###############################################################################
# Copyright (c), Forschungszentrum Jülich GmbH, IAS-1/PGI-1, Germany. #
# All rights reserved. #
# This file is part of the Masci-tools package. #
# (Material science tools) #
# #
# The code is hosted on GitHub at https://github.com/judftteam/masci-tools. #
# For further information on the license, see the LICENSE.txt file. #
# For further information please visit http://judft.de/. #
# #
###############################################################################
"""
Tools for the impurity calculation plugin and its workflows
"""
import numpy as np
import traceback
from masci_tools.io.common_functions import search_string, get_outfile_txt, get_version_info, convert_to_pystd
from masci_tools.io.parsers.kkrparser_functions import get_rms, find_warnings, get_charges_per_atom, get_core_states
from masci_tools.io.common_functions import get_Ry2eV
__copyright__ = ('Copyright (c), 2018, Forschungszentrum Jülich GmbH,'
'IAS-1/PGI-1, Germany. All rights reserved.')
__license__ = 'MIT license, see LICENSE.txt file'
__version__ = '0.8.1'
__contributors__ = ('Philipp Rüßmann', 'Fabian Bertoldo', 'David Antognini Silva')
####################################################################################
[docs]class KkrimpParserFunctions:
"""
Class of parser functions for KKRimp calculation
:usage: success, msg_list, out_dict = parse_kkrimp_outputfile().parse_kkrimp_outputfile(out_dict, files)
"""
### some helper functions ###
def _get_econt_info(self, out_log):
"""
extract energy contour information from out_log file
:param out_log: file that is parsed
:returns: econt (dict), dictionary containing the energy contour info
:note: econt contains the following keys
* 'emin', bottom of energy contour
* 'Nepts', number of points in energy contour
* 'epts', list of complex valued energy points
* 'weights', list of complex valued weights for energy integration
"""
tmptxt = get_outfile_txt(out_log)
econt = {}
itmp = search_string('[read_energy] number of energy points', tmptxt)
if itmp >= 0:
econt['Nepts'] = int(tmptxt.pop(itmp).split()[-1])
itmp = search_string('energies and weights are:', tmptxt)
if itmp >= 0:
tmp = []
for ie in range(econt['Nepts']):
tmpline = tmptxt[itmp + 4 + ie].split()[1:]
tmp.append([float(tmpline[0]), float(tmpline[1]), float(tmpline[2]), float(tmpline[3])])
tmp = np.array(tmp)
econt['epts'] = tmp[:, :2]
econt['weights'] = tmp[:, 2:]
econt['emin'] = tmp[0, 0]
return econt
def _get_scfinfo(self, file):
"""
extract scf infos (nunmber of iterations, max number of iterations, mixing info) from file
:param file:
:returns: niter (int), nitermax (int), converged (bool), nmax_reached (bool), mixinfo (dict)
:note: mixinfo contains information on mixing scheme and mixing factor used in the calculation
"""
tmptxt = get_outfile_txt(file)
# get rms and number of iterations
itmp, niter, rms = 0, -1, -1
while itmp >= 0:
itmp = search_string('average rms-error', tmptxt)
if itmp >= 0:
tmp = tmptxt.pop(itmp).replace('D', 'E').split()
niter = int(tmp[1])
rms = float(tmp[-1])
# get max number of scf steps
itmp = search_string('SCFSTEPS', tmptxt)
if itmp >= 0:
nitermax = int(tmptxt.pop(itmp).split()[-1])
# get qbound
itmp = search_string('QBOUND', tmptxt)
if itmp >= 0:
qbound = float(tmptxt.pop(itmp).split()[-1])
# get imix
itmp = search_string('IMIX', tmptxt)
if itmp >= 0:
imix = int(tmptxt.pop(itmp).split()[-1])
# get mixfac
itmp = search_string('MIXFAC', tmptxt)
if itmp >= 0:
mixfac = float(tmptxt.pop(itmp).split()[-1])
# get fcm
itmp = search_string('FCM', tmptxt)
if itmp >= 0:
fcm = float(tmptxt.pop(itmp).split()[-1])
# set mixinfo
mixinfo = [imix, mixfac, qbound, fcm]
# set converged and nmax_reached logicals
converged, nmax_reached = False, False
if nitermax == niter:
nmax_reached = True
if rms < qbound:
converged = True
