nip.utils.experiments.ExperimentFunctionArguments

class nip.utils.experiments.ExperimentFunctionArguments(combo: dict, run_id: str, cmd_args: ~argparse.Namespace, common_run_name: str, tqdm_func: callable, global_tqdm_step_fn: callable = <function ExperimentFunctionArguments.<lambda>>, log_level: int = 20)

Arguments to the function which runs a single experiment.

Parameters:

• combo (dict) – A single combination of hyperparameters.

• run_id (str) – A unique identifier for the run.

• cmd_args (Namespace) – The command line arguments.

• common_run_name (str) – A name for the experiment that is common to all runs.

• tqdm_func (callable) – A function used to create a tqdm progress bar.

• global_tqdm_step_fn (callable) – A function used to update the global progress bar.

• log_level (int) – The log level to use for the experiment. Can be used with logging.setLevel.

Methods Summary

__eq__(other)	Return self==value.
__init__(combo, run_id, cmd_args, ...[, ...])
__repr__()	Return repr(self).
global_tqdm_step_fn()

Attributes

log_level
combo
run_id
cmd_args
common_run_name
tqdm_func

Methods

__eq__(other)

Return self==value.

__init__(combo: dict, run_id: str, cmd_args: ~argparse.Namespace, common_run_name: str, tqdm_func: callable, global_tqdm_step_fn: callable = <function ExperimentFunctionArguments.<lambda>>, log_level: int = 20) → None

__repr__()

Return repr(self).

global_tqdm_step_fn()