def test_inverse(self):
batch_size = 10
shape = [2, 3, 4]
inputs = torch.randn(batch_size, *shape)
transform = standard.IdentityTransform()
outputs, logabsdet = transform.inverse(inputs)
self.assert_tensor_is_good(outputs, [batch_size] + shape)
self.assert_tensor_is_good(logabsdet, [batch_size])
self.assertEqual(outputs, inputs)
self.assertEqual(logabsdet, torch.zeros(batch_size))
def test_inverse(self):
batch_size = 10
shape = [2, 3, 4]
inputs = torch.randn(batch_size, *shape)
transforms = [
standard.AffineScalarTransform(scale=2.0),
standard.IdentityTransform(),
standard.AffineScalarTransform(scale=-0.25),
]
composite = base.CompositeTransform(transforms)
reference = standard.AffineScalarTransform(scale=-0.5)
outputs, logabsdet = composite.inverse(inputs)
outputs_ref, logabsdet_ref = reference.inverse(inputs)
self.assert_tensor_is_good(outputs, [batch_size] + shape)
self.assert_tensor_is_good(logabsdet, [batch_size])
self.assertEqual(outputs, outputs_ref)
self.assertEqual(logabsdet, logabsdet_ref)
def test_forward_inverse_are_consistent(self):
batch_size = 10
shape = [5, 10, 15]
inputs = torch.rand(batch_size, *shape)
transforms = [
nl.Tanh(),
nl.LogTanh(),
nl.LeakyReLU(),
nl.Sigmoid(),
nl.Logit(),
nl.CompositeCDFTransform(nl.Sigmoid(),
standard.IdentityTransform())
]
self.eps = 1e-3
for transform in transforms:
with self.subTest(transform=transform):
self.assert_forward_inverse_are_consistent(transform, inputs)
def test_forward(self):
batch_size = 10
shape = [5, 10, 15]
inputs = torch.rand(batch_size, *shape)
transforms = [
nl.Tanh(),
nl.LogTanh(),
nl.LeakyReLU(),
nl.Sigmoid(),
nl.Logit(),
nl.CompositeCDFTransform(nl.Sigmoid(),
standard.IdentityTransform())
]
for transform in transforms:
with self.subTest(transform=transform):
outputs, logabsdet = transform(inputs)
self.assert_tensor_is_good(outputs, [batch_size] + shape)
self.assert_tensor_is_good(logabsdet, [batch_size])
def test_forward_inverse_are_consistent(self):
batch_size = 10
shape = [2, 3, 4]
inputs = torch.randn(batch_size, *shape)
transform = standard.IdentityTransform()
self.assert_forward_inverse_are_consistent(transform, inputs)