ray/release/air_tests at master - hiro/ray - Forgejo: Beyond coding. We Forge.

hiro/ray

mirror of https://github.com/vale981/ray synced 2025-03-04 17:41:43 -05:00

History

Kai Fricke b91246a093 [air/benchmarks] Measure local training time in torch/tf benchmarks (#27902 ) We currently measure end-to-end training time in our benchmarks, which includes setup overhead. This is an unequal comparison, as setup overhead for vanilla training cannot be accurately expressed and was instead just disregarded. By comparing the raw training times in the actual training loop, we will get a more accurate expression of any potential overhead or benefit in using Ray vs. vanilla tensorflow/torch. Signed-off-by: Kai Fricke <kai@anyscale.com>	2022-08-16 19:16:08 +02:00
..
air_benchmarks	[air/benchmarks] Measure local training time in torch/tf benchmarks (#27902 )	2022-08-16 19:16:08 +02:00
horovod	[AIR] Remove ML code from `ray.util` (#27005 )	2022-07-27 14:24:19 +01:00

[air/benchmarks] Measure local training time in torch/tf benchmarks (#27902 )

We currently measure end-to-end training time in our benchmarks, which includes setup overhead. This is an unequal comparison, as setup overhead for vanilla training cannot be accurately expressed and was instead just disregarded.
By comparing the raw training times in the actual training loop, we will get a more accurate expression of any potential overhead or benefit in using Ray vs. vanilla tensorflow/torch.

Signed-off-by: Kai Fricke <kai@anyscale.com>

2022-08-16 19:16:08 +02:00

air_benchmarks [air/benchmarks] Measure local training time in torch/tf benchmarks (#27902 ) 2022-08-16 19:16:08 +02:00

horovod [AIR] Remove ML code from ray.util (#27005 ) 2022-07-27 14:24:19 +01:00