Struct NeuralNetworkTopology

pub struct NeuralNetworkTopology<const I: usize, const O: usize> {
    pub input_layer: [Arc<RwLock<NeuronTopology>>; I],
    pub hidden_layers: Vec<Arc<RwLock<NeuronTopology>>>,
    pub output_layer: [Arc<RwLock<NeuronTopology>>; O],
    pub mutation_rate: f32,
    pub mutation_passes: usize,
}

A stateless neural network topology. This is the struct you want to use in your agent’s inheritance. See NeuralNetwork::from for how to convert this to a runnable neural network.

Fields

input_layer: [Arc<RwLock<NeuronTopology>>; I]

The input layer of the neural network. Uses a fixed length of I.

hidden_layers: Vec<Arc<RwLock<NeuronTopology>>>

The hidden layers of the neural network. Because neurons have a flexible connection system, all of them exist in the same flat vector.

output_layer: [Arc<RwLock<NeuronTopology>>; O]

The output layer of the neural netowrk. Uses a fixed length of O.

mutation_rate: f32

The mutation rate used in NeuralNetworkTopology::mutate after crossover/division.

mutation_passes: usize

The number of mutation passes (and thus, maximum number of possible mutations that can occur for each entity in the generation).

Implementations

impl<const I: usize, const O: usize> NeuralNetworkTopology<I, O>

pub fn new( mutation_rate: f32, mutation_passes: usize, rng: &mut impl Rng, ) -> Self

Creates a new NeuralNetworkTopology.

pub fn add_connection( &mut self, from: NeuronLocation, to: NeuronLocation, weight: f32, ) -> bool

Creates a new connection between the neurons. If the connection is cyclic, it does not add a connection and returns false. Otherwise, it returns true.

pub fn get_neuron(&self, loc: NeuronLocation) -> Arc<RwLock<NeuronTopology>>

Gets a neuron pointer from a NeuronLocation. You shouldn’t ever need to directly call this unless you are doing complex custom mutations.

pub fn rand_neuron( &self, rng: &mut impl Rng, ) -> (Arc<RwLock<NeuronTopology>>, NeuronLocation)

Gets a random neuron and its location.

Trait Implementations

impl<const I: usize, const O: usize> Clone for NeuralNetworkTopology<I, O>

fn clone(&self) -> Self

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<const I: usize, const O: usize> Debug for NeuralNetworkTopology<I, O>

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

Formats the value using the given formatter.

impl<const I: usize, const O: usize> DivisionReproduction for NeuralNetworkTopology<I, O>

fn divide(&self, rng: &mut impl Rng) -> Self

Create a new child with mutation. Similar to RandomlyMutable::mutate, but returns a new instance instead of modifying the original. If it is simply returning a cloned and mutated version, consider using a constant mutation rate.

impl<const I: usize, const O: usize> From<&NeuralNetworkTopology<I, O>> for NNTSerde<I, O>

Available on crate feature serde only.

fn from(value: &NeuralNetworkTopology<I, O>) -> Self

Converts to this type from the input type.

impl<const I: usize, const O: usize> From<&NeuralNetworkTopology<I, O>> for NeuralNetwork<I, O>

fn from(value: &NeuralNetworkTopology<I, O>) -> Self

Converts to this type from the input type.

impl<const I: usize, const O: usize> From<NNTSerde<I, O>> for NeuralNetworkTopology<I, O>

fn from(value: NNTSerde<I, O>) -> Self

Converts to this type from the input type.

impl<const I: usize, const O: usize> PartialEq for NeuralNetworkTopology<I, O>

fn eq(&self, other: &Self) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<const I: usize, const O: usize> RandomlyMutable for NeuralNetworkTopology<I, O>

fn mutate(&mut self, rate: f32, rng: &mut impl Rng)

Mutate the genome with a given mutation rate (0..1)