commit b414123264cacca98a5ff8c91b2efe2d0f0b4b9f
author Tyler Davis <dvisnty@amazon.com> Tue Dec 22 15:52:38 2020 -0800
committer Facebook GitHub Bot <facebook-github-bot@users.noreply.github.com> Tue Dec 22 15:54:27 2020 -0800
tree 837024bb01a1b96f00a127025ad84caa91a2acc9
parent 67d0c1824178c45430ea6c81137f84c276d3b20b

Update `is_floating_point()` docs to mention bfloat16 (#49611)

Summary:
Fixes https://github.com/pytorch/pytorch/issues/49610 . Explicitly mentions that `is_floating_point()` will return `True` if passed a `bfloat16` tensor.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/49611

Reviewed By: mrshenli

Differential Revision: D25660723

Pulled By: VitalyFedyunin

fbshipit-source-id: 04fab2f6c1c5c2859c6efff1976a92a676b9efa3

torch/_torch_docs.py

diff --git a/torch/_torch_docs.py b/torch/_torch_docs.py
index 0294942..ae0ffd9 100644
--- a/torch/_torch_docs.py
+++ b/torch/_torch_docs.py
@@ -3865,7 +3865,7 @@
 is_floating_point(input) -> (bool)
 
 Returns True if the data type of :attr:`input` is a floating point data type i.e.,
-one of ``torch.float64``, ``torch.float32`` and ``torch.float16``.
+one of ``torch.float64``, ``torch.float32``, ``torch.float16``, and ``torch.bfloat16``.
 
 Args:
     {input}