Struct DenseLayer 

pub struct DenseLayer {
    pub name: String,
    pub weights: Tensor,
    pub bias: Tensor,
    pub activation: String,
}

A single dense (fully connected) layer with weights and biases.

Fields

name: String

Layer name.

weights: Tensor

Weight matrix: shape \[out_features, in_features\].

bias: Tensor

Bias vector: shape [out_features].

activation: String

Activation function name (e.g. "relu", "sigmoid", "tanh", "linear").

Implementations

impl DenseLayer

pub fn new( name: impl Into<String>, in_features: usize, out_features: usize, activation: impl Into<String>, ) -> Self

Construct a zero-initialised dense layer.

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

Forward pass: output\[i\] = sum_j(w\[i,j\] * input\[j\]) + bias\[i\], then activation.

pub fn param_count(&self) -> usize

Number of trainable parameters.

pub fn to_bytes(&self) -> Vec<u8>

Serialise to bytes (name length, name utf8, weights bytes, bias bytes).

Trait Implementations

impl Clone for DenseLayer

fn clone(&self) -> DenseLayer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DenseLayer

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

Formats the value using the given formatter.