class TestNoisedValueDivisionWithOtherPositiveNoisedValue(
TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) / NoisedValue(val=1.12, var=0.2)
expected_val = 1.339285
expected_var = 0.68458
expected_err = 0.82739
expected_relative_err = 0.617787
class TestNoisedValueMultiplicationWithOtherNegativeNoisedValue(
TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) * NoisedValue(val=-2.3, var=2.75)
expected_val = -3.45
expected_var = 8.8325
expected_err = 2.97195
expected_relative_err = 0.861435
class TestNoisedValueLeftDivisionWithNegativeConstant(TestCase,
NoisedValueBaseTestCase):
v = -1.12 / NoisedValue(val=1.5, var=0.5)
expected_val = -0.746666
expected_var = 0.12389
expected_err = 0.35198
expected_relative_err = 0.4714
class TestNoisedValueInitializationWithoutVarianceOrError(
TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5)
expected_val = 1.5
expected_var = 0
expected_err = 0
expected_relative_err = 0
def test_representation_of_zero_error(self):
v = NoisedValue(val=1.5)
self.assertEqual(
"1.5 \u00B1 0 (0.000% error)",
str(v),
msg="NoisedValue representation is different than expected",
)
class TestNoisedValueInitializationWithError(TestCase,
NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, err=0.5)
expected_val = 1.5
expected_var = 0.25
expected_err = 0.5
expected_relative_err = 0.33333
def test_representation_of_non_zero(self):
v = NoisedValue(val=1.5, err=0.5)
self.assertEqual(
"1.5 \u00B1 0.5 (33.333% error)",
str(v),
msg="NoisedValue representation is different than expected",
)
class TestNoisedValueRightMultiplicationWithPositiveConstant(
TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) * 1.12
expected_val = 1.68
expected_var = 0.6272
expected_err = 0.79196
expected_relative_err = 0.4714
class TestNoisedValueInitializationWithValueZero(TestCase,
NoisedValueBaseTestCase):
v = NoisedValue(val=0, err=0.5)
expected_val = 0
expected_var = 0.25
expected_err = 0.5
expected_relative_err = None
class TestNoisedValueLeftMultiplicationWithNegativeConstant(
TestCase, NoisedValueBaseTestCase):
v = -1.12 * NoisedValue(val=1.5, var=0.5)
expected_val = -1.68
expected_var = 0.6272
expected_err = 0.79196
expected_relative_err = 0.4714
def test_representation_of_zero_value(self):
v = NoisedValue(val=0, err=0.5)
self.assertEqual(
"0 \u00B1 0.5 (\u221E% error)",
str(v),
msg="NoisedValue representation is different than expected",
)
class TestNoisedValueRightDivisionWithPositiveConstant(TestCase,
NoisedValueBaseTestCase
):
v = NoisedValue(val=1.5, var=0.5) / 1.12
expected_val = 1.339285
expected_var = 0.3986
expected_err = 0.631345
expected_relative_err = 0.4714
class TestNoisedValueLeftConstantSubtraction(TestCase, NoisedValueBaseTestCase):
v = 2.2 - NoisedValue(val=1.5, var=0.5)
expected_val = 0.7
expected_var = 0.5
expected_err = 0.70711
expected_relative_err = 1.010157
class TestNoisedValueExponential(TestCase, NoisedValueBaseTestCase):
v = np.exp(NoisedValue(val=1.5, err=0.5))
expected_val = 4.48169
expected_var = 5.02138
expected_err = 2.24084
expected_relative_err = 0.5
class TestNoisedValueInteger3Power(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, err=0.5) ** 3
expected_val = 3.375
expected_var = 11.390625
expected_err = 3.375
expected_relative_err = 1
class TestNoisedValueSin(TestCase, NoisedValueBaseTestCase):
v = np.sin(NoisedValue(val=np.pi / 6, err=0.5))
expected_val = 0.5
expected_var = 0.1875
expected_err = 0.433013
expected_relative_err = 0.866025
class TestNoisedValueCos(TestCase, NoisedValueBaseTestCase):
v = np.cos(NoisedValue(val=np.pi / 6, err=0.5))
expected_val = 0.866025
expected_var = 0.0625
expected_err = 0.25
expected_relative_err = 0.288675
class TestNoisedValueSubtractionWithOtherNoisedValue(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) - NoisedValue(val=2.2, var=2.75)
expected_val = -0.7
expected_var = 3.25
expected_err = 1.802775
expected_relative_err = 2.57539
class TestNoisedValueRightConstantSubtraction(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) - 2.2
expected_val = -0.7
expected_var = 0.5
expected_err = 0.70711
expected_relative_err = 1.010157
class TestNoisedValueInteger2Power(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, err=0.5) ** 2
expected_val = 2.25
expected_var = 2.25
expected_err = 1.5
expected_relative_err = 0.666666
class TestNoisedValueIntegerNegativePower(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, err=0.5) ** -1
expected_val = 0.666666
expected_var = 0.049382
expected_err = 0.222222
expected_relative_err = 0.333333
class TestNoisedValueInteger1Power(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, err=0.5) ** 1
expected_val = 1.5
expected_var = 0.25
expected_err = 0.5
expected_relative_err = 0.33333
class TestNoisedValueAdditionWithOtherNoisedValue(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) + NoisedValue(val=2.2, var=2.75)
expected_val = 3.7
expected_var = 3.25
expected_err = 1.802775
expected_relative_err = 0.487236
class TestNoisedValueRightConstantAddition(TestCase, NoisedValueBaseTestCase):
v = NoisedValue(val=1.5, var=0.5) + 2.2
expected_val = 3.7
expected_var = 0.5
expected_err = 0.70711
expected_relative_err = 0.19111
class TestNoisedValueNegativeValue(TestCase, NoisedValueBaseTestCase):
v = -NoisedValue(val=1.5, var=0.5)
expected_val = -1.5
expected_var = 0.5
expected_err = 0.70711
expected_relative_err = 0.4714
from noised_value import NoisedValue
rho_b = NoisedValue(val=7.812, err=0.055)
rho_l = NoisedValue(val=1.26, err=5.7e-4)
g = NoisedValue(val=981, err=10)
L = NoisedValue(val=22.22, err=2.8e-3)
a = NoisedValue(val=0.1964, err=0.0091)
eta = (2 * g * a * (rho_b - rho_l)) / (9 * L)
expected_eta = NoisedValue(val=12.5, err=1.44)
print(f"experiment eta is: {eta}")
print(f"expected eta is: {expected_eta}")
print(f"n sigma is: {expected_eta.n_sigma(eta)}")
print(NoisedValue(val=0, err=1))