mlx.nn.init.he_uniform

he_uniform(dtype: Dtype = mlx.core.float32) → Callable[[array, str, float], array]

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) units according to:

\[\sigma = \gamma \sqrt{\frac{3.0}{\text{fan}}}\]

where \(\text{fan}\) is either the number of input units when the mode is "fan_in" or output units when the mode is "fan_out".

For more details see the original reference: Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

Parameters:

dtype (Dtype, optional) – The data type of the array. Default: float32.

Returns:

An initializer that returns an array with the same shape as the input, filled with samples from the He uniform distribution.

Return type:

Callable[[array, str, float], array]

Example

>>> init_fn = nn.init.he_uniform()
>>> init_fn(mx.zeros((2, 2)))  # uses fan_in
array([[0.0300242, -0.0184009],
       [0.793615, 0.666329]], dtype=float32)
>>> init_fn(mx.zeros((2, 2)), mode="fan_out", gain=5)
array([[-1.64331, -2.16506],
       [1.08619, 5.79854]], dtype=float32)