From b2309cc6bffb5b0676b559c5932f1322a59b0811 Mon Sep 17 00:00:00 2001 From: Arnav Sharma Date: Mon, 6 Jul 2020 18:45:22 +0530 Subject: [PATCH] Typo fix in `training` doc (#5495) --- docs/source/training.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/training.rst b/docs/source/training.rst index bb06750462..c7fdc02574 100644 --- a/docs/source/training.rst +++ b/docs/source/training.rst @@ -272,7 +272,7 @@ optimize. :func:`~transformers.Trainer` uses a built-in default function to collate batches and prepare them to be fed into the model. If needed, you can also use the ``data_collator`` argument to pass your own collator function which -takes in the data in the format provides by your dataset and returns a +takes in the data in the format provided by your dataset and returns a batch ready to be fed into the model. Note that :func:`~transformers.TFTrainer` expects the passed datasets to be dataset objects from ``tensorflow_datasets``.