def run_point(model, **input_pars):
stdouts = OrderedDict()
obj, stdout = tools.get_ctypes_streams(func=slha_generator.run,
args=[model], kwargs=input_pars[slha_generator.name])
stdouts.update({slhamodule.name: stdout})
slhafile = SLHA(obj)
predictor_output = OrderedDict()
for predictor in predictors:
is_modifier = predictor in slha_modifiers
result, stdout = tools.get_ctypes_streams(
func=slhamodule.send_to_predictor,
args=[slhafile,input_pars.get(predictor.name),predictor, is_modifier])
predictor_output.update(result)
stdouts.update({predictor.name: stdout})
return slhafile.process(), predictor_output , stdouts
def run_point( **input_pars):
"""
run_point is the core function of MC++
documentation is needed soon, however, there is extensive commenting already
"""
#==================
# define predictors
#==================
spectrum_generator=input_pars['spectrum_generator']
spectrum_modifiers=input_pars['spectrum_modifiers']
predictors=input_pars['predictors']
#=====================================
# define directory for temporary files
#=====================================
if 'tmp_dir' in input_pars:
tmp_dir_input={'tmp_dir':input_pars['tmp_dir']}
for predictor in [spectrum_generator] + predictors:
try:
input_pars[predictor.name].update(tmp_dir_input)
except:
input_pars[predictor.name]=tmp_dir_input
#define standard outs dict
stdouts = OrderedDict()
#==============================================================
# handle verbosity like this 'verbose': ['softsusy', 'FH', ...]
# verbosity helps with debugging
#==============================================================
if not input_pars.get('verbose'): input_pars['verbose']=[]
#make an option to select all verbosity
if 'all' in input_pars['verbose']:
input_pars['verbose']=[pred.name for pred in predictors]
input_pars['verbose'].append(spectrum_generator.name)
input_pars['verbose'].append(slhamodule.name)
input_pars['verbose'].append('spectrum')
input_pars['verbose'].append('mcspectrum')
input_pars['verbose'].append('predspectrum')
if input_pars.get('spectrumfile'):
#=============================================================
# if an slhafile is given, the spectrum generation is skipped
# the slhafile also does not get modified
#=============================================================
spectrum_modifiers=[]
slhafile = SLHA(input_pars.get('tmp_dir'))
# make slha object
none_return,stdout = tools.get_ctypes_streams(func=slhafile.read,args=[input_pars['spectrumfile']],kwargs={})
# handle standard outs
if slhamodule.name in input_pars['verbose']:
print(stdout)
stdouts.update({slhamodule.name:stdout})
# handle spectrum verbosity
if 'spectrum' in input_pars['verbose']:
print("NOTE: Running on spectrum file {0}.".format(input_pars['spectrumfile']))
print(' Spectrum is not modified.')
else:
# ======================
# run spectrum generator
# ======================
# check for verbosity
slha_gen_verbose=spectrum_generator.name in input_pars['verbose']
# run the spectrum calculator
(obj,err), stdout = tools.get_ctypes_streams(func=spectrum_generator.run,
args=[input_pars[spectrum_generator.name]], kwargs={'verbose':slha_gen_verbose})
# print standard out
if slha_gen_verbose:
print(stdout)
#error handling for slha generator
if err: print("ERROR: spectrum calculator")
# If the spectrum calculater fails then there is no hope to calculate anything else.
# Hence, return None is appropriate
if(err): return None,None,None
stdouts.update({slhamodule.name: stdout})
# if you want to see the slha file from the specturm generator, put 'verbose' in the
if 'spectrum' in input_pars['verbose']:
print(obj)
# make slha object
slhafile,stdout = tools.get_ctypes_streams(func=SLHA,args=[obj,input_pars.get('lookup'),input_pars.get('tmp_dir')],kwargs={})
# slhafile=SLHA(obj,input_pars.get('lookup'),input_pars.get('tmp_dir'))
# handle standard outs
if slhamodule.name in input_pars['verbose']:
print(stdout)
stdouts.update({slhamodule.name:stdout})
# ======================
# predictions start here
# ======================
predictor_output = OrderedDict()
# =============================================================
# save inputs to spectrum calculater like ('MINPAR','in_M0')
# Motivation: in NUHM1 and NUHM2, m0, m12 and A0 are translated
# into EXTPAR parameters. We need this for consistent naming
# for e.g. constraints
# =============================================================
in_dict={}
if not 'spectrumfile' in input_pars:
for key, val in input_pars[spectrum_generator.name].items():
if not (key=='model') and not (key=='version'):
in_key =(key[0],'in_{0}'.format(key[1]))
in_dict[in_key]=val
predictor_output.update(in_dict)
# ======================================
# save spectrum before any modifications
# ======================================
predictor_output.update(slhafile.process())
# =====================================================
# WARNING: here is a functionality needed by mastercode
# it is not generic
# "Manually" setting values in the slha file object
# if 'mc_slha_update' is in input_vars
# someone may define 'mc_slha_update : True',
# or provide a dictionary 'mc_slha_update':{ ('MASS','MZ') : ... }
# this hack should not escape from this file
# =====================================================
if input_pars.get('mc_slha_update'):
# by default, use set MW=80.4
values={('MASS','MW') : 80.4}
try:
values.update( input_pars['mc_slha_update'])
except TypeError:
pass
for oid, val in values.items():
slhafile[oid]=val
#explicitely save these in predictor_output as modified slha values
mod_key=(oid[0],'mod_{}'.format(oid[1]))
predictor_output[mod_key]=val
if 'mcspectrum' in input_pars['verbose']:
print(slhafile)
# ===========================================
# run predictors on slha file and save output
# ===========================================
for predictor in predictors:
is_modifier = predictor in spectrum_modifiers
pred_verbose=(predictor.name in input_pars['verbose'])
if pred_verbose:
verbose_true={'verbose': True }
if input_pars.get(predictor.name):
input_pars[predictor.name].update(verbose_true)
else:
input_pars[predictor.name]=verbose_true
result, stdout = tools.get_ctypes_streams(
func=slhamodule.send_to_predictor,
args=[slhafile,input_pars.get(predictor.name),predictor, is_modifier])
if pred_verbose: print(stdout)
predictor_output.update(result)
stdouts.update({predictor.name: stdout})
if 'predspectrum' in input_pars['verbose']:
print(slhafile)
# =====================================================
# save the modified slha obs as e.g. ('MASS','mod_mh0')
# =====================================================
mod_slha_obs=slhafile.process()
for key, val in mod_slha_obs.items():
if not predictor_output.get(key)==val:
mod_key=(key[0],'mod_{}'.format(key[1]))
predictor_output[mod_key]=val
return slhafile, predictor_output, stdouts
