Atenrev/diffusion_continual_learning

PyTorch implementation of various distillation approaches for continual learning of Diffusion Models.

25
/ 100
Experimental

This project helps machine learning practitioners who work with generative AI models to prevent 'catastrophic forgetting' when updating their models with new data. It takes an existing diffusion model and new datasets, then trains the model incrementally to retain its previous knowledge while learning new patterns. The outcome is a continually updated diffusion model that can generate high-quality images reflecting all learned data, without losing the ability to generate from earlier training.

No commits in the last 6 months.

Use this if you need to continually update your image-generating diffusion models with new data over time without having to retrain from scratch or sacrifice performance on previously learned tasks.

Not ideal if you are a non-technical user or are looking for a pre-trained, ready-to-use image generation tool rather than a framework for developing and evaluating continual learning strategies.

generative-ai diffusion-models continual-learning image-synthesis model-training
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

27

Forks

3

Language

Python

License

Last pushed

Mar 04, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Atenrev/diffusion_continual_learning"

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