update docstring in _load_duplicate_datasets

choderalab · Jul 26, 2023 · 3ccc44d · 3ccc44d
1 parent f4f23e8
commit 3ccc44d
Showing 1 changed file with 2 additions and 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
@@ -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)