def test_CenteredParameter_row():
"""Tests probflow.parameters.CenteredParameter w/ center_by=row + 2D"""
# Create the parameter
param = CenteredParameter([5, 6], center_by="row")
# posterior_mean should return mean
sample1 = param.posterior_mean()
sample2 = param.posterior_mean()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample2.shape[0] == 5
assert sample1.shape[1] == 6
assert sample2.shape[1] == 6
assert np.all(sample1 == sample2)
# mean of each column should be 0
assert np.all(np.abs(np.mean(sample1, axis=1)) < 1e-5)
# posterior_sample should return samples
sample1 = param.posterior_sample()
sample2 = param.posterior_sample()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample1.shape[1] == 6
assert sample2.shape[0] == 5
assert sample2.shape[1] == 6
assert np.all(sample1 != sample2)
# mean of each column should be 0
assert np.all(np.abs(np.mean(sample1, axis=1)) < 1e-5)
# posterior_sample should be able to return multiple samples
sample1 = param.posterior_sample(10)
sample2 = param.posterior_sample(10)
assert sample1.ndim == 3
assert sample2.ndim == 3
assert sample1.shape[0] == 10
assert sample1.shape[1] == 5
assert sample1.shape[2] == 6
assert sample2.shape[0] == 10
assert sample2.shape[1] == 5
assert sample2.shape[2] == 6
assert np.all(sample1 != sample2)
# mean of each column for each sample should be 0
assert np.all(np.abs(np.mean(sample1, axis=2)) < 1e-5)
def test_CenteredParameter_all_1d():
"""Tests probflow.parameters.CenteredParameter w/ center_by=all + 1D"""
# Create the parameter
param = CenteredParameter(5)
# posterior_mean should return mean
sample1 = param.posterior_mean()
sample2 = param.posterior_mean()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample2.shape[0] == 5
assert sample1.shape[1] == 1
assert sample2.shape[1] == 1
assert np.all(sample1 == sample2)
# mean should be 0! (that's the point of a centered parameter!)
assert is_close(np.mean(sample1), 0)
# posterior_sample should return samples
sample1 = param.posterior_sample()
sample2 = param.posterior_sample()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample1.shape[1] == 1
assert sample2.shape[0] == 5
assert sample2.shape[1] == 1
assert np.all(sample1 != sample2)
# mean should be 0! (that's the point of a centered parameter!)
assert is_close(np.mean(sample1), 0)
# posterior_sample should be able to return multiple samples
sample1 = param.posterior_sample(10)
sample2 = param.posterior_sample(10)
assert sample1.ndim == 3
assert sample2.ndim == 3
assert sample1.shape[0] == 10
assert sample1.shape[1] == 5
assert sample1.shape[2] == 1
assert sample2.shape[0] == 10
assert sample2.shape[1] == 5
assert sample2.shape[2] == 1
assert np.all(sample1 != sample2)
# mean should be 0!
assert np.all(np.abs(np.mean(sample1, axis=1)) < 1e-5)
def test_CenteredParameter_all_2d():
"""Tests probflow.parameters.CenteredParameter w/ center_by=all + 2D"""
# Shouldn't allow >2 dims
with pytest.raises(ValueError):
param = CenteredParameter([5, 6, 7])
# Create the parameter
param = CenteredParameter([5, 6])
# posterior_mean should return mean
sample1 = param.posterior_mean()
sample2 = param.posterior_mean()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample2.shape[0] == 5
assert sample1.shape[1] == 6
assert sample2.shape[1] == 6
assert np.all(sample1 == sample2)
# mean should be 0! (that's the point of a centered parameter!)
assert is_close(np.mean(sample1), 0)
# posterior_sample should return samples
sample1 = param.posterior_sample()
sample2 = param.posterior_sample()
assert sample1.ndim == 2
assert sample2.ndim == 2
assert sample1.shape[0] == 5
assert sample1.shape[1] == 6
assert sample2.shape[0] == 5
assert sample2.shape[1] == 6
assert np.all(sample1 != sample2)
# mean should be 0!
assert is_close(np.mean(sample1), 0)
# posterior_sample should be able to return multiple samples
sample1 = param.posterior_sample(10)
sample2 = param.posterior_sample(10)
assert sample1.ndim == 3
assert sample2.ndim == 3
assert sample1.shape[0] == 10
assert sample1.shape[1] == 5
assert sample1.shape[2] == 6
assert sample2.shape[0] == 10
assert sample2.shape[1] == 5
assert sample2.shape[2] == 6
assert np.all(sample1 != sample2)
# mean should be 0!
assert np.all(np.abs(np.mean(sample1.reshape((10, -1)), axis=1)) < 1e-5)
def __init__(self, di, do):
self.w = CenteredParameter([di, do], center_by="column")