def test_data():
hp = gvar.BufferDict({'log(sdev)': gvar.log(gvar.gvar(1, 1))})
x = np.linspace(0, 5, 10)
def gpfactory(hp):
gp = lgp.GP(lgp.ExpQuad() * hp['sdev']**2)
gp.addx(x, 'x')
return gp
truehp = gvar.sample(hp)
truegp = gpfactory(truehp)
trueprior = truegp.prior()
def makeerr(bd, err):
return gvar.BufferDict(bd, buf=np.full_like(bd.buf, err))
data_noerr = gvar.sample(trueprior)
error = makeerr(data_noerr, 0.1)
zeroerror = makeerr(data_noerr, 0)
zerocov = gvar.evalcov(gvar.gvar(data_noerr, zeroerror))
data_err = gvar.make_fake_data(gvar.gvar(data_noerr, error))
datas = [
[
data_noerr,
gvar.gvar(data_noerr),
(data_noerr, ),
(data_noerr, zerocov),
lambda _: data_noerr,
lambda _: gvar.gvar(data_noerr),
lambda _: (data_noerr, ),
lambda _: (data_noerr, zerocov),
],
[
data_err,
(data_err, ),
(gvar.mean(data_err), gvar.evalcov(data_err)),
lambda _: data_err,
lambda _: (data_err, ),
lambda _: (gvar.mean(data_err), gvar.evalcov(data_err)),
],
]
for datasets in datas:
fits = []
for data in datasets:
fit = lgp.empbayes_fit(hp, gpfactory, data)
fits.append(fit)
p = fits[0].minresult.x
for fit in fits[1:]:
np.testing.assert_allclose(fit.minresult.x, p, atol=1e-6)
def setUp(self):
gv.ranseed(1)
a = gv.gvar('1.000(1)')
b = gv.gvar('0.500(1)')
self.x = np.array([0.1, 0.2, 0.3, 0.4])
def fcn(p, x=self.x):
ans = gv.BufferDict()
ans['l'] = p['a'] + p['b'] * x
ans['c1'] = 4 * [p['a']]
ans['c2'] = 4 * [p['a']]
return ans
self.prior = gv.BufferDict([('a', a), ('b', b)])
self.data = gv.make_fake_data(fcn(self.prior))
self.fcn = fcn
# reference fit without using MultiFitter
self.ref_fit = lsqfit.nonlinear_fit(prior=self.prior,
fcn=self.fcn,
data=self.data)
# these data should be ignored
self.data['dummy'] = gv.gvar(['1(1)', '2(2)'])
return prior
#### FAKE DATA ####
truehp = gvar.sample(hyperprior)
truegp = makegp(truehp)
trueprior = makeprior(truegp, plot=True)
trueparams = gvar.sample(trueprior)
truedata = fcn(trueparams)
dataerr = {
k: np.full_like(v, 0.1 * (np.max(v) - np.min(v)))
for k, v in truedata.items()
}
data = gvar.make_fake_data(gvar.gvar(truedata, dataerr))
def check_constraints(y):
# integrate approximately with trapezoid rule
integ = np.sum((y[:, 1:] + y[:, :-1]) / 2 * np.diff(plotgrid), 1)
print('int dx x (Sigma(x) + g(x)) =', integ[0] + integ[-1])
for i in range(1, 5):
print(f'int dx {tpnames[i]}(x) =', integ[i])
for i, name in enumerate(tpnames):
print(f'{name}(1) =', y[i, -1])
print('\ncheck constraints in fake data:')
check_constraints(trueparams['plotgrid'])