 Unicode 编码。

表示 Unicode

在 TensorFlow 中有两种表示 Unicode 字符串的标准方式:

  • string 标量 - 使用已知字符编码对码位序列进行编码。
  • int32 向量 - 每个位置包含单个码位。

例如,以下三个值均表示 Unicode 字符串 "语言处理"

# Unicode string, represented as a UTF-8 encoded string scalar.
text_utf8 = tf.constant(u"语言处理")
text_utf8
<tf.Tensor: shape=(), dtype=string, numpy=b'\xe8\xaf\xad\xe8\xa8\x80\xe5\xa4\x84\xe7\x90\x86'>
# Unicode string, represented as a UTF-16-BE encoded string scalar.
text_utf16be = tf.constant(u"语言处理".encode("UTF-16-BE"))
text_utf16be
<tf.Tensor: shape=(), dtype=string, numpy=b'\x8b\xed\x8a\x00Y\x04t\x06'>
# Unicode string, represented as a vector of Unicode code points.
text_chars = tf.constant([ord(char) for char in u"语言处理"])
text_chars
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([35821, 35328, 22788, 29702], dtype=int32)>

在不同表示之间进行转换

TensorFlow 提供了在下列不同表示之间进行转换的运算:

tf.strings.unicode_decode(text_utf8,
                          input_encoding='UTF-8')
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([35821, 35328, 22788, 29702], dtype=int32)>
tf.strings.unicode_encode(text_chars,
                          output_encoding='UTF-8')
<tf.Tensor: shape=(), dtype=string, numpy=b'\xe8\xaf\xad\xe8\xa8\x80\xe5\xa4\x84\xe7\x90\x86'>
tf.strings.unicode_transcode(text_utf8,
                             input_encoding='UTF8',
                             output_encoding='UTF-16-BE')
<tf.Tensor: shape=(), dtype=string, numpy=b'\x8b\xed\x8a\x00Y\x04t\x06'>

批次维度

解码多个字符串时,每个字符串中的字符数可能不相等。返回结果是 tf.RaggedTensor,其中最里面的维度的长度会根据每个字符串中的字符数而变化:

# A batch of Unicode strings, each represented as a UTF8-encoded string.
batch_utf8 = [s.encode('UTF-8') for s in
              [u'hÃllo',  u'What is the weather tomorrow',  u'Göödnight', u'😊']]
batch_chars_ragged = tf.strings.unicode_decode(batch_utf8,
                                               input_encoding='UTF-8')
for sentence_chars in batch_chars_ragged.to_list():
  print(sentence_chars)
[104, 195, 108, 108, 111]
[87, 104, 97, 116, 32, 105, 115, 32, 116, 104, 101, 32, 119, 101, 97, 116, 104, 101, 114, 32, 116, 111, 109, 111, 114, 114, 111, 119]
[71, 246, 246, 100, 110, 105, 103, 104, 116]
[128522]

您可以直接使用此 tf.RaggedTensor,也可以使用 tf.RaggedTensor.to_tensortf.RaggedTensor.to_sparse 方法将其转换为带有填充的密集 tf.Tensortf.SparseTensor

batch_chars_padded = batch_chars_ragged.to_tensor(default_value=-1)
print(batch_chars_padded.numpy())
[[   104    195    108    108    111     -1     -1     -1     -1     -1
      -1     -1     -1     -1     -1     -1     -1     -1     -1     -1
      -1     -1     -1     -1     -1     -1     -1     -1]
 [    87    104     97    116     32    105    115     32    116    104
     101     32    119    101     97    116    104    101    114     32
     116    111    109    111    114    114    111    119]
 [    71    246    246    100    110    105    103    104    116     -1
      -1     -1     -1     -1     -1     -1     -1     -1     -1     -1
      -1     -1     -1     -1     -1     -1     -1     -1]
 [128522     -1     -1     -1     -1     -1     -1     -1     -1     -1
      -1     -1     -1     -1     -1     -1     -1     -1     -1     -1
      -1     -1     -1     -1     -1     -1     -1     -1]]
batch_chars_sparse = batch_chars_ragged.to_sparse()

在对多个具有相同长度的字符串进行编码时,可以将 tf.Tensor 用作输入:

tf.strings.unicode_encode([[99, 97, 116], [100, 111, 103], [ 99, 111, 119]],
                          output_encoding='UTF-8')
<tf.Tensor: shape=(3,), dtype=string, numpy=array([b'cat', b'dog', b'cow'], dtype=object)>

当对多个具有不同长度的字符串进行编码时,应将 tf.RaggedTensor 用作输入:

tf.strings.unicode_encode(batch_chars_ragged, output_encoding='UTF-8')
<tf.Tensor: shape=(4,), dtype=string, numpy=
array([b'h\xc3\x83llo', b'What is the weather tomorrow',
       b'G\xc3\xb6\xc3\xb6dnight', b'\xf0\x9f\x98\x8a'], dtype=object)>

