starry-sky6688 / MADDPG

Pytorch implementation of the MARL algorithm, MADDPG, which correspondings to the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments".
516 stars 80 forks source link

set evaluate==False ,no model saved in model #2

Closed xzxgg closed 3 years ago

starry-sky6688 commented 3 years ago

Directly modify the evaluate to False in arguments.py.

If you just pass the arguments by cmd,the evaluate still be True.