releso.callback.StepLogCallback

class releso.callback.StepLogCallback(step_log_location: Path, verbose: int = 0, update_n_steps: int = 0, log_infos: bool = False, logger=None)

Bases: BaseCallback

Step Callback class.

This class tracks all step-wise information that might come in handy during evaluation. Originally created to easily track the actions undertaken in each episode for better evaluation of the learned policy.

__init__(step_log_location: Path, verbose: int = 0, update_n_steps: int = 0, log_infos: bool = False, logger=None)

Constructor for the Callback using SB3 interface.

Parameters:

  • step_log_location (Path) – Path to the step log file.

  • verbose (int, optional) – Verbosity of the callback. Defaults to 0.

  • update_every (int, optional) – Update the step log file every n

  • episode. (steps. Defaults to 0 which triggers the update after every)

  • log_infos (bool, optional) – Whether to log the infos. Defaults to

  • False.

  • logger (optional) – Logger to use for logging. If None, a default

  • None. (logger is created. Defaults to)

Methods

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

logger

training_env

model