KyTea wrapper for Python

Mykytea-python is a Python wrapper module for KyTea, a general text analysis toolkit. KyTea is developed by KyTea Development Team.

Detailed information on KyTea can be found at: http://www.phontron.com/kytea

Installation

Install Mykytea-python via pip

You can install Mykytea-python via pip.

pip install kytea

You don't have to install KyTea anymore before installing Mykytea-python when you install it by using wheel on PyPI.

You should have any KyTea model on your machine.

Build Mykytea-python from source

Mykytea-python supports Python 3.11 or later.

To build from source, install KyTea first, then run

pip install .

SWIG is installed automatically as a build dependency from PyPI, and the wrapper is generated during the build. You don't need to run swig yourself.

If KyTea is installed outside /usr/local, pass its prefix with KYTEA_DIR. For example, with Homebrew on macOS:

brew install kytea
KYTEA_DIR=$(brew --prefix) pip install .

For development, make installs the package in editable mode and make test runs the tests.

How to use it?

Here is the example code to use Mykytea-python.

import Mykytea

def showTags(t):
    for word in t:
        out = word.surface + "\t"
        for t1 in word.tag:
            for t2 in t1:
                for t3 in t2:
                    out = out + "/" + str(t3)
                out += "\t"
            out += "\t"
        print(out)

def list_tags(t):
    def convert(t2):
        return (t2[0], type(t2[1]))
    return [(word.surface, [[convert(t2) for t2 in t1] for t1 in word.tag]) for word in t]

# Pass options for KyTea as keyword arguments:
mk = Mykytea.Mykytea(model="/usr/local/share/kytea/model.bin")
# or as a KyTea command line string:
# mk = Mykytea.Mykytea("-model /usr/local/share/kytea/model.bin")

s = "今日はいい天気です。"

# Fetch segmented words
for word in mk.getWS(s):
    print(word)

# Show analysis results
print(mk.getTagsToString(s))

# Fetch first best tag
t = mk.getTags(s)
showTags(t)

# Show all tags
tt = mk.getAllTags(s)
showTags(tt)

Options

Mykytea.Mykytea() accepts these KyTea analysis options as keyword arguments. Keyword arguments are applied after the option string, if both are given.

Keyword	KyTea option	Description
model	-model	Path to the model file
nows	-nows	Don't do word segmentation; the input must be segmented with spaces (True/False)
notags	-notags	Do only word segmentation, no tagging (True/False)
notag	-notag	Skip the n-th tag, starting at 1 (an int or a list of ints). What each tag is depends on the model; in the Japanese model, 1 is the part of speech and 2 is the pronunciation. A skipped tag is an empty list in getTags() and omitted in getTagsToString()
nounk	-nounk	Don't estimate the pronunciation of unknown words (True/False)
wsconst	-wsconst	Character types not to be segmented (e.g. "D" for digits)
numtags	-numtags	The number of tags to estimate
tagmax	-tagmax	The maximum number of tag candidates
unktag	-unktag	A tag appended to unknown words
deftag	-deftag	A tag for words that cannot be given any tag
unkbeam	-unkbeam	Beam width for unknown word search (0 for full search)
debug	-debug	Debugging level (0=silent, 1=simple, 2=detailed)

An unknown keyword argument raises TypeError.

License

MIT License