brightmart/roberta_zh
A Chinese RoBERTa pre-trained language model implementation in TensorFlow and PyTorch.
Not currently ranked — collecting fresh signals.
star history
This repository provides pre-trained RoBERTa models for Chinese language processing. It includes implementations for both TensorFlow and PyTorch frameworks. The models were trained on approximately 30GB of Chinese text data comprising nearly 300 million sentences and 10 billion Chinese tokens. Available model variants include 6-layer and 24/12-layer versions, compatible with standard Bert loading mechanisms.
Frequently asked
- What is brightmart/roberta_zh?
- A Chinese RoBERTa pre-trained language model implementation in TensorFlow and PyTorch.
- Is roberta_zh open source?
- Yes — brightmart/roberta_zh is an open-source project tracked on heatdrop.
- What language is roberta_zh written in?
- brightmart/roberta_zh is primarily written in Python.
- How popular is roberta_zh?
- brightmart/roberta_zh has 2.8k stars on GitHub.
- Where can I find roberta_zh?
- brightmart/roberta_zh is on GitHub at https://github.com/brightmart/roberta_zh.