multiagent-particle-envs
multiagent-particle-envs copied to clipboard
openai
Code for a multi-agent particle environment used in the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"
https://github.com/openai/multiagent-particle-envs/issues/53 Change a random function to the new gym, and delete `prng` function.
 I use python3.6 gym0.10.5 tensorflow1.8.0,but i always get INDEX ERROR when i render the environment... i think pyglet might be a problem,but i reinstalled and did not solve the...
Refactor MPE so we can use it with RLlib. Removed scenario system which I don't really like. Extract observations and rewards function to files so we can create new ones...
How to delete a landmark when training. Thanks a lot.
run interactive.py have error ctypes.ArgumentError: argument 2: : wrong type While you can see pictures,not animations
Hi, I'm wondering how can I use the checkpoint file to continue the training?
According to the paper "our policies are parameterized by a two-layer ReLU MLP with 64 units per layer. To support discrete communication messages, we use the Gumbel-Softmax estimator [14]." However,...
Greetings, I tried to run different scenarios rather than simple.py but it just pop-up 2 empty dialogs that no interaction happening what could be the reason?
When I run any scenario other than "Simple" the run time is very long and eventually, the result is not displayed. What is the average run time? what should I...
