composer.loggers.progress_bar_logger#
Logs metrics to the console and show a progress bar.
Classes
Log metrics to the console and optionally show a progress bar. |
- class composer.loggers.progress_bar_logger.ProgressBarLogger(progress_bar=True, log_to_console=None, console_log_level=LogLevel.EPOCH, stream=<_io.TextIOWrapper name='<stderr>' mode='w' encoding='utf-8'>)[source]#
Bases:
composer.loggers.logger_destination.LoggerDestination
Log metrics to the console and optionally show a progress bar.
Note
This logger is automatically instainatied by the trainer via the
progress_bar
,log_to_console
,log_level
, andconsole_stream
options. This logger does not need to be created manually.TQDM is used to display progress bars.
During training, the progress bar logs the batch and training loss. During validation, the progress bar logs the batch and validation accuracy.
Example progress bar output:
Epoch 1: 100%|โโโโโโโโโโ| 64/64 [00:01<00:00, 53.17it/s, loss/train=2.3023] Epoch 1 (val): 100%|โโโโโโโโโโ| 20/20 [00:00<00:00, 100.96it/s, accuracy/val=0.0995]
- Parameters
progress_bar (bool, optional) โ Whether to show a progress bar. (default:
True
)log_to_console (bool, optional) โ
Whether to print logging statements to the console. (default:
None
)The default behavior (when set to
None
) only prints logging statements whenprogress_bar
isFalse
.console_log_level (LogLevel | str | (State, LogLevel) -> bool, optional) โ
The maximum log level for which statements should be printed. (default:
LogLevel.EPOCH
)It can either be
LogLevel
, a string corresponding to aLogLevel
, or a callable that takes the trainingState
and theLogLevel
and returns a boolean of whether this statement should be printed.This parameter has no effect if
log_to_console
isFalse
or is unspecified whenprogress_bar
isTrue
.stream (str | TextIO, optional) โ The console stream to use. If a string, it can either be
'stdout'
or'stderr'
. (default:sys.stderr
)