qcd_ml.qcd

class qcd_ml.qcd.dirac.dirac_wilson(U, mass_parameter)[source]

Dirac Wilson operator. See arXiv:2302.05419.

class qcd_ml.qcd.dirac.dirac_wilson_clover(U, mass_parameter, csw)[source]

Dirac Wilson operator with clover term improvement.

See arXiv:2302.05419.

This module provides smearing of gauge links. Currenltly the following smearing algorithms are implemented:

class qcd_ml.qcd.gauge.smear.stout(rho: tensor)[source]

This class is used to construct the stout smearing operation. See [1]_ and [2]_ for details of the smearing algorithm.

This class is an abstract representation of the smearing operation with only the parameter ..math:rho being specified. The actual smearing operation is performed as such:

smearer = stout(rho)

for i in range(n_smearing_steps):
    U = smearer(U)

Classmethods are provided for common ..math:rho matrices:

  • constant_rho: ..math:rho is a constant matrix.

  • spatial_only: ..math:rho is a constant matrix

    with the temporal components set to zero.

QCD observables that are computed on the gauge field.

qcd_ml.qcd.gauge.observables.plaquette_field(U, _gpt_compat=False)[source]

Plaquette field of a gauge field. See [1]_ [2]_.

If _gpt_compat=True, the field is rescaled to match gpt’s conventions.

qcd_ml.qcd.gauge.observables.topological_charge_density_clover(U, _gpt_compat=False)[source]

The topological charge density field q(n) [1]_ [2]_ using the clover field strength [3]_.

qcd_ml.qcd.gauge.observables.topological_charge_density_plaquette(U)[source]

The topological charge density field q(n) using the plaquette field strength.

q(n) = \frac{1}{32\pi} \epsilon_{\mu\nu\rho\sigma} \mbox{Tr}\left(\frac{U_{\mu\nu}(x) - U_{\mu\nu}^\dagger(x)}{2i}\frac{U_{\rho\sigma}(x) - U_{\rho\sigma}^\dagger(x)}{2i}\right)

See 10.1103/PhysRevLett.128.032003.