Skip to content

GlorotUniform — Glorot Uniform Weight Initializer

Samples weights from a uniform distribution scaled by sqrt(6 / (fan_in + fan_out)).


Overview

GlorotUniform (Xavier Uniform) balances forward and backward variance over input and output dimensions. Scaled to sqrt(6 / (fan_in + fan_out)) for uniform sampling, ideal for networks using tanh/sigmoid activations.


Key Features

  • Balanced uniform scaling: sqrt(6 / (fan_in + fan_out))
  • Suitable for tanh/sigmoid-style networks
  • Reproducible via seed
  • Returns np.float64 array by default

Parameters

Parameter Type Default Description
seed int | None None Random seed for reproducibility
dtype np.dtype np.float64 Output data type

Methods

__call__(shape)

Return Glorot-scaled uniform samples with validated dimensions.


Example Usage

from machinegnostics.magnet import GlorotUniform
init = GlorotUniform(seed=42)
W = init((10, 20))
print(W.shape)  # (10, 20)

Author: Nirmal Parmar