commonpower.control.logging_utils.loggers.MARLTensorboardLogger

class MARLTensorboardLogger(log_dir: str, callback: ~commonpower.control.logging_utils.callbacks.MARLBaseCallback = <class 'commonpower.control.logging_utils.callbacks.MARLBaseCallback'>, format_strings: list = ['stdout', 'tensorboard'])

Bases: BaseLogger

Class for using tensorboard logging in multi-agent IPPO/MAPPO algorithms from the on-policy repository (https://github.com/marlbenchmark/on-policy/blob/main/README.md).

Parameters:

• log_dir (str) – relative path to logging directory

• callback (MARLBaseCallback) – object that implements actual logging during training. By defining a customized callback, additional information can be logged (apart from standard metrics like mean_eps_reward)

• format_strings (list) – list of output formats for the SB3 logger

Methods

finish_logging
get_callback	Hands over the callback.
get_log_dir
get_log_function	Hands over the logger we get from stable-baselines3
log_function

get_callback() → MARLBaseCallback

Hands over the callback.

Returns:

MARLBaseCallback – callback which is used during training to log additional information

get_log_function() → Callable

Hands over the logger we get from stable-baselines3

Returns:

Callable – Logger