pipeline.utils.scriptutils package#

A package that defines common utilies used in the scripts.

pipeline.utils.scriptutils.configure_logger(level=20, format='%(levelname)s:%(message)s', **kwargs)[source]#

Configure the logger with the basic logging.

All the arguments are passed to logging.basicConfig().

pipeline.utils.scriptutils.parse_args(**kwargs)[source]#

Parse the arguments of a parser with a single argument config, that allows to specify the configuration.

pipeline.utils.scriptutils.convenience_utils module#

pipeline.utils.scriptutils.convenience_utils.get_training_metrics(trainer)[source]#

Get the dataframe of the training metrics.

Parameters:: trainer (Trainer | str) – either a PyTorch Lighting Trainer object, or the path to the metric file to load directly.
Return type:: pd.DataFrame
Returns:: Dataframe of the training metrics (one row / epoch).

A module to handle the log.

pipeline.utils.scriptutils.loghandler.configure_logger(level=20, format='%(levelname)s:%(message)s', **kwargs)[source]#

Configure the logger with the basic logging.

All the arguments are passed to logging.basicConfig().

This module define the default parser used in all the scripts.

pipeline.utils.scriptutils.parser.add_predefined_arguments(parser, arguments, **kwargs)[source]#

Add predefined arguments defined by the predefined_arguments dictionary.

Parameters:

parser (ArgumentParser) – Argument parser which to add the predefined arguments to
arguments (Iterable[str]) – names of predefined arguments to add

pipeline.utils.scriptutils.parser.parse_args(**kwargs)[source]#

Parse the arguments of a parser with a single argument config, that allows to specify the configuration.