mirror of
https://github.com/pytorch/pytorch.git
synced 2025-11-04 08:00:58 +08:00
6f1d99b79f
Fixes https://github.com/pytorch/pytorch/issues/80839 Pull Request resolved: https://github.com/pytorch/pytorch/pull/80882 Approved by: https://github.com/jbschlosser
30 lines
778 B
ReStructuredText
30 lines
778 B
ReStructuredText
.. role:: hidden
|
|
:class: hidden-section
|
|
|
|
.. _nn-init-doc:
|
|
|
|
torch.nn.init
|
|
=============
|
|
|
|
.. warning::
|
|
All the functions in this module are intended to be used to initialize neural network
|
|
parameters, so they all run in :func:`torch.no_grad` mode and will not be taken into
|
|
account by autograd.
|
|
|
|
.. currentmodule:: torch.nn.init
|
|
.. autofunction:: calculate_gain
|
|
.. autofunction:: uniform_
|
|
.. autofunction:: normal_
|
|
.. autofunction:: constant_
|
|
.. autofunction:: ones_
|
|
.. autofunction:: zeros_
|
|
.. autofunction:: eye_
|
|
.. autofunction:: dirac_
|
|
.. autofunction:: xavier_uniform_
|
|
.. autofunction:: xavier_normal_
|
|
.. autofunction:: kaiming_uniform_
|
|
.. autofunction:: kaiming_normal_
|
|
.. autofunction:: trunc_normal_
|
|
.. autofunction:: orthogonal_
|
|
.. autofunction:: sparse_
|