Enum Init

pub enum Init {
    Const(f64),
    Randn {
        mean: f64,
        stdev: f64,
    },
    Uniform {
        lo: f64,
        up: f64,
    },
    Kaiming {
        dist: NormalOrUniform,
        fan: FanInOut,
        non_linearity: NonLinearity,
    },
}

Variable initializations.

Variants

Const(f64)

Constant value.

Randn

Random normal with some mean and standard deviation.

Fields

mean: f64

stdev: f64

Uniform

Uniform initialization between some lower and upper bounds.

Fields

lo: f64

up: f64

Kaiming

Kaiming uniform initialization. See “Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification” He, K. et al. (2015). This uses a uniform distribution.

Fields

dist: NormalOrUniform

fan: FanInOut

non_linearity: NonLinearity

Implementations

impl Init

pub fn var<S: Into<Shape>>( &self, s: S, dtype: DType, device: &Device, ) -> Result<Var>

Creates a new tensor with the specified shape, device, and initialization.

Trait Implementations

impl Clone for Init

fn clone(&self) -> Init

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Init

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for Init

fn default() -> Self

Returns the “default value” for a type.

impl Copy for Init