update docs

This commit is contained in:
Awni Hannun
2025-10-17 19:14:54 +00:00
committed by CircleCI Docs
parent f6ae46f713
commit 0f04ebb557
241 changed files with 15049 additions and 737 deletions
@@ -0,0 +1,6 @@
mlx.core.eye
============
.. currentmodule:: mlx.core
.. autofunction:: eye
@@ -0,0 +1,6 @@
mlx.core.identity
=================
.. currentmodule:: mlx.core
.. autofunction:: identity
@@ -0,0 +1,8 @@
mlx.nn.Mish
===========
.. currentmodule:: mlx.nn
.. autoclass:: Mish
@@ -0,0 +1,8 @@
mlx.nn.PReLU
============
.. currentmodule:: mlx.nn
.. autoclass:: PReLU
@@ -0,0 +1,8 @@
mlx.nn.SELU
===========
.. currentmodule:: mlx.nn
.. autoclass:: SELU
@@ -0,0 +1,8 @@
mlx.nn.Step
===========
.. currentmodule:: mlx.nn
.. autoclass:: Step
@@ -0,0 +1,8 @@
mlx.nn.losses.binary\_cross\_entropy
====================================
.. currentmodule:: mlx.nn.losses
.. autoclass:: binary_cross_entropy
@@ -0,0 +1,8 @@
mlx.nn.losses.cross\_entropy
============================
.. currentmodule:: mlx.nn.losses
.. autoclass:: cross_entropy
@@ -0,0 +1,8 @@
mlx.nn.losses.kl\_div\_loss
===========================
.. currentmodule:: mlx.nn.losses
.. autoclass:: kl_div_loss
@@ -0,0 +1,8 @@
mlx.nn.losses.l1\_loss
======================
.. currentmodule:: mlx.nn.losses
.. autoclass:: l1_loss
@@ -0,0 +1,8 @@
mlx.nn.losses.mse\_loss
=======================
.. currentmodule:: mlx.nn.losses
.. autoclass:: mse_loss
@@ -0,0 +1,8 @@
mlx.nn.losses.nll\_loss
=======================
.. currentmodule:: mlx.nn.losses
.. autoclass:: nll_loss
@@ -0,0 +1,8 @@
mlx.nn.mish
===========
.. currentmodule:: mlx.nn
.. autoclass:: mish
@@ -0,0 +1,8 @@
mlx.nn.prelu
============
.. currentmodule:: mlx.nn
.. autoclass:: prelu
@@ -0,0 +1,8 @@
mlx.nn.selu
===========
.. currentmodule:: mlx.nn
.. autoclass:: selu
@@ -0,0 +1,8 @@
mlx.nn.step
===========
.. currentmodule:: mlx.nn
.. autoclass:: step
+23 -1
View File
@@ -97,7 +97,7 @@ Updating the parameters
MLX modules allow accessing and updating individual parameters. However, most
times we need to update large subsets of a module's parameters. This action is
performed by :meth:`Module.update`.
performed by :meth:`Module.update`.
Value and grad
--------------
@@ -146,8 +146,12 @@ Neural Network Layers
Embedding
ReLU
PReLU
GELU
SiLU
Step
SELU
Mish
Linear
Conv1d
Conv2d
@@ -169,4 +173,22 @@ simple functions.
gelu_approx
gelu_fast_approx
relu
prelu
silu
step
selu
mish
Loss Functions
--------------
.. autosummary::
:toctree: _autosummary_functions
:template: nn-module-template.rst
losses.cross_entropy
losses.binary_cross_entropy
losses.l1_loss
losses.mse_loss
losses.nll_loss
losses.kl_div_loss
+2
View File
@@ -38,9 +38,11 @@ Operations
erfinv
exp
expand_dims
eye
full
greater
greater_equal
identity
less
less_equal
load