MarvinMartin24/MADA-PL

Multi-Adversarial Domain Adaptation (https://arxiv.org/abs/1809.02176) implementation in Pytorch-Lightning

34
/ 100
Emerging

This project helps machine learning engineers or researchers build image classification models that perform well even when the real-world images they encounter are significantly different from their training data. It takes labeled image datasets from one 'source' environment and unlabeled images from a 'target' environment, and outputs a more robust image classification model. This is for AI practitioners aiming to deploy models in varied real-world conditions.

No commits in the last 6 months.

Use this if you need to improve the performance of your image classification models on target datasets that have a different visual style, lighting, or background compared to your original training data.

Not ideal if your classification task involves text or numerical data, or if your source and target image datasets are already visually very similar.

image-classification computer-vision model-robustness machine-learning-engineering
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

18

Forks

3

Language

Jupyter Notebook

License

MIT

Last pushed

Aug 13, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MarvinMartin24/MADA-PL"

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