def testBijector(self):
loc = 0.3
scale = 5.
concentration = np.array([[[-5.5], [-20], [0.], [1.]]],
dtype=np.float32)
bijector = tfb.GeneralizedExtremeValueCDF(loc=loc,
scale=scale,
concentration=concentration,
validate_args=True)
self.assertStartsWith(bijector.name, "generalizedextremevalue")
x = np.array([[[0.], [-3.], [0.], [4.2]]], dtype=np.float32)
# GeneralizedExtremeValue distribution
gev_dist = stats.genextreme(-concentration, loc=loc, scale=scale)
y = gev_dist.cdf(x).astype(np.float32)
self.assertAllClose(y, self.evaluate(bijector.forward(x)))
self.assertAllClose(x, self.evaluate(bijector.inverse(y)))
self.assertAllClose(
np.squeeze(gev_dist.logpdf(x), axis=-1),
self.evaluate(bijector.forward_log_det_jacobian(x, event_ndims=1)))
self.assertAllClose(
self.evaluate(
-bijector.inverse_log_det_jacobian(y, event_ndims=1)),
self.evaluate(bijector.forward_log_det_jacobian(x, event_ndims=1)),
rtol=1e-4,
atol=0.)
def testScalarCongruency(self):
bijector_test_util.assert_scalar_congruency(
tfb.GeneralizedExtremeValueCDF(loc=0.3,
scale=20.,
concentration=0.2),
lower_x=1.,
upper_x=100.,
eval_func=self.evaluate,
rtol=0.05)
def testVariableScale(self):
x = tf.Variable(1.)
b = tfb.GeneralizedExtremeValueCDF(loc=0.,
scale=x,
concentration=0.6,
validate_args=True)
self.evaluate(x.initializer)
self.assertIs(x, b.scale)
self.assertEqual((), self.evaluate(b.forward(1.)).shape)
with self.assertRaisesOpError("Argument `scale` must be positive."):
with tf.control_dependencies([x.assign(-1.)]):
self.evaluate(b.forward(2.))
def testBijectiveAndFinite(self):
bijector = tfb.GeneralizedExtremeValueCDF(loc=0.,
scale=3.0,
concentration=2.0,
validate_args=True)
x = np.linspace(-1.4, 10., num=10).astype(np.float32)
y = np.linspace(0.01, 0.99, num=10).astype(np.float32)
bijector_test_util.assert_bijective_and_finite(bijector,
x,
y,
eval_func=self.evaluate,
event_ndims=0,
rtol=1e-3)