ja-ginza

library5.2.0pypypiunverified

Japanese NLP pipeline for spaCy, providing tokenization, POS tagging, named entity recognition, dependency parsing, and word vectors. Based on a multi-task CNN trained on UD-Japanese BCCWJ and GSK2014-A. Installs as a spaCy model package for Japanese. Current version 5.2.0.

ai-ml

pip install ja-ginza

github ↗homepage ↗

INSTALL

IMPORT

SIG · JA-GINZA

ja-ginza

ai-mlpythonv5.2.0

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

ja_ginza

✓ import ja_ginza

✗ import spacy; nlp = spacy.load('ja_ginza')

Load the Japanese model and process a simple sentence.

import spacy nlp = spacy.load('ja_ginza') doc = nlp('Pythonは素晴らしいプログラミング言語です。') for token in doc: print(token.text, token.lemma_, token.pos_, token.dep_)

Debug

Known issues

breakingja-ginza 5.x requires spaCy v3.6+. Using it with older spaCy versions will cause import errors.

fix

Upgrade spaCy: pip install -U spacy>=3.6

affects: ja-ginza >=5.0.0, spaCy <3.6

gotchaThe model 'ja_ginza' is not automatically downloaded on pip install. You must download it separately using 'python -m spacy download ja_ginza' or use spacy.load() which triggers download.

fix

Run: python -m spacy download ja_ginza

affects: all

deprecatedja-ginza versions prior to 5.0 used a different model name ('ja_ginza_electra'). Upgrading to 5.0 changes the model to 'ja_ginza' and removes the Electra variant.

fix

Update code to use spacy.load('ja_ginza') instead of 'ja_ginza_electra'.

affects: ja-ginza <5.0

Upgrade

Version history

5.2.0latest on PyPI · released Mar 30, 2024

Audit

Dependencies

spacyrequiredja-ginza is a spaCy model package; spaCy must be installed

Agent activity

15 hits · last 30 days

node

Resources

githubhttps://github.com/megagonlabs/ginza ↗

homepagehttps://github.com/megagonlabs/ginza ↗