This is an ELECTRA model pretrained on approximately 200M Japanese sentences extracted from the mC4 and finetuned by spaCy v3 on UD_Japanese_BCCWJ r2.8.

The base pretrain model is megagonlabs/transformers-ud-japanese-electra-base-discrimininator.

The entire spaCy v3 model is distributed as a python package named ja_ginza_electra from PyPI along with GiNZA v5 which provides some custom pipeline components to recognize the Japanese bunsetu-phrase structures. Try running it as below:

console

$ pip install ginza ja_ginza_electra
$ ginza

Licenses

The models are distributed under the terms of the MIT License.

Acknowledgments

This model is permitted to be published under the MIT License under a joint research agreement between NINJAL (National Institute for Japanese Language and Linguistics) and Megagon Labs Tokyo.

Citations

Contains information from mC4 which is made available under the ODC Attribution License.

@article{2019t5,
    author = {Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu},
    title = {Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer},
    journal = {arXiv e-prints},
    year = {2019},
    archivePrefix = {arXiv},
    eprint = {1910.10683},
}

UD_Japanese_BCCWJ r2.8

Asahara, M., Kanayama, H., Tanaka, T., Miyao, Y., Uematsu, S., Mori, S.,
Matsumoto, Y., Omura, M., & Murawaki, Y. (2018).
Universal Dependencies Version 2 for Japanese.
In LREC-2018.

GSK2014-A(2019)

megagonlabs/transformers-ud-japanese-electra-base-ginza-520

transformers-ud-japanese-electra-ginza-520 (sudachitra-wordpiece, mC4 Japanese)

Licenses

Acknowledgments

Citations

megagonlabs/
transformers-ud-japanese-electra-base-ginza-520