Struct NeuralLayer 

pub struct NeuralLayer {
    pub weights: Vec<f64>,
    pub biases: Vec<f64>,
    pub layer_type: LayerType,
    pub input_size: usize,
    pub output_size: usize,
}

A single layer in a neural network, carrying weights, biases and a type.

Fields

weights: Vec<f64>

Flattened weight matrix (row-major: [out, in]).

biases: Vec<f64>

Bias vector (length = number of output neurons).

layer_type: LayerType

Computational type of this layer.

input_size: usize

Number of input neurons / features.

output_size: usize

Number of output neurons.

Implementations

impl NeuralLayer

pub fn new(input_size: usize, output_size: usize, layer_type: LayerType) -> Self

Create a new layer with given dimensions and type.

Weights and biases are zero-initialised; call the builder helpers to set custom values.

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

Execute the forward pass of this layer on input.

For activation layers (ReLU, Sigmoid, Tanh, Softmax) the weights/biases are ignored and the input is transformed element-wise.

Trait Implementations

impl Clone for NeuralLayer

fn clone(&self) -> NeuralLayer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for NeuralLayer

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

Formats the value using the given formatter.