- Python >= 3.8
- Make
$ pip install git+https://github.com/peldszus/groupwise-stratified-kfold
from groupwise_stratified_kfold import GroupwiseStratifiedKFold
from groupwise_stratified_kfold import RepeatedGroupwiseStratifiedKFold
grouped_labels = {
"g01": "BBDE",
"g02": "CBEDF",
"g03": "BBAA",
"g04": "ABCD",
"g05": "ABBDF",
"g06": "ABC",
"g07": "ABBAA",
"g08": "ACBD",
"g09": "DEBBA",
"g10": "AABC",
"g11": "AAAAF",
"g12": "CCDACD",
"g13": "CCADB",
"g14": "CBAF",
"g15": "ABCD",
"g16": "CDBD",
}
k = 4
for train_group_ids, test_group_ids in GroupwiseStratifiedKFold(
k, grouped_labels, shuffle=False, seed=0
):
print(test_group_ids)
for train_group_ids, test_group_ids, iteration_id in RepeatedGroupwiseStratifiedKFold(
k, grouped_labels, shuffle=True, seed=0, repeats=2
):
print(iteration_id, test_group_ids)- The algorithm prefers similar class distributions over similar sized folds i.e. that folds might vary in size depending on how much the groups vary in size.
Run make to build, make test to test and make lint to lint.
This piece of code was taken from https://github.com/peldszus/evidencegraph/blob/fba4bc10dada1dc1574b5110be77263b5a8a1b6e/src/evidencegraph/folding.py and turned into a small library.