schola.ray.env.BaseEnv

class schola.ray.env.BaseEnv(unreal_connection, verbosity=0)[source]

Bases: BaseEnv

A Ray RLlib environment that wraps a Schola environment.

Parameters:

• unreal_connection (UnrealConnection) – The connection to the Unreal Engine environment.

• verbosity (int, default=0) – The verbosity level for the environment.

unwrapped

The underlying multi-agent environment.

Type:

MultiAgentEnv

last_reset_obs

The observations recorded during the last reset.

Type:

Dict[int,Dict[str,Any]]

last_reset_infos

The info dict recorded during the last reset.

Type:

Dict[int,Dict[str,str]]

Methods

__init__(unreal_connection[, verbosity])
get_agent_ids()	Return the agent ids for the sub_environment.
get_sub_environments([as_dict])	Return a reference to the underlying sub environments, if any.
last()	Returns the last observations, rewards, done- truncated flags and infos …
poll()	Poll the environment for the next observation, reward, termination, info and any off_policy_actions (Currently Unused).
send_actions(action_dict)	Called to send actions back to running agents in this env.
stop()	Releases all resources used.
to_base_env([make_env, num_envs, …])	Converts an RLlib-supported env into a BaseEnv object.
try_render([env_id])	Tries to render the sub-environment with the given id or all.
try_reset([env_id, seed, options])	Attempt to reset the sub-env with the given id or all sub-envs.
try_restart([env_id])	Attempt to restart the sub-env with the given id or all sub-envs.

Attributes

action_space	The action space for the environment.
num_envs	The number of sub-environments in the wrapped environment.
observation_space	The observation space for the environment.

__init__(unreal_connection, verbosity=0)[source]

Parameters:

• unreal_connection (UnrealConnection)

• verbosity (int)

property action_space: DictSpace

The action space for the environment.

Returns:

The action space for the environment

Return type:

DictSpace

property num_envs: int

The number of sub-environments in the wrapped environment.

Returns:

The number of sub-environments in the wrapped environment.

Return type:

int

property observation_space: DictSpace

The observation space for the environment.

Returns:

The observation space for the environment.

Return type:

DictSpace

poll()[source]

Poll the environment for the next observation, reward, termination, info and any off_policy_actions (Currently Unused).

Returns:

• observations (EnvAgentIdDict[Dict[str,Any]]) – A dictionary, keyed by the environment and agent Id, containing the observations for each agent.

• rewards (EnvAgentIdDict[float]) – A dictionary, keyed by the environment and agent Id, containing the reward for each agent.

• terminateds (EnvAgentIdDict[bool]) – A dictionary, keyed by the environment and agent Id, containing the termination flag for each agent.

• truncateds (EnvAgentIdDict[bool]) – A dictionary, keyed by the environment and agent Id, containing the truncation flag for each agent.

• infos (EnvAgentIdDict[Dict[str,str]]]:) – A dictionary, keyed by the environment and agent Id, containing the information dictionary for each agent.

• off_policy_actions (EnvAgentIdDict[Any]) – A dictionary, keyed by the environment and agent Id, containing the off-policy actions for each agent. Unused.

Return type:

Tuple[Dict[int, Dict[int, Dict[str, Any]]], Dict[int, Dict[int, float]], Dict[int, Dict[int, bool]], Dict[int, Dict[int, bool]], Dict[int, Dict[int, Dict[str, str]]], Dict[int, Dict[int, Any]]]

send_actions(action_dict)[source]

Called to send actions back to running agents in this env.

Actions should be sent for each ready agent that returned observations in the previous poll() call.

Parameters:

action_dict (Dict[int, Dict[int, Dict[str, Any]]]) – Actions values keyed by env_id and agent_id.

Return type:

None

stop()[source]

Releases all resources used.

Return type:

None

try_reset(env_id=None, seed=None, options=None)[source]

Attempt to reset the sub-env with the given id or all sub-envs.

If the environment does not support synchronous reset, a tuple of (ASYNC_RESET_REQUEST, ASYNC_RESET_REQUEST) can be returned here.

Note: A MultiAgentDict is returned when using the deprecated wrapper classes such as ray.rllib.env.base_env._MultiAgentEnvToBaseEnv, however for consistency with the poll() method, a MultiEnvDict is returned from the new wrapper classes, such as ray.rllib.env.multi_agent_env.MultiAgentEnvWrapper.

Parameters:

• env_id (int | None) – The sub-environment’s ID if applicable. If None, reset the entire Env (i.e. all sub-environments).

• seed (None | List[int] | int) – The seed to be passed to the sub-environment(s) when resetting it. If None, will not reset any existing PRNG. If you pass an integer, the PRNG will be reset even if it already exists.

• options (Dict[str, str] | None) – An options dict to be passed to the sub-environment(s) when resetting it.

Returns:

A tuple consisting of a) the reset (multi-env/multi-agent) observation dict and b) the reset (multi-env/multi-agent) infos dict. Returns the (ASYNC_RESET_REQUEST, ASYNC_RESET_REQUEST) tuple, if not supported.