fuzz1means = [sum(cuts[i1]) / len(cuts[i1]) for cuts in RCs_cuts]
fuzz1_ranks = TOPSIS.getCrispRanks(fuzz1means)
RC_UC_meas1 = [
[ID] for ID in alt_IDs
] #capture uncertainty as measured by ratio of 0.1 level support to value at m=1.0
for i in range(len(RC_UC_meas1)):
RC_UC_meas1[i].append((RCs_cuts[i][alphas.index(0.1)][1] - - RCs_cuts[i][alphas.index(0.1)][0])/ \
RCs_cuts[i][alphas.index(1.0)][1])
#get & diplay dominance (possibility) matrix
DM = [[a] for a in alt_IDs]
for i in range(len(alt_IDs)):
DM_row = []
for j in range(alts):
d = fuzzy.dominance_AlphaCut(alphas, RCs_cuts[i], alphas, RCs_cuts[j])
#get dominance of alt i over alt j
DM_row.append(d)
DM[i].append(DM_row)
print 'Fuzzy Dominance Matrix'
for i in range(len(DM)):
print DM[i][0], ':', [str(x)[0:5] for x in DM[i][1]], ':', str(
min(DM[i][1]))[0:5], ':', str(sum(DM[i][1]) / len(DM[i][1]))[0:5]
#rank monte carlo fit means
prob_ranks = TOPSIS.getCrispRanks([nf[0] for nf in norm_fits])
#get monte carlo CIs
CIs = []
for i in range(len(full_RCs)):
m, cl, cu = confidence_interval(full_RCs[i], confidence=alpha)
#get mean of alpha cut at 1.0 and rank them
i1 = alphas.index(1.0)
fuzz1means = [sum(cuts[i1])/len(cuts[i1]) for cuts in P_cuts]
fuzz1_ranks = AHP.getCrispRanks(fuzz1means)
P_UC_meas1 = [[ID] for ID in alt_IDs] #capture uncertainty as measured by ratio of 0.1 level support to value at m=1.0
for i in range(len(P_UC_meas1)):
P_UC_meas1[i].append((P_cuts[i][alphas.index(0.1)][1] - - P_cuts[i][alphas.index(0.1)][0])/ \
P_cuts[i][alphas.index(1.0)][1])
#get & diplay dominance (possibility) matrix
DM = [[a] for a in alt_IDs];
for i in range(alts):
DM_row = []
for j in range(alts):
d = fuzz.dominance_AlphaCut(alphas, P_cuts[i], alphas, P_cuts[j])
#get dominance of alt i over alt j
DM_row.append(d)
DM[i].append(DM_row)
print '\n'
print 'Fuzzy Dominance Matrix'
for i in range(len(DM)):
print DM[i][0], ':', [str(x)[0:5] for x in DM[i][1]], ':', str(min(DM[i][1]))[0:5], ':', str(sum(DM[i][1])/len(DM[i][1]))[0:5]
#rank monte carlo fit means
prob_ranks = AHP.getCrispRanks([nf[0] for nf in norm_fits])
#get monte carlo CIs
CIs = []
for i in range(len(full_Ps)):
m, cl, cu = confidence_interval(full_Ps[i], confidence=alpha)
#get mean of alpha cut at 1.0 and rank them
i1 = alphas.index(1.0)
fuzz1means = [sum(cuts[i1])/len(cuts[i1]) for cuts in RCs_cuts]
fuzz1_ranks = TOPSIS.getCrispRanks(fuzz1means)
RC_UC_meas1 = [[ID] for ID in alt_IDs] #capture uncertainty as measured by ratio of 0.1 level support to value at m=1.0
for i in range(len(RC_UC_meas1)):
RC_UC_meas1[i].append((RCs_cuts[i][alphas.index(0.1)][1] - - RCs_cuts[i][alphas.index(0.1)][0])/ \
RCs_cuts[i][alphas.index(1.0)][1])
#get & diplay dominance (possibility) matrix
DM = [[a] for a in alt_IDs];
for i in range(len(alt_IDs)):
DM_row = []
for j in range(alts):
d = fuzzy.dominance_AlphaCut(alphas, RCs_cuts[i], alphas, RCs_cuts[j])
#get dominance of alt i over alt j
DM_row.append(d)
DM[i].append(DM_row)
print 'Fuzzy Dominance Matrix'
for i in range(len(DM)):
print DM[i][0], ':', [str(x)[0:5] for x in DM[i][1]], ':', str(min(DM[i][1]))[0:5], ':', str(sum(DM[i][1])/len(DM[i][1]))[0:5]
#rank monte carlo fit means
prob_ranks = TOPSIS.getCrispRanks([nf[0] for nf in norm_fits])
#get monte carlo CIs
CIs = []
for i in range(len(full_RCs)):
m, cl, cu = confidence_interval(full_RCs[i], confidence=alpha)
CIs.append([m, cl, cu])
fuzz1means = [sum(cuts[i1]) / len(cuts[i1]) for cuts in P_cuts]
fuzz1_ranks = AHP.getCrispRanks(fuzz1means)
P_UC_meas1 = [
[ID] for ID in alt_IDs
] #capture uncertainty as measured by ratio of 0.1 level support to value at m=1.0
for i in range(len(P_UC_meas1)):
P_UC_meas1[i].append((P_cuts[i][alphas.index(0.1)][1] - - P_cuts[i][alphas.index(0.1)][0])/ \
P_cuts[i][alphas.index(1.0)][1])
#get & diplay dominance (possibility) matrix
DM = [[a] for a in alt_IDs]
for i in range(alts):
DM_row = []
for j in range(alts):
d = fuzz.dominance_AlphaCut(alphas, P_cuts[i], alphas, P_cuts[j])
#get dominance of alt i over alt j
DM_row.append(d)
DM[i].append(DM_row)
print '\n'
print 'Fuzzy Dominance Matrix'
for i in range(len(DM)):
print DM[i][0], ':', [str(x)[0:5] for x in DM[i][1]], ':', str(
min(DM[i][1]))[0:5], ':', str(sum(DM[i][1]) / len(DM[i][1]))[0:5]
#rank monte carlo fit means
prob_ranks = AHP.getCrispRanks([nf[0] for nf in norm_fits])
#get monte carlo CIs
CIs = []
for i in range(len(full_Ps)):