Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/direct-phonology/spacy-och
the old chinese language for spaCy
https://github.com/direct-phonology/spacy-och
chinese nlp spacy
Last synced: 24 days ago
JSON representation
the old chinese language for spaCy
- Host: GitHub
- URL: https://github.com/direct-phonology/spacy-och
- Owner: direct-phonology
- License: mit
- Created: 2021-09-05T18:14:35.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2022-09-21T18:45:22.000Z (about 2 years ago)
- Last Synced: 2024-09-29T13:01:17.232Z (about 1 month ago)
- Topics: chinese, nlp, spacy
- Language: Python
- Homepage:
- Size: 827 KB
- Stars: 2
- Watchers: 0
- Forks: 0
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# spacy-och
[](https://github.com/direct-phonology/spacy-och/actions/workflows/ci.yml)
[](https://pypi.org/project/spacy-och/)
the Old Chinese (`och`) language for the [spaCy NLP library](https://spacy.io/).
## installation
requires spacy v3.
```sh
$ pip install spacy-och
```
## usage
this package currently doesn't include trained models and is intended for basic NLP usage only, via `nlp.blank()`. it tokenizes texts by character and supports the `Token.like_num` and `Token.is_stop` attributes.
```python
>>> import spacy
>>> nlp = spacy.blank("och")
>>> from spacy_och.examples import sentences
>>> doc = nlp(sentences[0])
>>> doc.text
子曰：「上下无常，非為邪也。進退无恆，非離群也。君子進德脩業、欲及時也，故无咎。」
>>> [t for t in doc if t.is_stop] # all stop words
[曰, ：, 非, 也, 。, 非, 也, 。, 、, 欲, 及, 也, 故, 。]
```
more functionality is coming soon!
## developing
after cloning the repository:
```sh
$ pip install -e ".[dev]"
$ pre-commit install
```
## building
build a source archive and distribution for a release:
```sh
$ rm -rf dist/*
$ python -m build
```
publish the release on [test PyPI](https://test.pypi.org/) (useful for making sure everything worked):
```sh
$ python -m twine upload --repository testpypi dist/*
```
if everything looks ok, upload to the real PyPI:
```sh
$ python -m twine upload dist/*
```
## license
code is licensed under the [MIT license](LICENSE). some lookups data is derived from files licensed under the [unicode data files and software license](https://www.unicode.org/license.html).