Skip to content

Fix Glorot/He uniform init docstrings to label the uniform bound, not sigma #3831

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
vineethsaivs wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −6
Open

Fix Glorot/He uniform init docstrings to label the uniform bound, not sigma #3831

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 8 additions & 6 deletions python/mlx/nn/init.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -194,12 +194,13 @@ def glorot_uniform(
) -> Callable[[mx.array, float], mx.array]:
r"""A Glorot uniform initializer.

This initializer samples from a uniform distribution with a range
computed from the number of input (``fan_in``) and output (``fan_out``)
This initializer samples from a uniform distribution on the interval
:math:`[-\text{limit}, \text{limit}]`, where the bound :math:`\text{limit}`
is computed from the number of input (``fan_in``) and output (``fan_out``)
units according to:

.. math::
\sigma = \gamma \sqrt{\frac{6.0}{\text{fan\_in} + \text{fan\_out}}}
\text{limit} = \gamma \sqrt{\frac{6.0}{\text{fan\_in} + \text{fan\_out}}}

For more details see the original reference: `Understanding the difficulty
of training deep feedforward neural networks
Expand Down Expand Up @@ -295,13 +296,14 @@ def he_uniform(
) -> Callable[[mx.array, Literal["fan_in", "fan_out"], float], mx.array]:
r"""A He uniform (Kaiming uniform) initializer.

This initializer samples from a uniform distribution with a range
computed from the number of input (``fan_in``) or output (``fan_out``)
This initializer samples from a uniform distribution on the interval
:math:`[-\text{limit}, \text{limit}]`, where the bound :math:`\text{limit}`
is computed from the number of input (``fan_in``) or output (``fan_out``)
units according to:

.. math::

\sigma = \gamma \sqrt{\frac{3.0}{\text{fan}}}
\text{limit} = \gamma \sqrt{\frac{3.0}{\text{fan}}}

where :math:`\text{fan}` is either the number of input units when the
``mode`` is ``"fan_in"`` or output units when the ``mode`` is
Expand Down