YyzHarry/multi-domain-imbalance

[ECCV 2022] Multi-Domain Long-Tailed Recognition, Imbalanced Domain Generalization, and Beyond

40
/ 100
Emerging

This project helps machine learning practitioners address challenges when building classification models from image data that comes from multiple different sources or domains, and where some categories are under-represented or 'long-tailed'. It takes existing image datasets (like those used for object recognition or medical imaging) and provides tools to benchmark and improve model performance across these varied and imbalanced data distributions. Anyone building robust computer vision models from diverse, real-world data would benefit.

141 stars. No commits in the last 6 months.

Use this if your image classification models struggle with accuracy when applied to new environments or when classifying rare objects because your training data comes from various sources with uneven category representation.

Not ideal if your data is perfectly balanced, comes from a single, consistent source, or if you are not working with image classification tasks.

computer-vision image-recognition machine-learning-engineering dataset-analysis model-generalization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

141

Forks

16

Language

Python

License

MIT

Last pushed

Jan 26, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/YyzHarry/multi-domain-imbalance"

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