schola.scripts.sb3.utils.RewardCallback

class schola.scripts.sb3.utils.RewardCallback(verbose=0, frequency=10, num_envs=1)[source]

Bases: CallbackList

Callback for logging rewards and steps taken by each environment in a multi-env setting.

Parameters:

• verbose (int, default=0) – Verbosity level.

• frequency (int, default=10) – The frequency at which to log the rewards and steps taken.

• num_envs (int, default=1) – The number of environments to log rewards and steps for.

num_envs

The number of environments to log rewards and steps for.

Type:

int

callbacks

The list of RewardLoggingCallbacks for each environment.

Type:

List[SingleEnvRewardCallback]

summarize_every

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

Type:

int

curr_logging_interval

The current logging interval.

Type:

int

start_time

The time at which the callback was created.

Type:

int

Methods

__init__([verbose, frequency, num_envs])
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.
update_locals(locals_)	Update the references to the local variables.

Attributes

ready_to_log	Returns whether all environments are ready to log.
model
logger

__init__(verbose=0, frequency=10, num_envs=1)[source]

Parameters:

• verbose (int)

• frequency (int)

• num_envs (int)

property ready_to_log: bool

Returns whether all environments are ready to log.

Returns:

Whether all environments are ready to log.

Return type:

bool