如果您的张量具有填充或稀疏格式的多个字符串,请在调用 unicode_encode 之前将其转换为 tf.RaggedTensor

tf.strings.unicode_encode(
    tf.RaggedTensor.from_sparse(batch_chars_sparse),
    output_encoding='UTF-8')
<tf.Tensor: shape=(4,), dtype=string, numpy=
array([b'h\xc3\x83llo', b'What is the weather tomorrow',
       b'G\xc3\xb6\xc3\xb6dnight', b'\xf0\x9f\x98\x8a'], dtype=object)>
tf.strings.unicode_encode(
    tf.RaggedTensor.from_tensor(batch_chars_padded, padding=-1),
    output_encoding='UTF-8')
<tf.Tensor: shape=(4,), dtype=string, numpy=
array([b'h\xc3\x83llo', b'What is the weather tomorrow',
       b'G\xc3\xb6\xc3\xb6dnight', b'\xf0\x9f\x98\x8a'], dtype=object)>

Unicode 运算

字符长度

tf.strings.length 运算具有 unit 参数,该参数表示计算长度的方式。unit 默认为 "BYTE",但也可以将其设置为其他值(例如 "UTF8_CHAR""UTF16_CHAR"),以确定每个已编码 string 中的 Unicode 码位数量。

# Note that the final character takes up 4 bytes in UTF8.
thanks = u'Thanks 😊'.encode('UTF-8')
num_bytes = tf.strings.length(thanks).numpy()
num_chars = tf.strings.length(thanks, unit='UTF8_CHAR').numpy()
print('{} bytes; {} UTF-8 characters'.format(num_bytes, num_chars))
11 bytes; 8 UTF-8 characters

字符子字符串

类似地,tf.strings.substr 运算会接受 "unit" 参数,并用它来确定 "pos" 和 "len" 参数包含的偏移类型。

# default: unit='BYTE'. With len=1, we return a single byte.
tf.strings.substr(thanks, pos=7, len=1).numpy()
b'\xf0'
# Specifying unit='UTF8_CHAR', we return a single character, which in this case
# is 4 bytes.
print(tf.strings.substr(thanks, pos=7, len=1, unit='UTF8_CHAR').numpy())
b'\xf0\x9f\x98\x8a'

拆分 Unicode 字符串

tf.strings.unicode_split 运算会将 Unicode 字符串拆分为单个字符的子字符串:

tf.strings.unicode_split(thanks, 'UTF-8').numpy()
array([b'T', b'h', b'a', b'n', b'k', b's', b' ', b'\xf0\x9f\x98\x8a'],
      dtype=object)

字符的字节偏移量

为了将 tf.strings.unicode_decode 生成的字符张量与原始字符串对齐,了解每个字符开始位置的偏移量很有用。方法 tf.strings.unicode_decode_with_offsetsunicode_decode 类似,不同的是它会返回包含每个字符起始偏移量的第二张量。

codepoints, offsets = tf.strings.unicode_decode_with_offsets(u"🎈🎉🎊", 'UTF-8')

for (codepoint, offset) in zip(codepoints.numpy(), offsets.numpy()):
  print("At byte offset {}: codepoint {}".format(offset, codepoint))
At byte offset 0: codepoint 127880
At byte offset 4: codepoint 127881
At byte offset 8: codepoint 127882

Unicode 字符体系

每个 Unicode 码位都属于某个码位集合,这些集合被称作字符体系。某个字符的字符体系有助于确定该字符可能所属的语言。例如,已知 'Б' 属于西里尔字符体系,表明包含该字符的现代文本很可能来自某个斯拉夫语种(如俄语或乌克兰语)。

TensorFlow 提供了 tf.strings.unicode_script 运算来确定某一给定码位使用的是哪个字符体系。字符体系代码是对应于国际 Unicode 组件 (ICU) UScriptCode 值的 int32 值。

uscript = tf.strings.unicode_script([33464, 1041])  # ['芸', 'Б']

print(uscript.numpy())  # [17, 8] == [USCRIPT_HAN, USCRIPT_CYRILLIC]
[17  8]

tf.strings.unicode_script 运算还可以应用于码位的多维 tf.Tensortf.RaggedTensor

print(tf.strings.unicode_script(batch_chars_ragged))
<tf.RaggedTensor [[25, 25, 25, 25, 25],
 [25, 25, 25, 25, 0, 25, 25, 0, 25, 25, 25, 0, 25, 25, 25, 25, 25, 25, 25,
  0, 25, 25, 25, 25, 25, 25, 25, 25]                                      ,
 [25, 25, 25, 25, 25, 25, 25, 25, 25], [0]]>

示例:简单分词

分词是将文本拆分为类似单词的单元的任务。当使用空格字符分隔单词时,这通常很容易,但是某些语言(如中文和日语)不使用空格,而某些语言(如德语)中存在长复合词,必须进行拆分才能分析其含义。在网页文本中,不同语言和字符体系常常混合在一起,例如“NY株価”(纽约证券交易所)。

