Bert Base Ja Cased

🚀 bert-base-ja-cased

We are sharing smaller versions of bert-base-multilingual-cased that handle a custom number of languages, while maintaining the same representations and accuracy as the original model.

🚀 Quick Start

We are sharing smaller versions of bert-base-multilingual-cased that can handle a specific number of languages. Different from distilbert-base-multilingual-cased, our versions generate exactly the same representations as the original model, thus preserving the original accuracy.

For more information, please refer to our paper: Load What You Need: Smaller Versions of Multilingual BERT.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-ja-cased")
model = AutoModel.from_pretrained("Geotrend/bert-base-ja-cased")

To generate other smaller versions of multilingual transformers, please visit our Github repo.

How to cite

@inproceedings{smallermbert,
  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
  booktitle={SustaiNLP / EMNLP},
  year={2020}
}

📚 Documentation

Please contact amine@geotrend.fr for any questions, feedback, or requests.

📄 License

This project is under the Apache 2.0 license.