From 3ccc44d86e757c19c8eaeb525c6add4d1c40b6b6 Mon Sep 17 00:00:00 2001 From: kt Date: Wed, 26 Jul 2023 16:32:03 -0400 Subject: [PATCH] update docstring in _load_duplicate_datasets --- .../02-train/joint-improper-charge/charge-weight-1.0/train.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/openff-default/02-train/joint-improper-charge/charge-weight-1.0/train.py b/openff-default/02-train/joint-improper-charge/charge-weight-1.0/train.py index 80bfd63..48d0ff1 100644 --- a/openff-default/02-train/joint-improper-charge/charge-weight-1.0/train.py +++ b/openff-default/02-train/joint-improper-charge/charge-weight-1.0/train.py @@ -49,7 +49,8 @@ def _load_datasets(datasets, input_prefix): def _load_duplicate_datasets(ds_tr, input_prefix): """ - Load duplicated molecules (nonisomeric smiles) found from different datasets. + Load duplicated molecules (isomeric smiles) from different datasets + to avoid overlapping molecules in train, validate, test dataset. """ entries = glob.glob(os.path.join(input_prefix, "duplicated-isomeric-smiles-merge", "*")) random.seed(RANDOM_SEED)