def setUp(self):
particles = np.array([
[1., 2., 23.67, 0.],
[2., 4., 23.67, 1.],
[3., 6., 23.67, 3.],
[4., 8., 23.67, 6.],
])
self.emp = pb.EmpPdf(particles)
def test_shape(self):
# there were an error where EmpPdf returned nr of particles, so we use
# different data-set to detect such error
particles = np.array([
[1., 2.],
[2., 4.],
[3., 6.],
[4., 8.],
])
emp = pb.EmpPdf(particles)
self.assertEqual(emp.shape(), 2)
def test_sample(self):
"""Test LogNormPdf.sample() mean and variance."""
N = 500 # number of samples
samples = np.log(self.lognorm.samples(N)) # note the logarithm
emp = pb.EmpPdf(
samples) # Emipirical pdf computes sample mean and variance for us
mean, fuzz = 2., 0.2
self.assertTrue(np.all(abs(emp.mean() - mean) <= fuzz))
var, fuzz = 0.3, 0.1
self.assertTrue(np.all(abs(emp.variance() - var) <= fuzz))
def test_sample_multi(self):
"""Test GaussPdf.sample() mean and variance (multivariate case)."""
N = 500
mean = np.array([124.6, -1.5])
cov = np.array([[0.7953, 0.], [0., 1.7]])
samples = pb.GaussPdf(mean, cov).samples(N)
emp = pb.EmpPdf(samples)
self.assertAlmostEqual(np.max(abs(emp.mean() - mean)), 0., delta=0.2)
self.assertAlmostEqual(np.max(abs(emp.variance() - cov.diagonal())),
0.,
delta=0.3)
def test_sample(self):
"""Test GaussPdf.sample() mean and variance."""
N = 1000 # number of samples, variance is very sensible here
emp1 = pb.EmpPdf(self.gamma1.samples(
N)) # Emipirical pdf computes sample mean and variance for us
emp2 = pb.EmpPdf(self.gamma2.samples(
N)) # Emipirical pdf computes sample mean and variance for us
self.assertAlmostEqual(np.max(abs(emp1.mean() - self.gamma1.mean())),
0.,
delta=0.4)
self.assertAlmostEqual(np.max(abs(emp2.mean() - self.gamma2.mean())),
0.,
delta=0.033)
self.assertAlmostEqual(np.max(
abs(emp1.variance() - self.gamma1.variance())),
0.,
delta=34.0)
self.assertAlmostEqual(np.max(
abs(emp2.variance() - self.gamma2.variance())),
0.,
delta=0.06)
def test_sample(self):
"""Test GaussPdf.sample() mean and variance."""
N = 500 # number of samples
emp1 = pb.EmpPdf(self.gamma1.samples(
N)) # Emipirical pdf computes sample mean and variance for us
emp2 = pb.EmpPdf(self.gamma2.samples(
N)) # Emipirical pdf computes sample mean and variance for us
self.assertAlmostEqual(np.max(abs(emp1.mean() - self.gamma1.mean())),
0.,
delta=0.4)
self.assertAlmostEqual(np.max(abs(emp2.mean() - self.gamma2.mean())),
0.,
delta=0.35)
self.assertAlmostEqual(np.max(
abs(emp1.variance() - self.gamma1.variance())),
0.,
delta=3.0)
self.assertAlmostEqual(np.max(
abs(emp2.variance() - self.gamma2.variance())),
0.,
delta=1.4)
def test_sample_uni(self):
"""Test GaussPdf.sample() mean and variance (univariate case)."""
N = 500
mean = np.array([124.6])
cov = np.array([[0.7953]])
samples = pb.GaussPdf(mean, cov).samples(N)
emp = pb.EmpPdf(samples)
fuzz = 0.2
self.assertTrue(np.all(abs(emp.mean() - mean) <= fuzz))
var, fuzz = cov.diagonal(), 0.2
self.assertTrue(np.all(abs(emp.variance() - var) <= fuzz))
def test_sample(self):
"""Test GaussCPdf.sample() mean and variance."""
N = 500 # number of samples
samples = self.cgauss.samples(N, self.cond)
emp = pb.EmpPdf(
samples) # Emipirical pdf computes sample mean and variance for us
fuzz = np.array([0.3, 0.3])
mean = self.cgauss.mean(self.cond)
self.assertTrue(np.all(abs(emp.mean() - mean) <= fuzz))
var = self.cgauss.variance(self.cond)
self.assertTrue(np.all(abs(emp.variance() - var) <= fuzz))