def run(options):
if options.filename == None:
raise ValueError("\nNo input file supplied [-i filename]\n")
try:
if options.killfile is not None and os.path.exists(options.killfile):
print("The purpose of the killfile option is to create that file")
print("only when you want fitallb to stop")
print("If the file already exists when you run fitallb it is")
print("never going to get started")
raise ValueError("Your killfile " + options.killfile +
" already exists")
except:
pass
#Read and check input
far = check_input.parse_input(
input_file=options.filename) # Make instance of parse_input class
far.read() # read input file
far.check() # check validity of input
if far.missing == True: # if problem exit
logging.info('MISSING ITEMS')
sys.exit()
far.initialize() # if ok initialize
# mandatory farfield info
far.read_par(far.files['par_file']) # read detector.par file
far.read_res(
) # read paramters file to resume refinement NB! optional
far.read_flt(far.files['flt_file']) # read peaks_t##.flt file
if far.files['res_file'] == None:
far.read_log() # read grainspotter.log file
far.read_rej(
far.files['rej_file']
) # read file containing rejected peaks to resume refinement NB! optional
far.set_start() # set values and errors for refinement start
check_input.set_globals(far)
# optional nearfield info
if far.files['near_flt_file'] != None:
check_input.interrupt(options.killfile)
assert far.files[
'near_par_file'] != None, 'near_par_file parameter file for near-field detector is missing'
near = deepcopy(far)
# take special care of near-field keywords (eg copy near_dety_size to dety_size)
for key in list(near.fit.keys()):
if key[0:5] == 'near_':
near.fit[key[5:len(key)]] = near.fit[key]
near.fit['stem'] = far.fit['stem'] + '_near'
# print 'far.files', far.files
# print 'near.files', near.files
near.read_par(near.files['near_par_file'])
near.read_flt(near.files['near_flt_file'])
keys = [
'cell__a', 'cell__b', 'cell__c', 'cell_alpha', 'cell_beta',
'cell_gamma', 'cell_lattice_[P,A,B,C,I,F,R]', 'chi', 'omegasign',
'wavelength'
]
for key in keys:
assert near.param[key] == far.param[
key], '%s is different for near- and far-field detectors' % key
# in case of different wedge use farfield value and refine
if near.param['wedge'] != far.param['wedge']:
near.param['wedge'] = far.param['wedge']
near.values = far.values
near.errors = far.errors
near.fit['skip'] = far.fit['skip']
check_input.set_globals(near)
# match peaks on nearfield detector and reject outliers
from FitAllB import near_field
check_input.interrupt(options.killfile)
near_field.find_refl(near)
check_input.interrupt(options.killfile)
near_field.match(near)
near.read_rej(
near.files['near_rej_file']
) # read file containing rejected peaks to resume refinement NB! optional
from FitAllB import error
error.vars(near)
from FitAllB import build_fcn
build_fcn.FCN(near)
check_input.interrupt(options.killfile)
near.reject()
near.write_rej()
near.fit['reforder'] = ['start', 'rotpos', 'end']
near.fit['goon'] = near.fit['reforder'][0]
# nearfield refinement
from FitAllB import fit
fit.refine(near, options.killfile)
try:
far.fitarg = near.fitarg # needed to carry on parameters for iminuit
except:
pass
# Farfield outlier rejection
#check_input.set_globals(far)
if far.files['res_file'] != None and far.labels == None:
check_input.interrupt(options.killfile)
from FitAllB import near_field
near_field.find_refl(far)
check_input.interrupt(options.killfile)
near_field.match(far)
far.read_rej(
far.files['rej_file']
) # read file containing rejected peaks to resume refinement NB! optional
from FitAllB import error
error.vars(far)
from FitAllB import build_fcn
build_fcn.FCN(far)
check_input.interrupt(options.killfile)
far.reject()
far.write_rej()
# farfield refinement
far.fit['reforder'] = ['grain', 'final', 'end']
far.fit['goon'] = far.fit['reforder'][0]
from FitAllB import fit
fit.refine(far, options.killfile)
# program ends here after deleting fcn.py and fcn.pyc
print('\nNormal termination of FitAllB')
os.remove('%s/fcn.py' % far.fit['direc'])
# os.remove('%s/fcn.pyc' % far.fit['direc'])
# pyc file now goes inside a __pycache__ directory
import shutil
pycache_dir = os.path.join(far.fit['direc'], "__pycache__")
shutil.rmtree(pycache_dir)
from FitAllB import error
error.vars(far)
from FitAllB import build_fcn
build_fcn.FCN(far)
far.reject()
far.write_rej()
# farfield refinement
for k in range(far.fit['cycle']):
#while len(far.fit['newreject_grain']) > 0:
# refine grain paramters
far.fit['reforder'] = ['start%s' % k, 'rotpos%s' % k, 'end']
far.fit['goon'] = far.fit['reforder'][0]
far.fit['newreject_grain'] = list(range(far.no_grains + 1))
from FitAllB import fit
fit.refine(far)
# refine globals
far.fit['reforder'] = ['globals%s' % k, 'end']
far.fit['goon'] = far.fit['reforder'][0]
from FitAllB import fitga
fitga.refine(far)
far.residual = []
far.volume = []
far.mean_ia = []
for i in range(far.no_grains):
far.residual.append([])
far.volume.append([])
far.mean_ia.append([])
for j in range(far.nrefl[i]):
far.residual[i].append(1)
far.volume[i].append(1)
def run(options):
if options.filename == None:
#parser.print_help()
