def test_right_triangles(self):
right_triangle_min = self.right_triangles_min()
right_triangle_min_stats = {
'upper': 3.6648459510835116,
'lower': 1.0188257132115006,
'median': 1.8789162245549995,
'mode': 1.0,
'mean': 2
}
right_triangle_min_stats_expected_arr = np.array([
right_triangle_min_stats['upper'],
right_triangle_min_stats['lower'],
right_triangle_min_stats['median'],
right_triangle_min_stats['mode'],
right_triangle_min_stats['mean'],
])
calculated_stats_min = TriangularUncertainty.statistics(
right_triangle_min)
right_triangle_min_stats_calculated_arr = np.array([
calculated_stats_min['upper'],
calculated_stats_min['lower'],
calculated_stats_min['median'],
calculated_stats_min['mode'],
calculated_stats_min['mean'],
])
self.assertTrue(
np.allclose(right_triangle_min_stats_expected_arr,
right_triangle_min_stats_calculated_arr,
rtol=1e-03))
right_triangle_max = self.right_triangles_max()
right_triangle_max_stats = {
'upper': 3.981087302412936,
'lower': 1.3350311212247066,
'median': 3.1208194598959116,
'mode': 4.0,
'mean': 3
}
right_triangle_max_stats_expected_arr = np.array([
right_triangle_max_stats['upper'],
right_triangle_max_stats['lower'],
right_triangle_max_stats['median'],
right_triangle_max_stats['mode'],
right_triangle_max_stats['mean'],
])
calculated_stats_max = TriangularUncertainty.statistics(
right_triangle_max)
right_triangle_max_stats_calculated_arr = np.array([
calculated_stats_max['upper'],
calculated_stats_max['lower'],
calculated_stats_max['median'],
calculated_stats_max['mode'],
calculated_stats_max['mean'],
])
self.assertTrue(
np.allclose(right_triangle_max_stats_expected_arr,
right_triangle_max_stats_calculated_arr,
rtol=1e-03))
def test_triangular_pdf(self):
oneDparams = self.biased_params_1d()
xs, ys = TriangularUncertainty.pdf(oneDparams)
self.assertTrue(np.allclose(np.array([1, 3, 4]), xs))
self.assertTrue(np.allclose(np.array([0, 0.66666669, 0]), ys))
points = np.array([1, 2, 3, 4])
xs, ys = TriangularUncertainty.pdf(oneDparams, points)
self.assertTrue(np.allclose(points, xs))
self.assertTrue(
np.allclose(np.array([0, 0.99999997, 1.99999994, 0]), ys))
def test_triangular_random(self):
oneDparams = self.biased_params_1d()
params = self.biased_params_2d()
variables = TriangularUncertainty.random_variables(oneDparams, 5000)
self.assertTrue(2.61 < np.average(variables) < 2.71)
self.assertEqual((1, 5000), variables.shape)
variables = TriangularUncertainty.random_variables(params, 5000)
self.assertFalse(np.allclose(variables[0, :], variables[1, :]))
self.assertFalse(np.allclose(variables[0, :], variables[1, :]))
self.assertTrue(2.61 < np.average(variables[0, :]) < 2.71)
self.assertTrue(2.61 < np.average(variables[1, :]) < 2.71)
def test_triangular_cdf(self):
oneDparams = self.biased_params_1d()
params = self.biased_params_2d()
percentages = np.array([0, 0.5, 1])
self.assertTrue(
np.allclose(
TriangularUncertainty.cdf(oneDparams,
np.array([[1, 2.73205083, 4]])),
percentages))
self.assertTrue(
np.allclose(
TriangularUncertainty.cdf(params, np.array([1, 2.73205083])),
np.array([[0], [0.5]])))
def test_triangular_seeded_random(self):
oneDparams = self.biased_params_1d()
self.assertTrue(
np.allclose(
2.85968765,
TriangularUncertainty.random_variables(oneDparams, 1,
self.seeded_random())))
def test_triangular_statistics(self):
oneDparams = self.biased_params_1d()
tri_stats = {
'upper': 3.8063508384608244,
'lower': 1.2738612828334341,
'median': 2.7320508333784455,
'mode': 3.0,
'mean': 2.6666667461395264
}
self.assertEqual(TriangularUncertainty.statistics(oneDparams),
tri_stats)
def test_bounds(self):
params = self.make_params_array(1)
params['minimum'] = 1
params['loc'] = 0
params['maximum'] = 4
with self.assertRaises(ImproperBoundsError):
TriangularUncertainty.validate(params)
params['loc'] = 5
with self.assertRaises(ImproperBoundsError):
TriangularUncertainty.validate(params)
params['loc'] = 1
TriangularUncertainty.validate(params)
params['loc'] = 4
TriangularUncertainty.validate(params)