google-deepmind/distribution_shift_framework

This repository contains the code of the distribution shift framework presented in A Fine-Grained Analysis on Distribution Shift (Wiles et al., 2022).

47
/ 100
Emerging

This framework helps machine learning researchers understand how different training methods perform when the data distribution shifts between training and testing. It takes your specified dataset, model architecture, and training algorithm, and allows you to simulate various real-world data shift scenarios, outputting performance metrics for each. It's designed for researchers evaluating the robustness of machine learning models to changes in data.

Use this if you are a machine learning researcher focused on evaluating and improving model generalization under various forms of data distribution shift, such as unseen data or spurious correlations.

Not ideal if you are looking for a plug-and-play solution to deploy models, or if you are not actively researching machine learning generalization and distribution shifts.

machine-learning-research model-robustness dataset-shift out-of-distribution-generalization algorithmic-fairness
No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

86

Forks

9

Language

Python

License

Apache-2.0

Last pushed

Feb 20, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/google-deepmind/distribution_shift_framework"

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