Logging module

This module provides a number of automatic logging facilities to monitor continual learning experiments.
Loggers should be provided as input to the EvaluationPlugin class.

logging

Loggers

InteractiveLogger()

The InteractiveLogger class provides logging facilities for the console standard output.

TensorboardLogger([tb_log_dir, filename_suffix])

The TensorboardLogger provides an easy integration with Tensorboard logging.

WandBLogger([project_name, run_name, ...])

The WandBLogger provides an easy integration with Weights & Biases logging.

TextLogger([file])

The TextLogger class provides logging facilities printed to a user specified file.

CSVLogger([log_folder])

The CSVLogger logs accuracy and loss metrics into a csv file.

All the loggers inherit from the base class StrategyLogger.

StrategyLogger()

The base class for the strategy loggers.