def testRandom(self):
pdf = Normal(4, 2)
data = pdf.random(100000)
self.assertAlmostEqual(numpy.mean(data), pdf.mu, delta=0.05)
self.assertAlmostEqual(numpy.std(data), pdf.sigma, delta=0.05)
def testRandom(self):
pdf = Normal(4, 2)
data = pdf.random(100000)
self.assertAlmostEqual(numpy.mean(data), pdf.mu, delta=0.05)
self.assertAlmostEqual(numpy.std(data), pdf.sigma, delta=0.05)
def testParameters(self):
pdf = Normal(2.5, 1.5)
pdf.mu = 33
pdf.sigma = 44
self.assertEqual(pdf.mu, 33)
self.assertEqual(pdf.sigma, 44)
self.assertEqual(pdf.sigma, pdf['sigma'])
def testParameters(self):
pdf = Normal(2.5, 1.5)
pdf.mu = 33
pdf.sigma = 44
self.assertEqual(pdf.mu, 33)
self.assertEqual(pdf.sigma, 44)
self.assertEqual(pdf.sigma, pdf['sigma'])
def testParameterEstimation(self):
mu, sigma = 2.2, 0.3
data = numpy.random.normal(mu, sigma, 100000)
pdf = Normal(1, 1)
pdf.estimate(data)
self.assertAlmostEqual(pdf.mu, mu, delta=0.05)
self.assertAlmostEqual(pdf.sigma, sigma, delta=0.05)
def testParameterEstimation(self):
mu, sigma = 2.2, 0.3
data = numpy.random.normal(mu, sigma, 100000)
pdf = Normal(1, 1)
pdf.estimate(data)
self.assertAlmostEqual(pdf.mu, mu, delta=0.05)
self.assertAlmostEqual(pdf.sigma, sigma, delta=0.05)
def testSampleFromHistogram(self):
mu = 5.
sigma = 1.
normal = Normal(mu, sigma)
x = normal.random(10000)
hx, p = density(x, 100)
samples = hx[sample_from_histogram(p, n_samples=10000)]
self.assertAlmostEqual(mu, numpy.mean(samples), delta=0.5)
self.assertAlmostEqual(sigma, numpy.std(samples), delta=0.5)
def testSampleFromHistogram(self):
mu = 5.
sigma = 1.
normal = Normal(mu, sigma)
x = normal.random(10000)
hx, p = density(x, 100)
samples = hx[sample_from_histogram(p, n_samples=10000)]
self.assertAlmostEqual(mu, numpy.mean(samples), delta=0.5)
self.assertAlmostEqual(sigma, numpy.std(samples), delta=0.5)
def testLogProb(self):
n = Normal(0, 1)
self.assertAlmostEqual(n(0.),
1. / numpy.sqrt(2 * numpy.pi),
delta=1e-1)
self.assertAlmostEqual(n(1.),
numpy.exp(-0.5) / numpy.sqrt(2 * numpy.pi),
delta=1e-1)
