examples.rllib_agent module

class examples.rllib_agent.RLLibTFSavedModelAgent(path_to_model, observation_space)[source]
act(obs)[source]

The agent action. See documentation on observations, AgentSpec, and AgentInterface.

Expects an adapted observation and returns an unadapted action.

class examples.rllib_agent.TrainingModel(obs_space, action_space, num_outputs, model_config, name)[source]
NAME = 'FullyConnectedNetwork'
examples.rllib_agent.action_adapter(model_action)[source]
examples.rllib_agent.observation_adapter(env_observation)[source]
examples.rllib_agent.reward_adapter(env_obs, env_reward)[source]