1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
|
#!/usr/bin/env python
#
# Author: Mike McKerns (mmckerns @uqfoundation)
# Copyright (c) 2020-2026 The Uncertainty Quantification Foundation.
# License: 3-clause BSD. The full license text is available at:
# - https://github.com/uqfoundation/mystic/blob/master/LICENSE
"""
misc user-defined items (solver configuration, moment constraints)
"""
from mystic.solvers import DifferentialEvolutionSolver2, NelderMeadSimplexSolver
from mystic.termination import ChangeOverGeneration as COG
from mystic.monitors import Monitor, VerboseLoggingMonitor
from mystic.bounds import Bounds, MeasureBounds
# kwds for solver #TODO: tune
opts = dict(termination=COG(1e-6, 100))
param = dict(solver=DifferentialEvolutionSolver2,
npop=10,
maxiter=2000,
maxfun=1e+6,
x0=None, # use RandomInitialPoints
nested=None, # use SetNested
map=None, # use SetMapper
stepmon=VerboseLoggingMonitor(1,10, filename='log.txt'), # monitor
#evalmon=Monitor(), # monitor config (re-initialized in solve)
# kwds to pass directly to Solve(objective, **opt)
opts=opts,
)
# kwds for sampling
kwds = dict(npts=500, ipts=4, itol=1e-8, iter=5)
from mystic.constraints import and_, integers, sorting
from mystic.coupler import outer, additive
from emulators import cost3, x3, bounds3, error3, a_beta, a_beta_error
from ouq_misc import (flatten, unflatten, normalize_moments, constrained, check,
constrain_moments, constrain_expected, constrained_out)
# lower and upper bound for parameters and weights
xlb, xub = zip(*bounds3)
wlb = (0,0,0)
wub = (1,1,1)
# number of Dirac masses to use for each parameter
npts = (2,2,2) #NOTE: rv = (w0,w0,x0,x0,w1,w1,x1,x1,w2,w2,x2,x2)
index = (2,3,6,7,10,11) #NOTE: rv[index] -> x0,x0,x1,x1,x2,x2
ordered = lambda constraint: sorting(index=(0,1))(sorting(index=(4,5))(sorting(index=(8,9))(constraint)))
# moments and uncertainty in first parameter
a_ave = x3[0]
a_var = .5 * error3[0]**2
a_ave_err = 2 * a_var
a_var_err = a_var
# moments and uncertainty in second parameter
b_ave = x3[1]
b_var = .5 * error3[1]**2
b_ave_err = 2 * b_var
b_var_err = b_var
# moments and uncertainty in third parameter
c_ave = x3[2]
c_var = .5 * error3[2]**2
c_ave_err = 2 * c_var
c_var_err = c_var
# moments and uncertainty in output
o_ave = a_beta
o_var = .5 * a_beta_error**2
o_ave_err = 2 * o_var
o_var_err = o_var
# build a model representing 'truth' F(x)
from ouq_models import WrapModel
nx = 3; ny = None
Ns = None #500 # number of samples of F(x) in the objective
nargs = dict(nx=nx, ny=ny, rnd=(True if Ns else False))
model = WrapModel('model', cost3, **nargs)
# set the bounds
bnd = MeasureBounds(xlb, xub, n=npts, wlb=wlb, wub=wub)
# constrain parameters at given index(es) to be ints
integer_indices = integers(ints=float, index=index)(lambda rv: rv)
## moment-based constraints ##
normcon = normalize_moments()
momcon0 = constrain_moments(a_ave, a_var, a_ave_err, a_var_err, idx=0)
momcon1 = constrain_moments(b_ave, b_var, b_ave_err, b_var_err, idx=1)
momcon2 = constrain_moments(c_ave, c_var, c_ave_err, c_var_err, idx=2)
is_con0 = constrained(a_ave, a_var, a_ave_err, a_var_err, idx=0)
is_con1 = constrained(b_ave, b_var, b_ave_err, b_var_err, idx=1)
is_con2 = constrained(c_ave, c_var, c_ave_err, c_var_err, idx=2)
is_ocon = constrained_out(model, ave=o_ave, var=o_var, ave_err=o_ave_err, var_err=o_var_err, debug=False)
is_cons = lambda c: bool(additive(is_ocon)(additive(is_con2)(additive(is_con1)(is_con0)))(c))
## position-based constraints ##
# impose constraints sequentially (faster, but assumes are decoupled)
_scons = outer(momcon2)(outer(momcon1)(outer(momcon0)(unflatten(npts)(ordered(flatten(npts)(normcon))))))
scons = flatten(npts)(constrain_expected(model, ave=o_ave, var=o_var, ave_err=o_ave_err, var_err=o_var_err, bounds=bnd, constraints=_scons, maxiter=50))
#scons = flatten(npts)(outer(constrain_expected(model, ave=o_ave, var=o_var, ave_err=o_ave_err, var_err=o_var_err, bounds=bnd))(_scons))
# impose constraints concurrently (slower, but safer)
#_ccons = unflatten(npts)(and_(flatten(npts)(normcon), flatten(npts)(momcon0), flatten(npts)(momcon1), flatten(npts)(momcon2))) #FIXME: broadcasting error
#ccons = flatten(npts)(constrain_expected(model, ave=o_ave, var=o_var, ave_err=o_ave_err, var_err=o_var_err, bounds=bnd, constraints=_ccons, maxiter=50))
_ccons = flatten(npts)(constrain_expected(model, ave=o_ave, var=o_var, ave_err=o_ave_err, var_err=o_var_err, bounds=bnd, maxiter=1000, debug=False))
ccons = and_(ordered(flatten(npts)(normcon)), flatten(npts)(momcon0), flatten(npts)(momcon1), flatten(npts)(momcon2), _ccons)
# check parameters (instead of measures)
iscon = check(npts)(is_cons)
|