print "\nNo input file supplied [-i filename]\n"
sys.exit()
#Read and check input
far = check_input.parse_input(
input_file=options.filename) # Make instance of parse_input class
far.read() # read input file
far.check() # check validity of input
if far.missing == True: # if problem exit
logging.info('MISSING ITEMS')
sys.exit()
far.initialize() # if ok initialize
# mandatory farfield info
far.read_par(far.files['par_file']) # read detector.par file
far.read_res(
) # read paramters file to resume refinement NB! optional
far.read_flt(far.files['flt_file']) # read peaks_t##.flt file
if far.files['res_file'] == None:
far.read_log() # read grainspotter.log file
far.read_rej(
far.files['rej_file']
) # read file containing rejected peaks to resume refinement NB! optional
far.set_start() # set values and errors for refinement start
check_input.set_globals(far)
#for key in far.param.keys():
# print key,far.param[key]
# Farfield outlier rejection
if far.files['res_file'] != None and far.labels == None:
from FitAllB import near_field
near_field.find_refl(far)
near_field.match(far)
far.read_rej(
far.files['rej_file']
) # read file containing rejected peaks to resume refinement NB! optional
from FitAllB import error
error.vars(far)
from FitAllB import build_fcn
build_fcn.FCN(far)
far.reject()
far.write_rej()
# farfield refinement
for k in range(far.fit['cycle']):
#while len(far.fit['newreject_grain']) > 0:
# refine grain paramters
far.fit['reforder'] = ['start%s' % k, 'rotpos%s' % k, 'end']
far.fit['goon'] = far.fit['reforder'][0]
if k == 0: #start with globals in stead of grain refinement
far.fit['reforder'] = ['start%s' % k, 'end']
far.fit['newreject_grain'] = range(far.no_grains + 1)
from FitAllB import fit
fit.refine(far)
# refine globals
far.fit['reforder'] = ['globals%s' % k, 'end']
far.fit['goon'] = far.fit['reforder'][0]
from FitAllB import fitga
fitga.refine(far)
far.residual = []
far.volume = []
far.mean_ia = []
for i in range(far.no_grains):
far.residual.append([])
far.volume.append([])
far.mean_ia.append([])
for j in range(far.nrefl[i]):
far.residual[i].append(1)
far.volume[i].append(1)
far.mean_ia[i].append(1)
from FitAllB import reject
reject.residual(far, far.fit['rej_resmedian'])
reject.mean_ia(far, far.fit['rej_ia'] * float(k + 2) / float(k + 1))
reject.intensity(far)
from FitAllB import write_output
write_output.write_rej(far, message='globals%i' % k)
for i in range(far.no_grains):
if far.nrefl[i] < far.fit['min_refl'] and i + 1 not in far.fit[
'skip']:
far.fit['skip'].append(i + 1)
far.fit['skip'].sort()
# optional nearfield info
if far.files['near_flt_file'] != None:
assert far.files[
'near_par_file'] != None, 'near_par_file parameter file for near-field detector is missing'
near = deepcopy(far)
# take special care of near-field keywords (eg copy near_dety_size to dety_size)
for key in near.fit.keys():
if key[0:5] == 'near_':
near.fit[key[5:len(key)]] = near.fit[key]
near.fit['stem'] = far.fit['stem'] + '_near'
near.read_par(near.files['near_par_file'])
near.read_flt(near.files['near_flt_file'])
keys = [
'cell__a', 'cell__b', 'cell__c', 'cell_alpha', 'cell_beta',
'cell_gamma', 'cell_lattice_[P,A,B,C,I,F,R]', 'chi', 'omegasign',
'wavelength'
]
for key in keys:
assert near.param[key] == far.param[
key], '%s is different for near- and far-field detectors' % key
# in case of different wedge use farfield value and refine
if near.param['wedge'] != far.param['wedge']:
near.param['wedge'] = far.param['wedge']
near.values = far.values
near.errors = far.errors
near.fit['skip'] = far.fit['skip']
check_input.set_globals(near)
# match peaks on nearfield detector and reject outliers
from FitAllB import near_field
near_field.find_refl(near)
near_field.match(near)
from FitAllB import error
error.vars(near)
from FitAllB import build_fcn
build_fcn.FCN(near)
near.reject()
near.write_rej()
# nearfield refinement
for k in range(near.fit['cycle']):
# refine grain paramters
near.fit['reforder'] = ['start%s' % k, 'rotpos%s' % k, 'end']
near.fit['goon'] = near.fit['reforder'][0]
near.fit['newreject_grain'] = range(near.no_grains + 1)
from FitAllB import fit
fit.refine(near)
# refine globals
near.fit['reforder'] = ['globals%s' % k, 'end']
near.fit['goon'] = near.fit['reforder'][0]
from FitAllB import fitga
fitga.refine(near)
near.residual = []
near.volume = []
near.mean_ia = []
for i in range(near.no_grains):
near.residual.append([])
near.volume.append([])
near.mean_ia.append([])
for j in range(near.nrefl[i]):
near.residual[i].append(1)
near.volume[i].append(1)
near.mean_ia[i].append(1)
from FitAllB import reject
reject.residual(near, near.fit['rej_resmedian'])
reject.mean_ia(near,
near.fit['rej_ia'] * float(k + 2) / float(k + 1))
reject.intensity(near)
from FitAllB import write_output
write_output.write_rej(near, message='globals%i' % k)
for i in range(near.no_grains):
if near.nrefl[i] < near.fit[
'min_refl'] and i + 1 not in near.fit['skip']:
near.fit['skip'].append(i + 1)
near.fit['skip'].sort()
# program ends here after deleting fcn.py and fcn.pyc
print '\nNormal termination of FitGlobAll'
os.remove('%s/fcn.py' % far.fit['direc'])
os.remove('%s/fcn.pyc' % far.fit['direc'])