Struct neural_lib::Network

pub struct Network {
    pub inputs: Vec<f64>,
    pub output: Layer,
    /* private fields */
}

Define a Neural Network

Fields

inputs: Vec<f64>

output: Layer

Implementations

impl Network

pub fn new( nb_lay: usize, lays_params: Vec<Vec<(f64, f64)>>, lays_fn: Vec<Activate> ) -> Network

pub fn default(nb_lay: usize, nb_n: u32, lays_fn: Vec<Activate>) -> Network

pub fn prediction(&mut self) -> Vec<f64>

Launching a prediction with values -> This function can be call manually or by the training session

pub fn train( &mut self, inputs: &Vec<Vec<f64>>, targets: &Vec<f64>, learning_rate: f64, nb: usize )

Initiate a training session You can specifie the inputs and the number of repetitions from the nb variable

pub fn output(&self, n_file: &str) -> Result<(), Error>

Generate a file which contains the parameters of your neurons and layers