synbols.stratified_splits package

Module contents

synbols.stratified_splits.compositional_split(part_map_1, part_map_2)

Use 2 partition maps to create compositional split

synbols.stratified_splits.make_compositional_split(attr_list, axis1_name, axis2_name, ratios)

synbols.stratified_splits.make_default_splits(attr_list, ratios, random_seed)

synbols.stratified_splits.make_stratified_split(attr_list, axis_name, ratios)

synbols.stratified_splits.partition_array(values, ratios)

General function to partition an array into sub-arrays of relative size determined by ratios

synbols.stratified_splits.partition_map_to_mask(part_map)

Return a list of masks for each unique value in part_map

synbols.stratified_splits.percentile_partition(values, ratios)

Split according to percentiles of values.

synbols.stratified_splits.plot_split_2d(masks, attr_x, attr_y, name_x='x', name_y='y', mask_names=None)

synbols.stratified_splits.random_map(n, ratios, rng=<module 'numpy.random' from '/home/docs/checkouts/readthedocs.org/user_builds/synbols/envs/latest/lib/python3.7/site-packages/numpy/random/__init__.py'>)

synbols.stratified_splits.str_to_id(values)

synbols.stratified_splits.stratified_split(attr_list, attr_name, ratios, rng=<module 'numpy.random' from '/home/docs/checkouts/readthedocs.org/user_builds/synbols/envs/latest/lib/python3.7/site-packages/numpy/random/__init__.py'>)

synbols.stratified_splits.unique_class_based_partition(values, ratios, rng=None)

Split according to unique values

synbols.stratified_splits.verify_part_mask(part_mask, n, ratios, verify_ratios=True)