def test_one_dee_distributions(self):
p_1 = {(1, ) : 0.1,
(2, ) : 0.2,
(3, ) : 0.3,
(4, ) : 0.4,}
f = lambda state : (0, )
p = statistics.map_distribution(f, p_1)
assert len(p) == 1
assert ((0, ) in p) and (p[(0, )] == 1.0)
mu = statistics.expectation(p_1)
mu_goal = numpy.array([3.0])
assert_almost_equal(mu, mu_goal)
sigma_squared = statistics.variance(p_1)
sigma_squared_goal = numpy.asarray([1.0])
assert_almost_equal(sigma_squared, sigma_squared_goal)
def test_one_dee_distributions(self):
p_1 = {
(1, ): 0.1,
(2, ): 0.2,
(3, ): 0.3,
(4, ): 0.4,
}
f = lambda state: (0, )
p = statistics.map_distribution(f, p_1)
assert len(p) == 1
assert ((0, ) in p) and (p[(0, )] == 1.0)
mu = statistics.expectation(p_1)
mu_goal = numpy.array([3.0])
assert_almost_equal(mu, mu_goal)
sigma_squared = statistics.variance(p_1)
sigma_squared_goal = numpy.asarray([1.0])
assert_almost_equal(sigma_squared, sigma_squared_goal)
def test_two_dee_distributions(self):
p_2 = {(0, 0) : 0.2,
(0, 1) : 0.3,
(1, 0) : 0.2,
(3, 3) : 0.3,}
f = lambda state : (state[0], )
p = statistics.map_distribution(f, p_2)
assert len(p) == 3
assert ((0, ) in p) and (p[(0, )] == 0.5)
assert ((1, ) in p) and (p[(1, )] == 0.2)
assert ((3, ) in p) and (p[(3, )] == 0.3)
mu = statistics.expectation(p_2)
mu_goal = numpy.asarray([1.1, 1.2])
assert_almost_equal(mu, mu_goal)
cov = statistics.covariance(p_2)
cov_goal = -1.1*-1.2*0.2 -1.1*-0.2*0.3 -0.1*-1.2*0.2 +1.9*1.8*0.3
assert_almost_equal(cov, cov_goal)
def test_two_dee_distributions(self):
p_2 = {
(0, 0): 0.2,
(0, 1): 0.3,
(1, 0): 0.2,
(3, 3): 0.3,
}
f = lambda state: (state[0], )
p = statistics.map_distribution(f, p_2)
assert len(p) == 3
assert ((0, ) in p) and (p[(0, )] == 0.5)
assert ((1, ) in p) and (p[(1, )] == 0.2)
assert ((3, ) in p) and (p[(3, )] == 0.3)
mu = statistics.expectation(p_2)
mu_goal = numpy.asarray([1.1, 1.2])
assert_almost_equal(mu, mu_goal)
cov = statistics.covariance(p_2)
cov_goal = -1.1 * -1.2 * 0.2 - 1.1 * -0.2 * 0.3 - 0.1 * -1.2 * 0.2 + 1.9 * 1.8 * 0.3
assert_almost_equal(cov, cov_goal)