Struct TQParticleConservingLayer 

pub struct TQParticleConservingLayer { /* private fields */ }

Particle conserving layer using Givens rotations

This layer is specifically designed for quantum chemistry applications where particle number conservation is critical. It uses Givens rotations which naturally preserve the number of particles.

The layer can be used as an excitation preserving ansatz for molecular simulations in variational quantum algorithms.

Implementations

impl TQParticleConservingLayer

pub fn new(n_wires: usize, n_blocks: usize, pattern: GivensPattern) -> Self

Create a new particle conserving layer

pub fn adjacent(n_wires: usize, n_blocks: usize) -> Self

Create with adjacent pattern (default)

pub fn staircase(n_wires: usize, n_blocks: usize) -> Self

Create with staircase pattern

pub fn bricklayer(n_wires: usize, n_blocks: usize) -> Self

Create with bricklayer pattern

Trait Implementations

impl TQModule for TQParticleConservingLayer

fn forward(&mut self, qdev: &mut TQDevice) -> Result<()>

Execute the forward pass on the quantum device

fn parameters(&self) -> Vec<TQParameter>

Get all trainable parameters

fn n_wires(&self) -> Option<usize>

Get number of wires this module operates on

fn set_n_wires(&mut self, n_wires: usize)

Set number of wires

fn is_static_mode(&self) -> bool

Check if module is in static mode

fn static_on(&mut self)

Enable static mode for graph optimization

fn static_off(&mut self)

Disable static mode

fn name(&self) -> &str

Module name for debugging

fn zero_grad(&mut self)

Zero gradients of all parameters

fn forward_with_input( &mut self, qdev: &mut TQDevice, _x: Option<&Array2<f64>>, ) -> Result<()>

Execute forward pass with optional input data (for encoders)

fn get_unitary(&self) -> Option<Array2<CType>>

Get the unitary matrix representation (if available)

fn train(&mut self, _mode: bool)

Set training mode

fn training(&self) -> bool

Check if in training mode