neuralqx.lqx.qr.core.single_vertex_qrlg module

class LqxSVQRLG(hilbert, graph, gauge_group, *, computational=True, jax=True, immirzi=1.0, spacetime_dimensions=4, model_name='LqxSVQRLGModel')

Bases: AbstractLqxSVQRLGModel

property constraint

This property should return the constraint of the model to be considered in the optimisation process

property euclidean_constraint

Subclasses should implement the Euclidean constraint according to the single-vertex model of QRLG.

Returns:

property lorentzian_constraint

Implements the Lorentzian constraint according to the single-vertex model of QRLG.

Returns:

creation(edge, n, *, computational=True, jax=True)

Subclasses should implement a creation operator acting on the specified edge according to the single-vertex model of QRLG.

Parameters:
Returns:

annihilation(edge, n, *, computational=True, jax=True)

Subclasses should implement an annihilation operator acting on the specified edge according to the single-vertex model of QRLG.

Parameters:
Returns:

s(edge, no_i=False, *, computational=True, jax=True)

The symmetric holonomy operator s which increments or decrements the quantum number of the specified edge symmetrically by a value of 1.

c(edge, *, computational=True, jax=True)

The symmetric holonomy operator c which increments or decrements the quantum number of the specified edge symmetrically by a value of 1.

E(edge, power=1.0, *, computational=True, jax=True)

The flux operator which acts as a number operator but can additionally output eigenvalues raised to some power.

Parameters:
  • edge (int) – the site the flux operator should act on

  • power (float) – the power the eigenvalues should be raised to

E_inv(edge, power=1.0, *, computational=True, jax=True)

The flux operator which acts as an ‘inverse’ number operator in the sense that it outputs 1/spin instead of spin as the eigenvalue. Additionally, it can output eigenvalues raised to some power.

Parameters:
  • edge (int) – the site the inverted flux operator should act on

  • power (float) – the power the eigenvalues should be raised to