GWR flow with convergence studies

This script shows how to compute the G0W0 corrections in silicon. More specifically, we build a flow to analyze the convergence of the QP corrections wrt to the number of bands in the self-energy. More complicated convergence studies can be implemented on the basis of this example.

import os
import sys
import abipy.data as data
import abipy.abilab as abilab

from abipy import flowtk
from abipy.flowtk.gwr_works import DirectDiagoWork, GWRSigmaConvWork


def build_flow(options):
    # Working directory (default is the name of the script with '.py' removed and "run_" replaced by "flow_")
    if not options.workdir:
        options.workdir = os.path.basename(sys.argv[0]).replace(".py", "").replace("run_","flow_")

    # IMPORTANT: Note stringent table to have semi-core states.
    from abipy.flowtk.psrepos import get_oncvpsp_pseudos
    pseudos = get_oncvpsp_pseudos(xc_name="PBE", version="0.4",
                                  relativity_type="SR", accuracy="stringent")

    scf_input = abilab.AbinitInput(structure=data.cif_file("si.cif"), pseudos=pseudos)

    num_ele = scf_input.num_valence_electrons

    # Global variables.
    scf_input.set_vars(
        ecut=6,
        nband=num_ele // 2,
        tolvrs=1e-8,
        paral_kgb=1,
    )

    # IMPORTANT: k-grid for GWR must be Gamma-centered.
    scf_input.set_kmesh(ngkpt=[2, 2, 2], shiftk=[0.0, 0.0, 0.0])

    flow = flowtk.Flow(workdir=options.workdir)

    # GS-SCF run to get the DEN, followed by direct diago to obtain green_nband bands.
    green_nband = -1  # -1 this means full diago
    diago_work = DirectDiagoWork.from_scf_input(scf_input, green_nband)
    flow.register_work(diago_work)

    # Build template for GWR.
    gwr_template = scf_input.make_gwr_qprange_input(gwr_ntau=6, nband=8, ecuteps=4, ecutwfn=2)

    # Two possibilities:
    # 1) To change the value of one variable, use:

    varname_values = ("nband", [8, 12, 14])

    # 2) To take the Cartesian product of two or more variables use e.g.:
    #
    #varname_values = [
    #   ("nband", [50, 100]),
    #   ("ecuteps", [2, 4]),
    #]

    # Can also use strings with path to files for den_node and wfk_node
    # so that one does not need to recompute these files.

    gwr_work = GWRSigmaConvWork.from_varname_values(
            varname_values,
            gwr_template,
            den_node=diago_work.scf_task,
            wfk_node=diago_work.diago_task,
    )
    flow.register_work(gwr_work)

    return flow


# This block generates the thumbnails in the AbiPy gallery.
# You can safely REMOVE this part if you are using this script for production runs.
if os.getenv("READTHEDOCS", False):
    __name__ = None
    import tempfile
    options = flowtk.build_flow_main_parser().parse_args(["-w", tempfile.mkdtemp()])
    build_flow(options).graphviz_imshow()


@flowtk.flow_main
def main(options):
    return build_flow(options)


if __name__ == "__main__":
    sys.exit(main())