# return values
return niter, nitermax, converged, nmax_reached, mixinfo
def _get_newsosol(self, file):
"""
Check if spin orbit coupling solver is used
:param file: absolute path to out_log.000.txt of KKRimp calculation
:returns: True(False) if SOC solver is (not) used
"""
tmptxt = get_outfile_txt(file)
itmp = search_string('Spin orbit coupling used?', tmptxt)
itmp = int(tmptxt.pop(itmp).split()[-1])
newsosol = itmp == 1
return newsosol
def _get_natom(self, file):
"""
Extract number of atoms in impurity cluster
:param file: file that is parsed to find number of atoms
:returns: natom (int), number of atoms in impurity cluster
"""
tmptxt = get_outfile_txt(file)
itmp = search_string('NATOM is', tmptxt)
natom = int(tmptxt.pop(itmp).split()[-1])
return natom
def _get_magtot(self, file, natom, debug=False):
"""
Extract total magnetic moment of all atoms in imp. cluster,
magnetic moment for each atom in the cluster and magn. moment
for all atoms and all iterations of the calculation
:param file: file that is parsed to find magnetic moments
:param natom: number of atoms in the cluster
:returns: magn. moment for all atoms in the cluster for the last iteration (saved in z-comp. of 3d vector)
magn. moment for all atoms in the cluster for all iterations (saved in z-comp. of 3d vector)
total magnetic moments of all atoms for last iteration
"""
tmptxt = get_outfile_txt(file)
itmp = 0
spinmom_all = []
while itmp >= 0:
itmp = search_string('spin magnetic moment =', tmptxt)
if itmp >= 0:
spinmom_all.append(float(tmptxt.pop(itmp).split()[-1]))
# if no spin
spinmom = spinmom_all[len(spinmom_all) - natom:]
if len(spinmom) > 0: # this means we found something
spinmom_vec = np.array([[0, 0, spinmom[0]]])
spinmom_vec_all = np.array([[0, 0, spinmom_all[0]]])
for i in range(1, natom):
spinmom_vec = np.append(spinmom_vec, [[0, 0, spinmom[i]]], axis=0)
for i in range(1, len(spinmom_all)):
spinmom_vec_all = np.append(spinmom_vec_all, [[0, 0, spinmom_all[i]]], axis=0)
magtot = sum(spinmom)
else:
# otherwise return empty lists
spinmom_vec, spinmom_vec_all, magtot = [], [], []
return spinmom_vec, spinmom_vec_all, magtot
def _extract_timings(self, outfile):
"""
Extract timings for the different parts in the KKRimp code
:param outfile: timing file of the KKRimp run
:returns: res (dict) timings in seconds, averaged over iterations
"""
tmptxt = get_outfile_txt(outfile)
search_keys = [
'time until scf starts', 'vpot->tmat', 'gref->gmat', 'gonsite->density', 'energyloop', 'Iteration number',
'Total running time'
]
res = {}
for isearch in search_keys:
tmpval = []
itmp = 0
while itmp >= 0:
itmp = search_string(isearch, tmptxt)
if itmp >= 0:
tmpval.append(float(tmptxt.pop(itmp).split()[-1]))
if len(tmpval) > 0:
res[isearch] = tmpval
# average over iterations
niter = len(res.get(search_keys[-2], []))
if niter > 0:
for key in search_keys[1:6]:
if key in res:
res[key] = sum(res[key]) / niter
for key in [search_keys[0], search_keys[-1]]:
if key in res:
res[key] = res[key][0]
return res
def _get_nspin(self, file):
"""
Extract nspin from file
:param file: file that is parsed
:returns: 1 if calculation is paramagnetic, 2 otherwise
"""
tmptxt = get_outfile_txt(file)
itmp = search_string('NSPIN', tmptxt)
nspin = int(tmptxt.pop(itmp).split()[-1])
return nspin
def _get_spinmom_per_atom(self, file, natom):
"""
Extract spin moment for all atoms
:param file: file that is parsed
:param natom: number of atoms in impurity cluster
:returns: spinmom_at (array of spin moments for all atoms and the last iteration),
spinmom_at_all (array of spin moments for all atoms and iterations),
spinmom_at_tot (total spinmoment for the last iteration)
"""
from math import sqrt #pylint: disable=no-name-in-module
lines = get_outfile_txt(file)
startline = len(lines) - natom
spinmom_at = np.array([lines[startline].split()])
spinmom_at_all = np.array([lines[1].split()])
for i in range(1, natom):
