def test_visiting_dist_high_temperature(self):
lu = list(zip(*self.ld_bounds))
lower = np.array(lu[0])
upper = np.array(lu[1])
vd = VisitingDistribution(lower, upper, self.qv, self.rs)
values = np.zeros(self.nbtestvalues)
for i in np.arange(self.nbtestvalues):
values[i] = vd.visit_fn(self.high_temperature)
# Visiting distribution is a distorted version of Cauchy-Lorentz
# distribution, and as no 1st and higher moments (no mean defined,
# no variance defined).
# Check that big tails values are generated
assert_array_less(np.min(values), 1e-10)
assert_array_less(1e+10, np.max(values))
def test_visiting_stepping(self):
lu = list(zip(*self.ld_bounds))
lower = np.array(lu[0])
upper = np.array(lu[1])
dim = lower.size
vd = VisitingDistribution(lower, upper, self.qv, self.rs)
values = np.zeros(dim)
x_step_low = vd.visiting(values, 0, self.high_temperature)
# Make sure that only the first component is changed
assert_equal(np.not_equal(x_step_low, 0), True)
values = np.zeros(dim)
x_step_high = vd.visiting(values, dim, self.high_temperature)
# Make sure that component other than at dim has changed
assert_equal(np.not_equal(x_step_high[0], 0), True)
def test_visiting_stepping(self):
lu = list(zip(*self.ld_bounds))
lower = np.array(lu[0])
upper = np.array(lu[1])
dim = lower.size
vd = VisitingDistribution(lower, upper, self.qv, self.rs)
values = np.zeros(dim)
x_step_low = vd.visiting(values, 0, self.high_temperature)
# Make sure that only the first component is changed
assert_equal(np.not_equal(x_step_low, 0), True)
values = np.zeros(dim)
x_step_high = vd.visiting(values, dim, self.high_temperature)
# Make sure that component other than at dim has changed
assert_equal(np.not_equal(x_step_high[0], 0), True)
def test_visiting_dist_high_temperature(self):
lu = list(zip(*self.ld_bounds))
lower = np.array(lu[0])
upper = np.array(lu[1])
vd = VisitingDistribution(lower, upper, self.qv, self.rs)
values = np.zeros(self.nbtestvalues)
for i in np.arange(self.nbtestvalues):
values[i] = vd.visit_fn(self.high_temperature)
# Visiting distribution is a distorted version of Cauchy-Lorentz
# distribution, and as no 1st and higher moments (no mean defined,
# no variance defined).
# Check that big tails values are generated
assert_array_less(np.min(values), 1e-10)
assert_array_less(1e+10, np.max(values))