Kai Fricke d527c7b335 [air/benchmarks] Drop OMP_NUM_THREADS in vanilla torch/tf training (#27256) ... Ray automatically sets OMP_NUM_THREADS=1, potentially limiting multithreading in native pytorch/tensorflow. If this leads to performance differences, we should address this either in Ray Train or in Ray core. Signed-off-by: Kai Fricke <kai@anyscale.com>		2022-08-02 13:38:01 +01:00
..
air_benchmarks	[air/benchmarks] Drop OMP_NUM_THREADS in vanilla torch/tf training (#27256)	2022-08-02 13:38:01 +01:00
horovod	[AIR] Remove ML code from ray.util (#27005)	2022-07-27 14:24:19 +01:00