diff --git a/pina/data/data_module.py b/pina/data/data_module.py index 9ef419f..127694d 100644 --- a/pina/data/data_module.py +++ b/pina/data/data_module.py @@ -81,7 +81,7 @@ class Collator: :param dict max_conditions_lengths: ``dict`` containing the maximum number of data points to consider in a single batch for each condition. - :param bool automatic_batching: Whether to enable automatic batching. + :param bool automatic_batching: Whether to enable automatic batching. If ``True``, automatic PyTorch batching is performed, which consists of extracting one element at a time from the dataset and collating them into a batch. This is useful