cage-challenge / cage-challenge-4

The TTCP CAGE Challenges are a series of public challenges instigated to foster the development of autonomous cyber defensive agents. This CAGE Challenge 4 (CC4) returns to a defence industry enterprise environment, and introduces a Multi-Agent Reinforcement Learning (MARL) scenario.
https://cage-challenge.github.io/cage-challenge-4/
Other
41 stars 10 forks source link

need to call policy_mapper, not policy_mapper_function in example code? #9

Closed bodacea closed 9 months ago

bodacea commented 9 months ago

in https://github.com/cage-challenge/cage-challenge-4/blob/main/documentation/docs/pages/tutorials/01_Getting_Started/3_Training_Agents.md the code below uses policy_mapper_func, when the code above creates policy_mapper. Should the example call be to policy_mapper instead?

algo_config = ( PPOConfig() .environment(env="CC4") .multi_agent(policies={ ray_agent: PolicySpec( policy_class=None, observation_space=env.observation_space(cyborg_agent), action_space=env.action_space(cyborg_agent), config={"gamma": 0.85}, ) for cyborg_agent, ray_agent in POLICY_MAP.items() }, policy_mapping_fn=policy_mapper_func ))