我们可以利用字符体系的变化进行粗略分词(不实现任何 ML 模型),从而估算词边界。这对类似上面“NY株価”示例的字符串都有效。这种方法对大多数使用空格的语言也都有效,因为各种字符体系中的空格字符都归类为 USCRIPT_COMMON,这是一种特殊的字符体系代码,不同于任何实际文本。

# dtype: string; shape: [num_sentences]
#
# The sentences to process.  Edit this line to try out different inputs!
sentence_texts = [u'Hello, world.', u'世界こんにちは']

首先,我们将句子解码为字符码位,然后查找每个字符的字符体系标识符。

# dtype: int32; shape: [num_sentences, (num_chars_per_sentence)]
#
# sentence_char_codepoint[i, j] is the codepoint for the j'th character in
# the i'th sentence.
sentence_char_codepoint = tf.strings.unicode_decode(sentence_texts, 'UTF-8')
print(sentence_char_codepoint)

# dtype: int32; shape: [num_sentences, (num_chars_per_sentence)]
#
# sentence_char_scripts[i, j] is the unicode script of the j'th character in
# the i'th sentence.
sentence_char_script = tf.strings.unicode_script(sentence_char_codepoint)
print(sentence_char_script)
<tf.RaggedTensor [[72, 101, 108, 108, 111, 44, 32, 119, 111, 114, 108, 100, 46],
 [19990, 30028, 12371, 12435, 12395, 12385, 12399]]>
<tf.RaggedTensor [[25, 25, 25, 25, 25, 0, 0, 25, 25, 25, 25, 25, 0],
 [17, 17, 20, 20, 20, 20, 20]]>

接下来,我们使用这些字符体系标识符来确定添加词边界的位置。我们在每个句子的开头添加一个词边界;如果某个字符与前一个字符属于不同的字符体系,也为该字符添加词边界。

# dtype: bool; shape: [num_sentences, (num_chars_per_sentence)]
#
# sentence_char_starts_word[i, j] is True if the j'th character in the i'th
# sentence is the start of a word.
sentence_char_starts_word = tf.concat(
    [tf.fill([sentence_char_script.nrows(), 1], True),
     tf.not_equal(sentence_char_script[:, 1:], sentence_char_script[:, :-1])],
    axis=1)

# dtype: int64; shape: [num_words]
#
# word_starts[i] is the index of the character that starts the i'th word (in
# the flattened list of characters from all sentences).
word_starts = tf.squeeze(tf.where(sentence_char_starts_word.values), axis=1)
print(word_starts)
tf.Tensor([ 0  5  7 12 13 15], shape=(6,), dtype=int64)

然后,我们可以使用这些起始偏移量来构建 RaggedTensor,它包含了所有批次的单词列表:

# dtype: int32; shape: [num_words, (num_chars_per_word)]
#
# word_char_codepoint[i, j] is the codepoint for the j'th character in the
# i'th word.
word_char_codepoint = tf.RaggedTensor.from_row_starts(
    values=sentence_char_codepoint.values,
    row_starts=word_starts)
print(word_char_codepoint)
<tf.RaggedTensor [[72, 101, 108, 108, 111], [44, 32], [119, 111, 114, 108, 100], [46],
 [19990, 30028], [12371, 12435, 12395, 12385, 12399]]>

最后,我们可以将词码位 RaggedTensor 划分回句子中:

# dtype: int64; shape: [num_sentences]
#
# sentence_num_words[i] is the number of words in the i'th sentence.
sentence_num_words = tf.reduce_sum(
    tf.cast(sentence_char_starts_word, tf.int64),
    axis=1)

# dtype: int32; shape: [num_sentences, (num_words_per_sentence), (num_chars_per_word)]
#
# sentence_word_char_codepoint[i, j, k] is the codepoint for the k'th character
# in the j'th word in the i'th sentence.
sentence_word_char_codepoint = tf.RaggedTensor.from_row_lengths(
    values=word_char_codepoint,
    row_lengths=sentence_num_words)
print(sentence_word_char_codepoint)
<tf.RaggedTensor [[[72, 101, 108, 108, 111], [44, 32], [119, 111, 114, 108, 100], [46]],
 [[19990, 30028], [12371, 12435, 12395, 12385, 12399]]]>

为了使最终结果更易于阅读,我们可以将其重新编码为 UTF-8 字符串:

tf.strings.unicode_encode(sentence_word_char_codepoint, 'UTF-8').to_list()
[[b'Hello', b', ', b'world', b'.'],
 [b'\xe4\xb8\x96\xe7\x95\x8c',
  b'\xe3\x81\x93\xe3\x82\x93\xe3\x81\xab\xe3\x81\xa1\xe3\x81\xaf']]