README.md

metadata

language:
  - ja
library_name: spacy
tags:
  - spacy
  - ner
  - token-classification
  - japanese
pipeline_tag: token-classification
license: mit
model-index:
  - name: ja-ner-ja
    results:
      - task:
          type: token-classification
          name: NER
        metrics:
          - type: f1
            value: 0.922
            name: F1
          - type: precision
            value: 0.931
            name: Precision
          - type: recall
            value: 0.914
            name: Recall

ja-ner-ja

Japanese Named Entity Recognition model built with spaCy.

Labels

Label	Precision	Recall	F1
PERSON	0.950	0.949	0.949
DATE_OF_BIRTH	0.872	0.872	0.872
ADDRESS	0.904	0.967	0.934
ORGANIZATION	0.929	0.828	0.876
BANK_ACCOUNT	0.947	0.857	0.900

Architecture

• Pipeline: tok2vec → ner
• spaCy version: >=3.8.13,<3.9.0

Install

pip install https://huggingface.co/0xhikae/ja-ner-ja/resolve/main/ja_ner_ja-0.2.0-py3-none-any.whl

Usage

import spacy
nlp = spacy.load("ja_ner_ja")
doc = nlp("田中太郎は東京都渋谷区に住んでいます。")
for ent in doc.ents:
    print(ent.text, ent.label_)