Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Use math.fsum for duration aggregation in SimpleProfiler (#21525)
* Update test_mlflow.py * test_mlflow.py * Optimize SimpleProfiler duration aggregation * re run ci * Add tests for SimpleProfiler extended report functionality --------- Co-authored-by: bhimrazy <bhimrajyadav977@gmail.com>
K
Kavyansh Tyagi committed
215d7d9737fc20168b188b71003f355357904d53
Parent: 7d2de87
Committed by GitHub <noreply@github.com>
on 2/3/2026, 12:05:29 PM