qlat.vector_utils — Legacy Vector Utilities

Source: qlat/qlat/vector_utils.pyx

Note: Update this document when updating the source file.

Outline

  • diff_gauge(g0, g1) — difference between two gauge fields.

  • diff_prop(p0, p1) — difference between two propagators.

  • load_gwu_link(g0, path) — read a gauge field in gwu link format.

  • save_gwu_prop(prop, path) — write a propagator in gwu format.

  • load_gwu_prop(prop, path) — read a propagator in gwu format.

  • save_gwu_noiP(prop, path) — write the noiP projection of a propagator.

  • load_gwu_noiP(prop, path) — read the noiP projection of a propagator.

  • load_qlat_link(g0, path) — read a gauge field in qlat format.

  • save_qlat_link(g0, path) — write a gauge field in qlat format.

  • random_point_src(prop, seed) — random point source.

  • make_point_prop(prop, sp) — point propagator at sp.

  • make_volume_src(prop, seed, mix_color, mix_spin, tini) — volume source.

  • local_sequential_source(res, src, tseq, gammai) — local sequential source.

  • meson_corr(p0, p1, filename, g0, g1, tini, invmode, info, shift_end, mom) — meson correlator.

  • corr_dat_create(filename, key_T, dimN, info) — create a .dat correlator file.

  • corr_dat_info(filename, info) — append info to a .dat correlator file.

  • prop4d_conj(prop, rotate) — complex conjugate a propagator.

  • prop4d_src_gamma(prop, g0, Conj) — multiply a source gamma matrix.

  • prop4d_sink_gamma(prop, g0, Conj) — multiply a sink gamma matrix.

Overview

This module exposes the legacy vector-utility routines that used to live in the hand-written qlat/cqlat/vector_utils.cpp extension. The functions operate on the standard Prop and GaugeField types and delegate directly to the C++ templates in qlat/qlat/include/qlat/vector_utils/.

It is reached through the qlat.c re-export module, i.e. the functions are normally called as q.c.<name>(...) (aliased as qc in the examples).

API Reference

diff_gauge(g0, g1)

Return the maximum difference between the gauge fields g0 and g1.

Parameter

Type

Description

g0

GaugeField

First gauge field.

g1

GaugeField

Second gauge field.

Returns: float — the difference measure.

diff_prop(p0, p1)

Compare two propagators p0 and p1 (no return value; the C++ routine reports the difference).

Parameter

Type

Description

p0

Prop

First propagator.

p1

Prop

Second propagator.

save_gwu_prop(prop, path) / load_gwu_prop(prop, path)

Write / read a propagator in the gwu format.

save_gwu_noiP(prop, path) / load_gwu_noiP(prop, path)

Write / read the noiP (diagonal noise) projection of a propagator.

random_point_src(prop, seed=0)

Fill prop with a random point source.

make_point_prop(prop, sp=None)

Fill prop with the point propagator located at sp (a coordinate list or Coordinate; defaults to the origin).

make_volume_src(prop, seed=0, mix_color=0, mix_spin=0, tini=-1)

Fill prop with a volume source. seed == -1 produces an all-ones wall source; mix_color / mix_spin select the colour/spin mixing; tini restricts the source to time slices >= tini.

local_sequential_source(res, src, tseq, gammai=-1)

Build the local sequential source res from src for the time sequence tseq and gamma index gammai.

meson_corr(p0, p1, filename, g0, g1, tini=0, invmode=1, info="NONE", shift_end=1, mom=None)

Compute the meson correlator between p0 and p1 and write it to filename.

Parameter

Type

Description

p0, p1

Prop

Propagators.

filename

str

Output file name.

g0, g1

int

Gamma matrix indices.

tini

int

Initial time slice.

invmode

int

Inversion mode.

info

str

Free-form info string.

shift_end

int

End-shift flag.

mom

coordinate

Momentum (defaults to zero).

corr_dat_create(filename, key_T, dimN, info="NONE")

Create a .dat correlator file with the given key/dimension metadata.

corr_dat_info(filename, info="NONE")

Append the info string to an existing .dat correlator file.

prop4d_conj(prop, rotate=1)

Replace prop by its complex conjugate (CPS gamma convention).

prop4d_src_gamma(prop, g0=0, Conj=0)

Multiply prop by the source gamma matrix g0 (optionally conjugated).

prop4d_sink_gamma(prop, g0=0, Conj=0)

Multiply prop by the sink gamma matrix g0 (optionally conjugated).

Examples

import qlat as q
import qlat.c as qc

q.begin_with_mpi([[1, 1, 1, 4]])

geo = q.Geometry(q.Coordinate([4, 4, 4, 8]))

p0 = q.Prop(geo)
p0.set_rand(q.RngState("p0"))
p1 = q.Prop(geo)
p1 @= p0

qc.save_qlat_link(q.GaugeField(geo), "links.qlat")
qc.prop4d_sink_gamma(p1, 0, 0)
qc.diff_prop(p0, p1)

q.end_with_mpi()