kamalkraj/BERT-NER

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1249 stars · 272 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch library for training and running named entity recognition (NER) models based on Google's BERT, evaluated on the CoNLL-2003 dataset. It also provides a C++ inference path so trained models can be deployed without a Python runtime.

Use cases

• extract names, locations, and organizations from text
• train a BERT model for named entity recognition on CoNLL-2003
• run NER inference in C++ for production deployment
• fine-tune pretrained BERT models on custom NER data
• tag tokens in a sentence with entity labels and confidence scores

When to choose

• you need a simple, proven BERT-based NER pipeline in PyTorch
• you want fast C++ inference without Python dependencies
• you are working with the CoNLL-2003 entity extraction task

When to avoid

• you need multi-language or multilingual NER out of the box
• you want a maintained library with recent updates and modern transformer support
• you need a full NLP toolkit rather than a focused NER implementation

Facets

library · maturity maintenance

machine-learning nlp llm-training machine-learning deep-learning python cpp bert named-entity-recognition pytorch conll-2003 ner inference transfer-learning natural-language-processing linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="kamalkraj/BERT-NER")