def _test_multinomial_goodness_of_fit(dim):
seed_all(0)
sample_count = int(1e5)
probs = numpy.random.dirichlet([1] * dim)
counts = numpy.random.multinomial(sample_count, probs)
p_good = multinomial_goodness_of_fit(probs, counts, sample_count)
assert_greater(p_good, TEST_FAILURE_RATE)
unif_counts = numpy.random.multinomial(sample_count, [1. / dim] * dim)
p_bad = multinomial_goodness_of_fit(probs, unif_counts, sample_count)
assert_less(p_bad, TEST_FAILURE_RATE)
def test_scipy_stats():
seed_all(0)
for name in dir(scipy.stats):
if hasattr(getattr(scipy.stats, name), 'rvs'):
yield _test_scipy_stats, name
]
def split_example(i):
example = split_examples[i]
discrete, continuous = split_discrete_continuous(example['mixed'])
assert_equal(discrete, example['discrete'])
assert_almost_equal(continuous, example['continuous'])
def test_split_continuous_discrete():
for i in xrange(len(split_examples)):
yield split_example, i
seed_all(0)
default_params = {
'bernoulli': [(0.2,)],
'binom': [(40, 0.4)],
'dirichlet': [
(1.0 + rand(2),),
(1.0 + rand(3),),
(1.0 + rand(4),),
],
'erlang': [(7,)],
'dlaplace': [(0.8,)],
'frechet': [tuple(2 * rand(1)) + (0,) + tuple(2 * rand(2))],
'geom': [(0.1,)],
'hypergeom': [(40, 14, 24)],
'logser': [(0.9,)],
'multivariate_normal': [