spinmom_at = np.append(spinmom_at, [lines[startline + i].split()], axis=0)
for j in range(2, len(lines)):
spinmom_at_all = np.append(spinmom_at_all, [lines[j].split()], axis=0)
spinmom_at_tot = 0
for i in range(0, natom):
spinmom_at_tot += sqrt(float(spinmom_at[i][0])**2 + float(spinmom_at[i][1])**2 + float(spinmom_at[i][2])**2)
# make sure the values are converted from string to float
spinmom_at = np.array(spinmom_at, dtype=float)
spinmom_at_all = np.array(spinmom_at_all, dtype=float)
return spinmom_at, spinmom_at_all, spinmom_at_tot
def _get_orbmom_per_atom(self, file, natom):
"""
Extract orbital moment for all atoms (orbmom_at: all atoms in last iteration,
orbmom_at_all: all atoms in all iterations). For each atom there are six values:
first -> x-component real part, second -> x-component imaginary part,
third -> y-component real part, ... sixth -> z-component imaginary part.
:param file: file that is parsed
:param natom: number of atoms in impurity cluster
:returns: orbmom_at (list), orbital moments for all atoms
"""
lines = get_outfile_txt(file)
startline = len(lines) - natom
orbmom_at = []
for i in range(natom):
tmp = lines[startline + i].split()
orbmom_at.append([tmp[1], tmp[3], tmp[5]]) # [1,3,5] needed since full complex number is written
orbmom_at = np.array(orbmom_at, dtype=float) # convert to float array
# do the same for all iterations
orbmom_at_all = []
for i in range(1, len(lines)):
tmp = lines[i].split()
orbmom_at_all.append([tmp[1], tmp[3], tmp[5]])
orbmom_at_all = np.array(orbmom_at_all, dtype=float) # convert to float array
return orbmom_at, orbmom_at_all
def _get_EF_potfile(self, potfile):
"""
Extract EF value from potential file
:param potfile: file that is parsed
:returns: EF (float), value of the Fermi energy in Ry
"""
tmptxt = get_outfile_txt(potfile)
EF = float(tmptxt[3].split()[1])
return EF
def _get_Etot(self, file):
"""
Extract total energy file
:param file: file that is parsed
:returns: Etot (list), values of the total energy in Ry for all iterations
"""
tmptxt = get_outfile_txt(file)
itmp = 0
Etot = []
while itmp >= 0:
itmp = search_string('TOTAL ENERGY', tmptxt)
if itmp >= 0:
Etot.append(float(tmptxt.pop(itmp).split()[-1]))
return Etot
def _get_energies_atom(self, file1, file2, natom):
"""
Extract single particle and total energies in Ry for all atoms from file 1 and file 2
:param file1: file containing all single particle energies
:param file2: file containing all total energies
:returns: esp_at (list), etot_at (list)
"""
esp = np.loadtxt(file1)
etot = np.loadtxt(file2)
if natom > 1:
esp_at = esp[-natom:, 1]
etot_at = etot[-natom:, 1]
else:
esp_at = esp
etot_at = etot
return esp_at, etot_at
### end helper functions ###
[docs] def parse_kkrimp_outputfile(self, out_dict, file_dict, debug=False, ignore_nan=False):
"""
Main parser function for kkrimp, read information from files in file_dict and fills out_dict
:param out_dict: dictionary that is filled with parsed output of the KKRimp calculation
:param file_dict: dictionary of files that are parsed
:param debug: True/False to activate debug output
:param ignore_nan: bool replace NaN by zero with numpy's nan_to_num function
:returns: success (bool), msg_list(list of error/warning messages of parser), out_dict (filled dict of parsed output)
:note: file_dict should contain the following keys
- 'outfile', the std_out of the KKRimp calculation
- 'out_log', the out_log.000.txt file
- 'out_pot', the output potential
- 'out_enersp_at', the out_energysp_per_atom_eV file
- 'out_enertot_at', the out_energytotal_per_atom_eV file
- 'out_timing', the timing file
- 'kkrflex_llyfac', the file for the Lloyd factor
- 'kkrflex_angles', the nonco_angles file for the KKRimp calculation
- 'out_spinmoms', the output spin moments file
- 'out_orbmoms', the output orbital moments file
"""
msg_list = []
files = file_dict
try:
code_version, compile_options, serial_number = get_version_info(files['out_log'])
