Fix unintended zero-only target in iou example (#4262)

pytorch_lightning/metrics/functional/classification.py

@@ -1024,11 +1024,11 @@ def iou(
 
     Example:
 
-        >>> target = torch.randint(0, 1, (10, 25, 25))
+        >>> target = torch.randint(0, 2, (10, 25, 25))
         >>> pred = torch.tensor(target)
         >>> pred[2:5, 7:13, 9:15] = 1 - pred[2:5, 7:13, 9:15]
         >>> iou(pred, target)
-        tensor(0.4914)
+        tensor(0.9660)
 
     """
     num_classes = get_num_classes(pred=pred, target=target, num_classes=num_classes)