From: serhii-havrylov <sergii.gavrylov@gmail.com>
Date: Wed, 13 Mar 2019 10:16:40 +0000 (-0700)
Subject: Update docs for `mark_non_differentiable` method (#17891)
X-Git-Tag: accepted/tizen/6.5/unified/20211028.231830~835
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=f6de833cac63016de7f0f04b260c734f19e46505;p=platform%2Fupstream%2Fpytorch.git

Update docs for `mark_non_differentiable` method (#17891)

Summary:
The current documentation doesn't reflect the real values of tensors during the backward pass.
This issue is mentioned in https://github.com/pytorch/pytorch/issues/12631
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17891

Differential Revision: D14419949

Pulled By: soumith

fbshipit-source-id: 8b495628c3f017bc880f8096682cd176a53974e5
---

diff --git a/torch/autograd/function.py b/torch/autograd/function.py
index ea48a4f..89e930a 100644
--- a/torch/autograd/function.py
+++ b/torch/autograd/function.py
@@ -51,7 +51,8 @@ class _ContextMethodMixin(object):
         This will mark outputs as not requiring gradients, increasing the
         efficiency of backward computation. You still need to accept a gradient
         for each output in :meth:`~Function.backward`, but it's always going to
-        be ``None``.
+        be a zero tensor with the same shape as the shape of a corresponding
+        output.
 
         This is used e.g. for indices returned from a max :class:`Function`.
         """