tmp_dict = {}
tmp_dict['code_version'] = code_version
tmp_dict['compile_options'] = compile_options
tmp_dict['calculation_serial_number'] = serial_number
out_dict['code_info_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: Version Info'
msg_list.append(msg)
if debug:
traceback.print_exc()
tmp_dict = {} # used to group convergence info (rms, rms per atom, charge neutrality)
# also initialize convegence_group where all info stored for all iterations is kept
out_dict['convergence_group'] = tmp_dict
try:
rms_charge, rms_ldau, rms_spin, result_atoms_last_charge, result_atoms_last_spin = get_rms(files['outfile'],
files['out_log'],
debug=debug,
is_imp_calc=True)
tmp_dict['rms'] = rms_charge[-1]
tmp_dict['rms_all_iterations'] = rms_charge
tmp_dict['rms_LDAU'] = rms_ldau[-1]
tmp_dict['rms_LDAU_all_iterations'] = rms_ldau
tmp_dict['rms_per_atom'] = result_atoms_last_charge
if len(rms_spin) > 0:
tmp_dict['rms_spin'] = rms_spin[-1]
else:
tmp_dict['rms_spin'] = None
tmp_dict['rms_spin_all_iterations'] = rms_spin
tmp_dict['rms_spin_per_atom'] = result_atoms_last_spin
tmp_dict['rms_unit'] = 'unitless'
out_dict['convergence_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: rms-error'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
nspin = self._get_nspin(files['out_log'])
natom = self._get_natom(files['out_log'])
newsosol = self._get_newsosol(files['out_log'])
out_dict['nspin'] = nspin
out_dict['number_of_atoms_in_unit_cell'] = natom
out_dict['use_newsosol'] = newsosol
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: nspin/natom'
msg_list.append(msg)
if debug:
traceback.print_exc()
tmp_dict = {} # used to group magnetism info (spin and orbital moments)
try:
result = self._get_magtot(files['out_log'], natom, debug=debug)
if len(result) > 0:
tmp_dict['total_spin_moment'] = result[-1]
out_dict['convergence_group']['total_spin_moment_all_iterations'] = result
tmp_dict['total_spin_moment_unit'] = 'mu_Bohr'
out_dict['magnetism_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: total magnetic moment'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
if nspin > 1 and newsosol:
#result, vec, angles = get_spinmom_per_atom(outfile, natom, nonco_out_file)
spinmom_atom, spinmom_atom_vec_all_iter, spin_tot_abs = self._get_spinmom_per_atom(
files['out_spinmoms'], natom)
if ignore_nan:
spinmom_atom = np.nan_to_num(spinmom_atom)
spinmom_atom_vec_all_iter = np.nan_to_num(spinmom_atom_vec_all_iter)
spin_tot_abs = np.nan_to_num(spin_tot_abs)
if len(result) > 0:
tmp_dict['total_abs_spin_moment'] = spin_tot_abs
tmp_dict['spin_moment_per_atom'] = spinmom_atom
out_dict['convergence_group']['spin_moment_per_atom_all_iterations'] = spinmom_atom_vec_all_iter
tmp_dict['spin_moment_unit'] = 'mu_Bohr'
out_dict['magnetism_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: spin moment per atom'
msg_list.append(msg)
if debug:
traceback.print_exc()
# add orbital moments to magnetis group in parser output
try:
if nspin > 1 and newsosol:
orbmom_atom, orbmom_atom_all_iter = self._get_orbmom_per_atom(files['out_orbmoms'], natom)
if len(result) > 0:
tmp_dict['total_orbital_moment'] = sum(orbmom_atom)
tmp_dict['orbital_moment_per_atom'] = orbmom_atom
out_dict['convergence_group']['orbital_moment_per_atom_all_iterations'] = orbmom_atom_all_iter
tmp_dict['orbital_moment_unit'] = 'mu_Bohr'
out_dict['magnetism_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: orbital moment'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
result = self._get_EF_potfile(files['out_pot'])
out_dict['fermi_energy'] = result
out_dict['fermi_energy_units'] = 'Ry'
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: EF'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
result = self._get_Etot(files['out_log'])
out_dict['energy'] = result[-1] * get_Ry2eV()
out_dict['energy_unit'] = 'eV'
out_dict['total_energy_Ry'] = result[-1]
