search

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

dev -> main #670

Closed priyakasimbeg closed 6 months ago

priyakasimbeg commented 6 months ago

Tests passed in https://github.com/mlcommons/algorithmic-efficiency/actions/runs/8133680125/job/22272986163?pr=670.

Workload variant additions and fixes:

Add prize qualification logs for external tuning ruleset. Note: FastMRI trials with dropout are not yet added due to https://github.com/mlcommons/algorithmic-efficiency/issues/664.

Add missing funcitonality to Docker startup script for self_tuning ruleset. Add self_tuning ruleset option to script that runs all workloads for scoring.

Datasetup fixes.

Fix tests that check training differences in PyTorch and JAX on GPU.

github-actions[bot] commented 6 months ago

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