manyoso / tibetan-part-of-speech-tool Goto Github PK

I pay homage to my Spiritual Heroes:
The Virtuous Teachers,
Practicing the Excellent Path,
Praised by the Wise.

This repository contains the parameter file and training corpus for Tibetan language support for the Tree Tagger tool from http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/

The lexicon, corpus, and enumeration of tags are in the training/ directory and come from http://larkpie.net/tibetancorpus/ courtesy of Dr. Nathan Hill. These documents needed to be edited to remove blank lines and add a end-of-sentence tag to the lexicon of '།'

The python script tibetan-pos.py acts as a wrapper/driver for the Tree Tagger and is invoked like so:

./tibetan-pos.py --file test/test.txt

The optional flags dictate the form of the output. With no flags all parts of speech operations are run and the output displayed to stdout.

Required arguments are:

script Specify the tibetan script directly on the command line OR -f --file Specify the file containing the tibetan script

Optional flags are:

-s --syllables Output tagged syllables only -w --words Output word break of input only -p --pos Output tagged part of speech only

May whatever merit
Achieved by this task
Accrue to the cause
For the liberation of all.