I’m currently relying on tune.Tuner to run my experiment on a machine that has 28 CPUs and 2 GPUs. Being that I’m not the only one who has access to this machine, I’d like to restrict my experiment to a single GPU. Despite specifying with_resources(Trainer, {"cpu": 1, "gpu": 1}) both GPUs are used.…

Hi @mtt , tune.with_resources sets the resources per trial, and since you have 5 trials with 2 GPUs, Tune will schedule 2 trials at a time, each taking one of the GPUs. You can specify fractional GPUs per trial, as well as limit concurrency so you never go above some GPU usage. For example, this wi…

How do I run my experiment on a single GPU?

Xinchengzelin August 20, 2023, 12:53am 5

Thanks @matthewdeng, I create a new topic: How to use fraction GPU in `ray.tune.Tuner`?

Topic		Replies	Views
Multiple trials on each GPU Ray Tune	1	493	February 19, 2021
How to use fraction GPU in `ray.tune.Tuner`? Ray Train	6	1228	August 24, 2023
Using specific GPUs in a shared machine Ray Tune	6	3059	March 24, 2022
Different trial on CPU and GPU separately? Ray Tune	8	560	April 4, 2022
Training trials in parallel on multi-gpu machine Ray Tune	8	1764	August 23, 2021

How do I run my experiment on a single GPU?

Related topics