diff --git a/spacy/training/loop.py b/spacy/training/loop.py
index 7f67aa2cf..71fcc65cb 100644
--- a/spacy/training/loop.py
+++ b/spacy/training/loop.py
@@ -511,12 +511,15 @@ def train_while_improving(
             break
 
 
-def subdivide_batch(batch, accumulate_gradient):
+def subdivide_batch(
+    batch: Union[Iterable[Doc], Iterable[Example]], accumulate_gradient
+):
     batch = list(batch)
-    if isinstance(batch, Example):
-        batch.sort(key=lambda eg: len(eg.predicted))
-    elif isinstance(batch, Doc):
-        batch.sort(key=lambda doc: len(doc))
+    if len(batch):
+        if isinstance(batch[0], Example):
+            batch.sort(key=lambda eg: len(eg.predicted))
+        else:
+            batch.sort(key=lambda doc: len(doc))
     sub_len = len(batch) // accumulate_gradient
     start = 0
     for i in range(accumulate_gradient):