search

datamllab / rlcard

Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.
http://www.rlcard.org
MIT License
2.87k stars 619 forks source link

Training only uses ~50% of CPU #307

Closed amcneel closed 9 months ago

amcneel commented 9 months ago

Are there any parameter tweaks or ways of optimizing that you're aware of to have the training make as much use of the CPU as possible? I'm running the nlh agent.

amcneel commented 9 months ago

note this was resolved with torch.set_num_threads :)