rl_a3c_pytorch icon indicating copy to clipboard operation
rl_a3c_pytorch copied to clipboard

run a3c on 8 cpus, it still slow.

Open TIANRENK opened this issue 1 year ago • 1 comments

When I runing a3c on 8 cpus, it still slow.
My cpu is Xeon(R) Platinum 8255C. Is this the reason for my poor cpu performance or torch multiprocessing 's problem?

TIANRENK avatar Jun 21 '23 09:06 TIANRENK

I suspect you need more resources (CPUs/GPUs/Compute Time). It took me hours to days to build successful networks on this machine: Lambda Deep Learning GPU Workstation 3x 1080Ti Quad DevBox Server/Workstation (presently for sale): https://www.ebay.com/itm/185934565514 . If you want a platform that runs the models to mastery, you'll need workstation-level computing power.

Solution Videos: Space Invaders: https://www.instagram.com/p/BtXNrJCnbDU/?utm_source=ig_web_copy_link&igshid=MzRlODBiNWFlZA==

Beam Rider: https://www.instagram.com/p/BtXH-50nXtO/?utm_source=ig_web_copy_link&igshid=MzRlODBiNWFlZA==

StrategicVisionary avatar Jun 21 '23 14:06 StrategicVisionary