def test_reduce_moment_matching_multivariate():
int_inputs = [('i', bint(4))]
real_inputs = [('x', reals(2))]
inputs = OrderedDict(int_inputs + real_inputs)
int_inputs = OrderedDict(int_inputs)
real_inputs = OrderedDict(real_inputs)
loc = numeric_array([[-10., -1.], [+10., -1.], [+10., +1.], [-10., +1.]])
precision = zeros(4, 1, 1) + ops.new_eye(loc, (2, ))
discrete = Tensor(zeros(4), int_inputs)
gaussian = Gaussian(loc, precision, inputs)
gaussian -= gaussian.log_normalizer
joint = discrete + gaussian
with interpretation(moment_matching):
actual = joint.reduce(ops.logaddexp, 'i')
assert_close(actual.reduce(ops.logaddexp), joint.reduce(ops.logaddexp))
expected_loc = zeros(2)
expected_covariance = numeric_array([[101., 0.], [0., 2.]])
expected_precision = _inverse(expected_covariance)
expected_gaussian = Gaussian(expected_loc, expected_precision, real_inputs)
expected_gaussian -= expected_gaussian.log_normalizer
expected_discrete = Tensor(ops.log(numeric_array(4.)))
expected = expected_discrete + expected_gaussian
assert_close(actual, expected, atol=1e-5, rtol=None)
def test_eager_subs_origin(int_inputs, real_inputs):
int_inputs = OrderedDict(sorted(int_inputs.items()))
real_inputs = OrderedDict(sorted(real_inputs.items()))
inputs = int_inputs.copy()
inputs.update(real_inputs)
g = random_gaussian(inputs)
# Check that Gaussian log density at origin is zero.
origin = {k: zeros(d.shape) for k, d in real_inputs.items()}
actual = g(**origin)
expected_data = zeros(tuple(d.size for d in int_inputs.values()))
expected = Tensor(expected_data, int_inputs)
assert_close(actual, expected)
def test_block_vector_batched(batch_shape):
shape = batch_shape + (10, )
expected = zeros(shape)
actual = BlockVector(shape)
expected[..., 1] = randn(batch_shape)
actual[..., 1] = expected[..., 1]
expected[..., 3:5] = randn(batch_shape + (2, ))
actual[..., 3:5] = expected[..., 3:5]
assert_close(actual.as_tensor(), expected)
def test_block_vector():
shape = (10, )
expected = zeros(shape)
actual = BlockVector(shape)
expected[1] = randn(())
actual[1] = expected[1]
expected[3:5] = randn((2, ))
actual[3:5] = expected[3:5]
assert_close(actual.as_tensor(), expected)
def test_block_matrix(sparse):
shape = (10, 10)
expected = zeros(shape)
actual = BlockMatrix(shape)
expected[1, 1] = randn(())
actual[1, 1] = expected[1, 1]
if not sparse:
expected[1, 3:5] = randn((2, ))
actual[1, 3:5] = expected[1, 3:5]
expected[3:5, 1] = randn((2, ))
actual[3:5, 1] = expected[3:5, 1]
expected[3:5, 3:5] = randn((2, 2))
actual[3:5, 3:5] = expected[3:5, 3:5]
assert_close(actual.as_tensor(), expected)
def test_block_matrix_batched(batch_shape, sparse):
shape = batch_shape + (10, 10)
expected = zeros(shape)
actual = BlockMatrix(shape)
expected[..., 1, 1] = randn(batch_shape)
actual[..., 1, 1] = expected[..., 1, 1]
if not sparse:
expected[..., 1, 3:5] = randn(batch_shape + (2, ))
actual[..., 1, 3:5] = expected[..., 1, 3:5]
expected[..., 3:5, 1] = randn(batch_shape + (2, ))
actual[..., 3:5, 1] = expected[..., 3:5, 1]
expected[..., 3:5, 3:5] = randn(batch_shape + (2, 2))
actual[..., 3:5, 3:5] = expected[..., 3:5, 3:5]
assert_close(actual.as_tensor(), expected)
def test_to_data_error():
data = zeros((3, 3))
x = Tensor(data, OrderedDict(i=bint(3)))
with pytest.raises(ValueError):
funsor.to_data(x)
def test_to_data():
data = zeros((3, 3))
x = Tensor(data)
assert funsor.to_data(x) is data