Skip to main content

Tools for tokenizer develope and evaluation

Project description

Tokenizer Tools

https://img.shields.io/pypi/v/tokenizer_tools.svg https://travis-ci.com/howl-anderson/tokenizer_tools.svg?branch=master Documentation Status Updates

Tools/Utils for NLP (including dataset reading, tagset encoding & decoding, metrics computing) | NLP 工具集(包含数据集读取、tagset 编码和解码、指标的计算等)

Features

功能

语料集读写

本软件提供了一种语料存储的磁盘文件格式(暂定名为 conllx)和内存对象格式(暂定名为 offset)。

语料集读取

任务:读取 corpus.collx 文件,遍历打印每一条语料。

代码:

from tokenizer_tools.tagset.offset.corpus import Corpus

corpus = Corpus.read_from_file("corpus.conllx")
for document in corpus:
    print(document)  # document 就是单条语料对象
语料集写入

任务:将多条语料写入 corpus.conllx 文件

代码:

from tokenizer_tools.tagset.offset.corpus import Corpus

corpus_list = [corpus_item_one, corpus_item_two]

corpus = Corpus(corpus_list)
corpus.write_to_file("corpus.conllx")

Document 属性和方法

每一个单条语料都是一个 Document 对象,现在介绍这个对象所拥有的属性和方法

属性
text

类型是 list, 代表文本的字段

domain

类型是 string, 代表领域

function

类型是 string, 代表功能点

sub_function

类型是 string,代表子功能点

intent

类型是 string, 代表意图

entities

类型是 SpanSet, 代表实体,下文有详细介绍

方法
compare_entities

比较文本和实体是否匹配

convert_to_md

将文本和实体转换成 markdown 格式,用于文本化渲染输出

SpanSet 属性和方法

方法

__iter__

可以像列表一样访问,得到的每一个元素都是 Span 对象

check_overlap

检查 span 是否重叠

Span 属性和方法

属性

start

int, 从 0 开始,包含该位置

end

int, 从0开始,不包含该位置

entity

string, 实体类型

value

string, 实体的值

TODO

  • 改变项目的名字,tokenizer_tools 已经无法正确描述现在项目的功能

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.1.0 (2018-09-05)

  • First release on PyPI.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tokenizer_tools-0.46.1.tar.gz (69.8 kB view details)

Uploaded Source

Built Distribution

tokenizer_tools-0.46.1-py2.py3-none-any.whl (69.2 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file tokenizer_tools-0.46.1.tar.gz.

File metadata

  • Download URL: tokenizer_tools-0.46.1.tar.gz
  • Upload date:
  • Size: 69.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/50.3.0 requests-toolbelt/0.9.1 tqdm/4.49.0 CPython/3.7.4

File hashes

Hashes for tokenizer_tools-0.46.1.tar.gz
Algorithm Hash digest
SHA256 7bdfe8249b3422b77e2fe798d2ef38e05283c229834e79a12bdd8784d759733a
MD5 89877fda9a5612e2a6ed0cc689276361
BLAKE2b-256 a1ec56971a0b4adb6286713ab6dc85792d87dafd544e58367e678f6baaf49f2f

See more details on using hashes here.

File details

Details for the file tokenizer_tools-0.46.1-py2.py3-none-any.whl.

File metadata

  • Download URL: tokenizer_tools-0.46.1-py2.py3-none-any.whl
  • Upload date:
  • Size: 69.2 kB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/50.3.0 requests-toolbelt/0.9.1 tqdm/4.49.0 CPython/3.7.4

File hashes

Hashes for tokenizer_tools-0.46.1-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 e73d27f575e357202272ab68a9b3e6cbec32bc3aae8268d537ab090bdebf74bb
MD5 0a41f9fbf963caa4bd2c9af3e6986bb7
BLAKE2b-256 93851f5bd5df02cfc4a7d7edd7d4a9b62a29f092e64f6660fbe159337569196f

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page