AOoligei
AOoligei
Anothon question.当我运行到100时， `Episode: 96, reward = -90.833405 Episode: 97, reward = -82.624748 Episode: 98, reward = -79.773178 Episode: 99, reward = -82.121429 training now begins... MADDPG on WaterWorld scale_reward=0.010000 agent=2,...
> I fixed this warning by changing objects of type ByteTensor to BoolTensor instead. This was the source of the warnings for me.
Hello, I can't get the result in the picture(All the agents are on the side). How did you set the parameters to get the result in the PNG（demo.gif）?
Have you solved the problem？Can you tell me how do I change the landmark from stationary to random walk?