Struct StandardOptNetLayer 

pub struct StandardOptNetLayer {
    pub qp: DifferentiableQP,
    pub config: DiffQPConfig,
}

A standard OptNet layer wrapping a differentiable QP.

Fields

qp: DifferentiableQP

The underlying differentiable QP.

config: DiffQPConfig

Configuration for forward/backward passes.

Implementations

impl StandardOptNetLayer

pub fn new(qp: DifferentiableQP, config: DiffQPConfig) -> Self

Create a new OptNet layer from a differentiable QP and config.

pub fn forward_batch( qps: &[DifferentiableQP], config: &DiffQPConfig, ) -> OptimizeResult<Vec<DiffQPResult>>

Solve a batch of QPs sharing the same config.

Trait Implementations

impl Clone for StandardOptNetLayer

fn clone(&self) -> StandardOptNetLayer

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for StandardOptNetLayer

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

Formats the value using the given formatter.

impl OptNetLayer for StandardOptNetLayer

type ForwardResult = DiffQPResult

The result type returned by the forward pass.

fn forward(&self) -> OptimizeResult<DiffQPResult>

Solve the optimization problem (forward pass).

fn backward( &self, result: &DiffQPResult, dl_dx: &[f64], ) -> OptimizeResult<ImplicitGradient>

Compute parameter gradients (backward pass).