pybo 0.6.2

Python utils for processing Tibetan

PYBO - Tibetan NLP in Python

Overview

pybo is a word tokenizer for the Tibetan language written in Python. pybo takes in chunks of text and returns lists of words. It provides an easy-to-use, high-performance tokenization pipeline that can serve as a stand-alone solution or be adapted as a complement.

Getting started

pip install pybo

Or to install from the latest master branch:

pip install git+https://github.com/Esukhia/pybo.git

How to use pybo

To initiate the tokenizer together with part-of-speech capability:

# Initialize the tokenizer
tok = bo.WordTokenizer('POS')

# Feed it some Tibetan text
input_str = '༄༅། །རྒྱ་གར་སྐད་དུ། བོ་དྷི་སཏྭ་ཙརྻ་ཨ་བ་ཏ་ར། བོད་སྐད་དུ། བྱང་ཆུབ་སེམས་དཔའི་སྤྱོད་པ་ལ་འཇུག་པ། །སངས་རྒྱས་དང་བྱང་ཆུབ་སེམས་དཔའ་ཐམས་ཅད་ལ་ཕྱག་འཚལ་ལོ། །བདེ་གཤེགས་ཆོས་ཀྱི་སྐུ་མངའ་སྲས་བཅས་དང༌། །ཕྱག་འོས་ཀུན་ལའང་གུས་པར་ཕྱག་འཚལ་ཏེ། །བདེ་གཤེགས་སྲས་ཀྱི་སྡོམ་ལ་འཇུག་པ་ནི། །ལུང་བཞིན་མདོར་བསྡུས་ནས་ནི་བརྗོད་པར་བྱ། །'

# Run the tokenizer
tokens = tok.tokenize(input_str)

Now in 'tokens' you have an iterable where each token consist of several meta-data:

# Access the first token in the iterable
tokens[0]

This yields:

text: "༄༅། "
char_types: |PUNCT|PUNCT|PUNCT|SPACE|
chunk_type: PUNCT
start: 0
len: 4
syls: None
pos: PUNCT
skrt: False
freq: 0

notes:

• start is the starting index of the current token in the input string.
• syls is a list of cleaned syllables, each syllable being represented as a list of indices. Each index leads to a constituting character within the input string.

How to access all the words in a list

# iterate through the tokens object to get all the words in a list
[t.content for t in tokens]

How to get all the nouns in a text

# extract nouns from the tokens
[t.content for t in tokens if t.tag == 'NOUNᛃᛃᛃ']

These examples highlight the basic principle of accessing attributes within each token object.

Acknowledgements

pybo is an open source library for Tibetan NLP.

We are always open to cooperation in introducing new features, tool integrations and testing solutions.

Many thanks to the companies and organizations who have supported pybo's development, especially:

• Khyentse Foundation for contributing USD22,000 to kickstart the project
• The Barom/Esukhia canon project for sponsoring training data curation
• BDRC for contributing 2 staff for 6 months for data curation

Maintainance

Build the source dist:

rm -rf dist/
python3 setup.py clean sdist

and upload on twine (version >= 1.11.0) with:

twine upload dist/*

License

The Python code is Copyright (C) 2019 Esukhia, provided under Apache 2.

author: Drupchen

contributors:

• Élie Roux
• Thubten Rinzin
• Ngawang Trinley
• Mikko Kotila
• Tenzin
• Joyce Mackzenzie for reworking the logo