Struct Conv2DLayer 

pub struct Conv2DLayer {
    pub in_channels: usize,
    pub out_channels: usize,
    pub kernel_size: usize,
    pub stride: usize,
    pub padding: usize,
    pub kernels: Vec<f64>,
}

Stochastic 2D convolutional layer.

Kernels stored as probabilities in [0,1]. Forward pass uses P(A∧B) = P(A)·P(B) for unipolar stochastic multiplication.

Fields

in_channels: usize

out_channels: usize

kernel_size: usize

stride: usize

padding: usize

kernels: Vec<f64>

Flat kernels: [out_ch, in_ch, k, k]

Implementations

impl Conv2DLayer

pub fn new( in_channels: usize, out_channels: usize, kernel_size: usize, stride: usize, padding: usize, seed: u64, ) -> Self

pub fn forward( &self, input: &[f64], h: usize, w: usize, ) -> (Vec<f64>, usize, usize)

Forward pass: input shape [C_in, H, W] (flat), returns [C_out, H_out, W_out] (flat).

Trait Implementations

impl Clone for Conv2DLayer

fn clone(&self) -> Conv2DLayer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Conv2DLayer

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

Formats the value using the given formatter.