scnc
/
abinit
mirror of https://github.com/abinit/abinit.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
abinit/doc/tutorial/paw3_assets/scripts/comp_bands_abinit2elk.py

241 lines
8.4 KiB
Python
Executable File
Raw Permalink Blame History

#!/usr/bin/python
import sys, string
if len(sys.argv)<2 or len(sys.argv)>11:
print ('Usage: python comp_band_abinit2elk.py [abinit _EIG file] ...')
print (' ... [elk BAND.OUT file] [align #kpt #band] ...')
print (' ... [nbdbuf #] [Fermi #abinit #elk]')
print (' (align, eV, nbdbuf and Fermi sections optional)')
sys.exit()
# Check arguments
eVconv = ''
str_nbdbuf = ''
nbdbuf = 0
align_values = 0
align_ikpt = 0
align_iband = 0
align_values_fermi = 0
align_abinit_fermi = 0.0
align_elk_fermi = 0.0
iarg = 3
while iarg < len(sys.argv):
if str(sys.argv[iarg])=='eV':
eVconv = str(sys.argv[iarg]) # Assume Ha values and convert to eV
print ('# Values assumed to be in Ha and converted to eV')
if str(sys.argv[iarg])=='align':
align_values = 1
align_ikpt = int(sys.argv[iarg+1])
align_iband = int(sys.argv[iarg+2])
print ('# Values aligned at kpt:','%4i'%align_ikpt,\
' and band:','%4i'%align_iband)
iarg = iarg + 2
if (str(sys.argv[iarg])=='Fermi' or \
str(sys.argv[iarg])=='fermi'): # Align Fermi energies
align_values_fermi = 1
align_abinit_fermi = float(sys.argv[iarg+1])
align_elk_fermi = float(sys.argv[iarg+2])
iarg = iarg + 2
if str(sys.argv[iarg])=='nbdbuf':
nbdbuf = int(sys.argv[iarg+1])
print ('# nbdbuf set, last:','%4i'%nbdbuf,' bands will be ignored')
iarg = iarg + 1
iarg = iarg + 1
# Parse abinit file
input_file1_name = str(sys.argv[1]) # name of first input file (first argument)
input_file1_r = open(input_file1_name,'r') # open it as read file
abinit_file_data = input_file1_r.readlines()
input_file1_r.close()
# Read in k-point data as a list of numbers, by k-point
abinit_band_data = []
k_point_list = []
for iline in range(2,len(abinit_file_data)):
# Skip line if it is a k-point spec.
if abinit_file_data[iline].find('kpt#') > -1:
continue
# Accumulate values
new_values = map(float,abinit_file_data[iline].split())
k_point_list.extend(new_values)
# If we are on last line, finish appending
if iline == len(abinit_file_data)-1:
abinit_band_data.append(k_point_list)
break
# If the next line is a k-point spec., append.
if abinit_file_data[iline+1].find('kpt#') > -1:
abinit_band_data.append(k_point_list)
k_point_list = []
continue
#print 'nkpt:',len(abinit_band_data)
#print 'nbnds:',len(abinit_band_data[0])
nkpt = len(abinit_band_data)
nbands = len(abinit_band_data[0])
# Initialise variables for elk file
input_file2_name = str(sys.argv[2]) # name of second input file(second argument)
input_file2_r = open(input_file2_name, "r") # open it as read file
# If alignment is done, parse elk file and find alignment value
if align_values:
elk_file_data = input_file2_r.readlines()
band = 0
kpt = -1
previous_kpt_val = -1.0
for iline in range(0,len(elk_file_data)-1):
kpt = kpt + 1
if elk_file_data[iline].split() == []: # skip blank lines
kpt = -1
band = band + 1
continue
val_list = list(map(float,elk_file_data[iline].split()))
elk_align_value = val_list[1]
if previous_kpt_val==elk_align_value: # skip repeated kpts
kpt = kpt - 1
band = band - 1
continue
if kpt+1==align_ikpt and band+1==align_iband:
break
abinit_align_value = abinit_band_data[align_ikpt-1][align_iband-1]
print ('# Abinit alignment value: ',\
'%18.9E'%abinit_align_value)
print ('# Elk alignment value: ',\
'%18.9E'%elk_align_value)
input_file2_r.close()
input_file2_r = open(input_file2_name, "r")
# If Fermi alignment is done, print info
if align_values_fermi:
print ('# Abinit Fermi energy (Ha): ',\
'%18.9E'%align_abinit_fermi)
print ('# Elk Fermi energy (Ha): ',\
'%18.9E'%align_elk_fermi)
print ('# k-point abinit val elk val diff %diff')
# Now we begin the reading and comparison of the data
Ha_to_eV = 27.21138386
