def print_chi2_breakdown(chain,mcf):
import models
import variables as v
model = models.get_model_from_file(mcf)
lhoods = models.get_lhood_names(mcf)
MCVdict=v.mc_variables()
if len( model ) > 0 :
print "\nchi2 penalties from gaussian constraints :"
print "==================================================================="
print " Penalty Prediction Name Type Constraint"
print "==================================================================="
for constraint in model:
sn=constraint.short_name
MCV=MCVdict[sn]
v_index = MCV.get_index(mcf)
chi2=chain.treeVars["contributions"][v_index]
pred=chain.treeVars["predictions"][v_index]
#print "{:11g} {:<{width}{precision}{base}}{c!r}".format(chi2, pred, base='g', width=1, precision=4, c=constraint)
print "{:11.2f} {:11.4g} {!r}".format(chi2, pred, constraint)
#print "{chi2:>f} {".format(chi2=chi2)
print "==================================================================="
if len(lhoods.keys()) > 0 : print "\nThe likelihoods give penalties:\n"
for i, lhood in enumerate(lhoods.items()):
chi2=chain.treeVars["lhoods"][i]
print "{:11.2f} {:16} {:16}". format( chi2, lhood[0], lhood[1] )
def get_p_value_n_dof(chain,mcf):
import models
import variables as v
model = models.get_model_from_file(mcf)
lhoods = models.get_lhood_names(mcf)
MCVdict=v.mc_variables()
chi2_s=[]
for constraint in model:
sn=constraint.short_name
chi2_s.append(get_contribution(chain,mcf,sn))
for i, lhood in enumerate(lhoods.items()):
chi2_s.append(chain.treeVars["lhoods"][i])
count=0
for chi2 in chi2_s:
if chi2 > 0 : count += 1
n_dof = count - mcf.Inputs
chi2_tot = chain.treeVars["predictions"][0]
p_value= r.TMath.Prob(chi2_tot, n_dof)
return p_value, n_dof
