john-osborne-j/quantized-clinicalbert

This repository contains a 4-bit quantized ClinicalBERT model for disease classification based on clinical text. Inspired by CheXNet, this model can predict diseases from patient symptom descriptions, particularly focusing on chest-related conditions.

24
/ 100
Experimental

Implements 4-bit quantization via BitsAndBytes to reduce model footprint while maintaining inference performance, enabling deployment on memory-constrained environments. Trained on synthetic clinical text datasets to avoid privacy constraints, with a Flask web interface for direct symptom-to-disease prediction. Integrates with Hugging Face Transformers and supports Google Colab deployment for accessible prototyping.

No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 3 / 25
Maturity 9 / 25
Community 12 / 25

How are scores calculated?

Stars

4

Forks

1

Language

Python

License

MIT

Last pushed

Apr 04, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/john-osborne-j/quantized-clinicalbert"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.