ai-forever/ner-bert
BERT-based Named Entity Recognition system using BERTBiLSTMAttnCRF architecture on PyTorch.
Not currently ranked — collecting fresh signals.
star history
This repository implements Named Entity Recognition using BERT as a feature extractor combined with bidirectional LSTM and attention mechanisms, topped with a CRF layer for sequence labeling. It loads pre-trained TensorFlow BERT checkpoints and supports joint models for multilingual NER tasks. The system provides data loading, model training, and prediction pipelines tailored for NER applications.
Frequently asked
- What is ai-forever/ner-bert?
- BERT-based Named Entity Recognition system using BERTBiLSTMAttnCRF architecture on PyTorch.
- Is ner-bert open source?
- Yes — ai-forever/ner-bert is open source, released under the MIT license.
- What language is ner-bert written in?
- ai-forever/ner-bert is primarily written in Jupyter Notebook.
- How popular is ner-bert?
- ai-forever/ner-bert has 408 stars on GitHub.
- Where can I find ner-bert?
- ai-forever/ner-bert is on GitHub at https://github.com/ai-forever/ner-bert.