Crate concision_init

concision-init

This library provides various random distribution and initialization routines for the concision framework. It includes implementations for different initialization strategies optimized for neural networks, such as Glorot (Xavier) initialization, LeCun initialization, etc.

Re-exports

pub use rand;

pub use rand_distr;

Modules

distr

this module implements various random distributions optimized for neural network initialization.

error

Structs

LecunNormal

LecunNormal is a truncated normal distribution centered at 0 with a standard deviation that is calculated as:

TruncatedNormal

The TruncatedNormal distribution is similar to the StandardNormal distribution, differing in that is computes a boundary equal to two standard deviations from the mean. More formally, the boundary is defined as:

XavierNormal

Normal Xavier initializers leverage a normal distribution centered around 0 and using a standard deviation ($\sigma$) computed by:

XavierUniform

Uniform Xavier initializers use a uniform distribution to initialize the weights of a neural network within a given range.

Enums

InitError

Traits

Init

A trait for creating custom initialization routines for models or other entities.

InitInplace

This trait enables models to implement custom, in-place initialization methods.

Initialize

This trait provides the base methods required for initializing tensors with random values. The trait is similar to the RandomExt trait provided by the ndarray_rand crate, however, it is designed to be more generic, extensible, and optimized for neural network initialization routines. Initialize is implemented for ArrayBase as well as ParamsBase allowing you to randomly initialize new tensors and parameters.

Functions

randc

Generate a random array of complex numbers with real and imaginary parts in the range [0, 1)

stdnorm

Given a shape, generate a random array using the StandardNormal distribution

stdnorm_from_seed

uniform_from_seed

Creates a random array from a uniform distribution using a given key