def test_frozen_matrix_normal(self):
for i in range(1, 5):
for j in range(1, 5):
M = 0.3 * np.ones((i, j))
U = 0.5 * np.identity(i) + 0.5 * np.ones((i, i))
V = 0.7 * np.identity(j) + 0.3 * np.ones((j, j))
frozen = matrix_normal(mean=M, rowcov=U, colcov=V)
rvs1 = frozen.rvs(random_state=1234)
rvs2 = matrix_normal.rvs(mean=M,
rowcov=U,
colcov=V,
random_state=1234)
assert_equal(rvs1, rvs2)
X = frozen.rvs(random_state=1234)
pdf1 = frozen.pdf(X)
pdf2 = matrix_normal.pdf(X, mean=M, rowcov=U, colcov=V)
assert_equal(pdf1, pdf2)
logpdf1 = frozen.logpdf(X)
logpdf2 = matrix_normal.logpdf(X, mean=M, rowcov=U, colcov=V)
assert_equal(logpdf1, logpdf2)
def pdf(x, cov_scale=0.25):
# added trace
mu01 = np.array([[-0.5, 0.5]])
mu02 = np.array([[0.5, -0.5]])
mu11 = np.array([[0.5, 0.5]])
mu12 = np.array([[-0.5, -0.5]])
cov = cov_scale * np.eye(2)
inv_cov = np.linalg.inv(cov)
p01 = matrix_normal.pdf(x, mean=mu01, colcov=cov)
p02 = matrix_normal.pdf(x, mean=mu02, colcov=cov)
p11 = matrix_normal.pdf(x, mean=mu11, colcov=cov)
p12 = matrix_normal.pdf(x, mean=mu12, colcov=cov)
# return np.max([p01,p02])/(p01+p02+p11+p12)
return (p01 + p02) / (p01 + p02 + p11 + p12)
def matrix_normal_testcase():
from scipy.stats import matrix_normal
M = np.arange(6).reshape(3, 2)
U = np.diag([1, 2, 3])
V = 0.3 * np.identity(2)
X = M + 0.1
logging.debug(matrix_normal.pdf(X, mean=M, rowcov=U, colcov=V))
vectorised_X = X.T.flatten()
equiv_mean = M.T.flatten()
equiv_cov = np.kron(V, U)
mn_pdf = multivariate_normal.pdf(vectorised_X,
mean=equiv_mean,
cov=equiv_cov)
logging.debug(mn_pdf)
def test_frozen_matrix_normal(self):
for i in range(1,5):
for j in range(1,5):
M = 0.3 * np.ones((i,j))
U = 0.5 * np.identity(i) + 0.5 * np.ones((i,i))
V = 0.7 * np.identity(j) + 0.3 * np.ones((j,j))
frozen = matrix_normal(mean=M, rowcov=U, colcov=V)
rvs1 = frozen.rvs(random_state=1234)
rvs2 = matrix_normal.rvs(mean=M, rowcov=U, colcov=V,
random_state=1234)
assert_equal(rvs1, rvs2)
X = frozen.rvs(random_state=1234)
pdf1 = frozen.pdf(X)
pdf2 = matrix_normal.pdf(X, mean=M, rowcov=U, colcov=V)
assert_equal(pdf1, pdf2)
logpdf1 = frozen.logpdf(X)
logpdf2 = matrix_normal.logpdf(X, mean=M, rowcov=U, colcov=V)
assert_equal(logpdf1, logpdf2)