gather_exp_stats

hybrid_learning.experimentation.fuzzy_exp.fuzzy_exp_eval.gather_exp_stats(conf, iterator, thresh=0.001, save_as=None, force_reevaluate=False, show_progress=True)

Create a DataFrame summary of statistics of the masks created during a fuzzy logic experiment. The dataframe will feature

• rows: indexed by image file names

• columns: indexed by double index of (mask_type, value_type); mask_type``s: formula, ped, [gt, ] <concepts>; ``value_type``s: stddev, max, min, num_entries, num_interesting, mean_interesting, stddev_interesting, mean_formula_non_one, stddev_formula_non_one, mean_ped_non_zero, stddev_ped_non_zero, mean_concepts_non_zero, stddev_concepts_non_zero; ``*_interesting / *_ped_non_zero / *_concepts_non_zero means values after application of masks where the formula mask is below 1-thresh/ person mask above thresh / all concept masks above thresh.

Parameters

• iterator (Iterable[Tuple[str, Tuple[Tensor, Dict[str, Tensor]]]]) –

• thresh (float) –

• save_as (Optional[str]) –

• force_reevaluate (bool) –

• show_progress (bool) –

Return type

DataFrame