def build_gp_model(data, x_std=1.0, y_std=0.1):
dim = data.query_points.shape[-1]
empirical_variance = tf.math.reduce_variance(data.observations)
prior_lengthscales = [0.2 * x_std * np.sqrt(dim)] * dim
prior_scale = tf.cast(1.0, dtype=tf.float64)
x_std = tf.cast(x_std, dtype=tf.float64)
y_std = tf.cast(y_std, dtype=tf.float64)
kernel = gpflow.kernels.Matern52(
variance=empirical_variance,
lengthscales=prior_lengthscales,
)
kernel.variance.prior = tfp.distributions.LogNormal(
tf.math.log(y_std), prior_scale)
kernel.lengthscales.prior = tfp.distributions.LogNormal(
tf.math.log(kernel.lengthscales), prior_scale)
gpr = gpflow.models.GPR(
data.astuple(),
kernel,
mean_function=gpflow.mean_functions.Constant(),
noise_variance=1e-5,
)
gpflow.set_trainable(gpr.likelihood, False)
return GaussianProcessRegression(
model=gpr,
optimizer=Optimizer(gpflow.optimizers.Scipy(),
minimize_args={"options": dict(maxiter=100)}),
num_kernel_samples=100,
)
def test_gpr_raises_for_invalid_num_kernel_samples() -> None:
x_np = np.arange(5, dtype=np.float64).reshape(-1, 1)
x = tf.convert_to_tensor(x_np, x_np.dtype)
y = fnc_3x_plus_10(x)
with pytest.raises(ValueError):
GaussianProcessRegression(gpr_model(x, y), num_kernel_samples=-1)
def build_model(data: Dataset) -> GaussianProcessRegression:
variance = tf.math.reduce_variance(data.observations)
kernel = gpflow.kernels.Matern52(variance,
tf.constant([0.2, 0.2], tf.float64))
gpr = gpflow.models.GPR((data.query_points, data.observations),
kernel,
noise_variance=1e-5)
gpflow.utilities.set_trainable(gpr.likelihood, False)
return GaussianProcessRegression(gpr)
def build_model(data: Dataset) -> GaussianProcessRegression:
variance = tf.math.reduce_variance(data.observations)
kernel = gpflow.kernels.Matern52(variance, tf.constant([0.2, 0.2], tf.float64))
scale = tf.constant(1.0, dtype=tf.float64)
kernel.variance.prior = tfp.distributions.LogNormal(
tf.constant(-2.0, dtype=tf.float64), scale
)
kernel.lengthscales.prior = tfp.distributions.LogNormal(
tf.math.log(kernel.lengthscales), scale
)
gpr = gpflow.models.GPR((data.query_points, data.observations), kernel, noise_variance=1e-5)
gpflow.utilities.set_trainable(gpr.likelihood, False)
return GaussianProcessRegression(gpr)
def build_stacked_independent_objectives_model(
data: Dataset) -> ModelStack:
gprs = []
for idx in range(2):
single_obj_data = Dataset(
data.query_points, tf.gather(data.observations, [idx], axis=1))
variance = tf.math.reduce_variance(single_obj_data.observations)
kernel = gpflow.kernels.Matern52(
variance, tf.constant([0.2, 0.2], tf.float64))
gpr = gpflow.models.GPR(single_obj_data.astuple(),
kernel,
noise_variance=1e-5)
gpflow.utilities.set_trainable(gpr.likelihood, False)
gprs.append((GaussianProcessRegression(gpr), 1))
return ModelStack(*gprs)
def build_model(data):
variance = tf.math.reduce_variance(data.observations)
kernel = gpflow.kernels.RBF(variance=variance)
gpr = gpflow.models.GPR(data.astuple(), kernel, noise_variance=1e-5)
gpflow.set_trainable(gpr.likelihood, False)
return GaussianProcessRegression(gpr)
def test_sgpr_raises_for_covariance_between_points() -> None:
data = mock_data()
model = GaussianProcessRegression(sgpr_model(*data))
with pytest.raises(NotImplementedError):
model.covariance_between_points(data[0], data[0])