avg_diff = 0.0
avg_percentage = 0.0
nvals = 0
min_diff = 1000000000.0
max_diff = -100000000.0
input_file2_current_line='\n'
band = 0
kpt = -1
previous_kpt_val = -1.0
reset_done = 1
while input_file2_current_line.endswith('\n'):
kpt = kpt + 1
#print kpt
input_file2_current_line = input_file2_r.readline()
if not input_file2_current_line.endswith('\n'): # stop at last line
break
if input_file2_current_line.split() == []: # skip blank lines
kpt = -1
band = band + 1
if band == nbands-nbdbuf:
break
print (' ')
continue
# Parse the records of the lines
numbers_list2 = []
numbers2 = map(float, input_file2_current_line.split())
numbers_list2.extend(numbers2)
# Cycle if k-point value is repeated in elk
if numbers_list2[0] == previous_kpt_val:
kpt = kpt - 1
continue
previous_kpt_val = numbers_list2[0]
# Calculate difference,average,max,min
if align_values:
abinit_band_data[kpt][band] = abinit_band_data[kpt][band] \
- abinit_align_value
numbers_list2[1] = numbers_list2[1] - elk_align_value
if align_values_fermi:
abinit_band_data[kpt][band] = abinit_band_data[kpt][band] \
- align_abinit_fermi
numbers_list2[1] = numbers_list2[1] - align_elk_fermi
if eVconv=='eV':
abinit_band_data[kpt][band] = abinit_band_data[kpt][band]*Ha_to_eV
numbers_list2[1] = numbers_list2[1]*Ha_to_eV
diff = numbers_list2[1] - abinit_band_data[kpt][band]
if numbers_list2[1]!=0.0:
percentage = (abinit_band_data[kpt][band]/numbers_list2[1] - 1.0)*100.0
else:
percentage = 0.0
avg_diff = avg_diff + abs(diff)
avg_percentage = avg_percentage + abs(percentage)
nvals = nvals + 1
if diff<min_diff:
min_diff = diff
min_percentage = percentage
if diff>max_diff:
max_diff = diff
max_percentage = percentage
if align_values and band+1==align_iband and reset_done: # Save current averages for the
occ_avg_diff = avg_diff # occupied states and reset
occ_avg_percentage = avg_percentage
occ_min_diff = min_diff
occ_min_percentage = min_percentage
occ_max_diff = max_diff
occ_max_percentage = max_percentage
occ_nvals = nvals
nvals = 0
avg_diff = 0.0
avg_percentage = 0.0
min_diff = 1000000000.0
max_diff = -100000000.0
reset_done = 0
# Print the output data
if abs(abinit_band_data[kpt][band])<0.1 or abs(numbers_list2[1])<0.1:
print ('%14.9f'%numbers_list2[0],'%18.9E'%abinit_band_data[kpt][band],\
'%18.9E'%numbers_list2[1],'%12.3E'%diff,' (','%7.3f'%percentage,'%)')
else:
print ('%14.9f'%numbers_list2[0],'%14.9f'%abinit_band_data[kpt][band],\
'%18.9f'%numbers_list2[1],'%16.3E'%diff,' (','%7.3f'%percentage,'%)')
input_file2_r.close
avg_diff = avg_diff/float(nvals)
avg_percentage = avg_percentage/float(nvals)
if eVconv=='eV':
eVconv=' eV'
else:
eVconv=' Ha'
if align_values:
occ_avg_diff = occ_avg_diff/float(occ_nvals)
occ_avg_percentage = occ_avg_percentage/float(occ_nvals)
print ('#')
print ('# AVERAGES FOR OCCUPIED STATES:')
print ('# nvals:','%5i'%occ_nvals)
print ('# average diff:','%12.6F'%occ_avg_diff,eVconv)
print ('# minimum diff:','%12.6F'%occ_min_diff,eVconv)
print ('# maximum diff:','%12.6F'%occ_max_diff,eVconv)
print ('#')
print ('# AVERAGES FOR UNOCCUPIED STATES:')
print ('# nvals:','%5i'%nvals)
print ('# average diff:','%12.6F'%avg_diff,eVconv)
print ('# minimum diff:','%12.6F'%min_diff,eVconv)
print ('# maximum diff:','%12.6F'%max_diff,eVconv)
print ('#')
print ('# NOTE: Abinit values are read in fixed format with five decimal')
print ('# places. For low values, ~E-04 or ~E-03 may be the highest')
print ('# precision you can get.')
else:
print ('# nvals:','%5i'%nvals)
print ('# average diff:','%12.6F'%avg_diff,eVconv)
print ('# minimum diff:','%12.6F'%min_diff,eVconv)
print ('# maximum diff:','%12.6F'%max_diff,eVconv)
print ('#')
print ('# NOTE: Abinit values are read in fixed format with five decimal')
print ('# places. For low values, ~E-04 or ~E-03 may be the highest')
print ('# precision you can get.')
Reference in New Issue View Git Blame Copy Permalink