Revision 6c05add8ae69fb25e09b3bb537ea90a2a3033b95 authored by yashkatariya on 15 November 2022, 20:59:00 UTC, committed by yashkatariya on 15 November 2022, 20:59:00 UTC
1 parent 233bf21
jax.nn.initializers.rst
jax.nn.initializers package
===========================
.. currentmodule:: jax.nn.initializers
.. automodule:: jax.nn.initializers
Initializers
------------
This module provides common neural network layer initializers,
consistent with definitions used in Keras and Sonnet.
An initializer is a function that takes three arguments:
``(key, shape, dtype)`` and returns an array with dimensions ``shape`` and
data type ``dtype``. Argument ``key`` is a :class:`jax.random.PRNGKey` random
key used when generating random numbers to initialize the array.
.. autosummary::
:toctree: _autosummary
constant
delta_orthogonal
glorot_normal
glorot_uniform
he_normal
he_uniform
lecun_normal
lecun_uniform
normal
ones
orthogonal
uniform
variance_scaling
zeros
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...