out_dict['total_energy_Ry_unit'] = 'Rydberg'
out_dict['convergence_group']['total_energy_Ry_all_iterations'] = result
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: total energy'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
result = find_warnings(files['outfile'])
tmp_dict = {}
tmp_dict['number_of_warnings'] = len(result)
tmp_dict['warnings_list'] = result
out_dict['warnings_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: search for warnings'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
result = self._extract_timings(files['out_timing'])
out_dict['timings_group'] = result
out_dict['timings_unit'] = 'seconds'
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: timings'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
esp_at, etot_at = self._get_energies_atom(files['out_enersp_at'], files['out_enertot_at'], natom)
out_dict['single_particle_energies'] = esp_at * get_Ry2eV()
out_dict['single_particle_energies_unit'] = 'eV'
out_dict['total_energies_atom'] = etot_at * get_Ry2eV()
out_dict['total_energies_atom_unit'] = 'eV'
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: single particle energies'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
result_WS, result_tot, result_C = get_charges_per_atom(files['out_log'])
niter = len(out_dict['convergence_group']['rms_all_iterations'])
natyp = int(len(result_tot) / niter)
out_dict['total_charge_per_atom'] = result_WS[-natyp:]
out_dict['charge_core_states_per_atom'] = result_C[-natyp:]
# this check deals with the DOS case where output is slightly different
if len(result_WS) == len(result_C):
out_dict['charge_valence_states_per_atom'] = result_WS[-natyp:] - result_C[-natyp:]
out_dict['total_charge_per_atom_unit'] = 'electron charge'
out_dict['charge_core_states_per_atom_unit'] = 'electron charge'
out_dict['charge_valence_states_per_atom_unit'] = 'electron charge'
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: charges'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
econt = self._get_econt_info(files['out_log'])
tmp_dict = {}
tmp_dict['emin'] = econt.get('emin')
tmp_dict['emin_unit'] = 'Rydberg'
tmp_dict['number_of_energy_points'] = econt.get('Nepts')
tmp_dict['epoints_contour'] = econt.get('epts')
tmp_dict['epoints_contour_unit'] = 'Rydberg'
tmp_dict['epoints_weights'] = econt.get('weights')
out_dict['energy_contour_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: energy contour'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
ncore, emax, lmax, descr_max = get_core_states(files['out_pot'])
tmp_dict = {}
tmp_dict['number_of_core_states_per_atom'] = ncore
tmp_dict['energy_highest_lying_core_state_per_atom'] = emax
tmp_dict['energy_highest_lying_core_state_per_atom_unit'] = 'Rydberg'
tmp_dict['descr_highest_lying_core_state_per_atom'] = descr_max
out_dict['core_states_group'] = tmp_dict
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: core_states'
msg_list.append(msg)
if debug:
traceback.print_exc()
try:
niter, nitermax, converged, nmax_reached, mixinfo = self._get_scfinfo(files['out_log'])
out_dict['convergence_group']['number_of_iterations'] = niter
out_dict['convergence_group']['number_of_iterations_max'] = nitermax
out_dict['convergence_group']['calculation_converged'] = converged
out_dict['convergence_group']['nsteps_exhausted'] = nmax_reached
out_dict['convergence_group']['imix'] = mixinfo[0]
out_dict['convergence_group']['strmix'] = mixinfo[1]
out_dict['convergence_group']['qbound'] = mixinfo[2]
out_dict['convergence_group']['fcm'] = mixinfo[3]
out_dict['convergence_group']['brymix'] = mixinfo[1]
except: # pylint: disable=bare-except
msg = 'Error parsing output of KKRimp: scfinfo'
msg_list.append(msg)
if debug:
traceback.print_exc()
#convert numpy arrays to standard python lists
out_dict = convert_to_pystd(out_dict)
# return output with error messages if there are any
return len(msg_list) == 0, msg_list, out_dict