mashijie1028/TrustDD

(Pattern Recognition 2025) Towards Trustworthy Dataset Distillation

21
/ 100
Experimental

This helps deep learning practitioners create smaller, synthetic datasets that efficiently train models for image classification and detect unusual, 'out-of-distribution' images. You input a large, real image dataset, and it outputs a tiny, distilled dataset. Data scientists or machine learning engineers in charge of deploying reliable deep learning models would use this.

No commits in the last 6 months.

Use this if you need to train deep learning models efficiently on smaller datasets while ensuring they can reliably identify data that doesn't fit their training distribution, making them safer for real-world use.

Not ideal if your primary concern is solely in-distribution classification performance without any need for identifying out-of-distribution data or if you prefer to use the full, original dataset for training.

deep-learning-efficiency image-classification outlier-detection model-trustworthiness dataset-optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

14

Forks

Language

Python

License

MIT

Last pushed

Dec 08, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mashijie1028/TrustDD"

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