Struct rusty_machine::learning::optim::grad_desc::GradientDesc [−] [src]

pub struct GradientDesc {
    // some fields omitted
}

Methods

impl GradientDesc
[src]

fn new() -> GradientDesc

Constructs a new gradient descent algorithm.

Currently defaults to 1000 iterations and step size of 0.3. More control coming soon!

Trait Implementations

impl<M: Optimizable> OptimAlgorithm<M> for GradientDesc
[src]

fn optimize(&self, model: &M, start: &[f64], data: &M::Data, outputs: &M::Target) -> Vec<f64>

Return the optimized parameter using gradient optimization. Read more