schola.scripts.sb3.utils.SingleEnvRewardCallback

class schola.scripts.sb3.utils.SingleEnvRewardCallback(verbose=0, id=0, frequency=10)[source]

Bases: BaseCallback

Callback for logging rewards and steps taken by a single environment inside a vector environment.

Parameters:

• verbose (int) – Verbosity level.

• id (int) – The id of the environment to log rewards and steps for.

• frequency (int) – The frequency at which to log the rewards and steps taken.

episode_reward

The reward for the current episode.

Type:

float

episode_rewards

The rewards for each episode.

Type:

List[float]

episode_steps

The number of steps taken in the current episode.

Type:

int

step_count

The number of steps taken in each episode.

Type:

List[int]

last_logging_interval

The last interval that was logged.

Type:

int

logging_interval_size

The frequency at which to log the rewards and steps taken.

Type:

int

id

The id of the environment to log rewards and steps for.

Type:

int

Methods

__init__([verbose, id, frequency])
get_reward_interval()	Returns the rewards for the last logging interval.
get_step_interval()	Returns the steps taken for each episode in the last logging interval.
increment_logging_interval()	Increments the logging interval by self.logging_interval_size steps.
init_callback(model)	Initialize the callback by saving references to the RL model and the training environment for convenience.
on_rollout_end()
on_rollout_start()
on_step()	This method will be called by the model after each call to env.step().
on_training_end()
on_training_start(locals_, globals_)
update_child_locals(locals_)	Update the references to the local variables on sub callbacks.
update_locals(locals_)	Update the references to the local variables.

Attributes

ready_to_log	Returns whether the environment is ready to log, by checking if there are more episodes completed than self.logging_interval_size since we last logged.
model
logger

__init__(verbose=0, id=0, frequency=10)[source]

get_reward_interval()[source]

Returns the rewards for the last logging interval.

Returns:

The rewards for the last logging interval.

Return type:

List[float]

get_step_interval()[source]

Returns the steps taken for each episode in the last logging interval.

Returns:

The steps taken for each episode in the last logging interval.

Return type:

List[int]

increment_logging_interval()[source]

Increments the logging interval by self.logging_interval_size steps.

Return type:

None

property ready_to_log: bool

Returns whether the environment is ready to log, by checking if there are more episodes completed than self.logging_interval_size since we last logged.

Returns:

Whether the environment is ready to log.

Return type:

bool