Skip to content

Clarify explanation of requires_grad in PyTorch #3717

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
FlightVin wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Clarify explanation of requires_grad in PyTorch #3717

Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
6772537
Clarify explanation of requires_grad in PyTorch
FlightVin Jan 4, 2026
d00fbc6
Merge branch 'main' into patch-1
sekyondaMeta Jan 8, 2026
207c0d3
Update nn_tutorial.py
sekyondaMeta Jan 8, 2026
bd36235
Merge branch 'main' into patch-1
svekars Jan 8, 2026
dba5b9b
Remove trailing spaces
FlightVin Jan 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions beginner_source/nn_tutorial.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,9 +110,10 @@
# gradient. This causes PyTorch to record all of the operations done on the tensor,
# so that it can calculate the gradient during back-propagation *automatically*!
#
# For the weights, we set ``requires_grad`` **after** the initialization, since we
# don't want that step included in the gradient. (Note that a trailing ``_`` in
# PyTorch signifies that the operation is performed in-place.)
# We set ``requires_grad`` **after** the initialization math. If enabled earlier,
# PyTorch would view the weights as the result of a calculation (non-leaf) rather
# than a source parameter, making them impossible to optimize. (Note that a trailing
# ``_`` in PyTorch signifies that the operation is performed in-place.)
#
# .. note:: We are initializing the weights here with
# `Xavier initialisation <http://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf>`_
Expand Down
Loading