Use separate `rng` for dropout in FastMRI `model_init`

mlcommons / algorithmic-efficiency

MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvements in both training algorithms and models.

https://mlcommons.org/en/groups/research-algorithms/

Apache License 2.0

321 stars 62 forks source link

Use separate `rng` for dropout in FastMRI `model_init` #672

Closed runame closed 6 months ago

runame commented 6 months ago

Fixes #664. Not sure why this worked before and stopped working now?

github-actions[bot] commented 6 months ago

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