def test_acquisition():
n_params = 2
n = 10
n2 = 5
parameter_names = ['a', 'b']
bounds = {'a': [-2, 3], 'b': [5, 6]}
target_model = GPyRegression(parameter_names, bounds=bounds)
x1 = np.random.uniform(*bounds['a'], n)
x2 = np.random.uniform(*bounds['b'], n)
x = np.column_stack((x1, x2))
y = np.random.rand(n)
target_model.update(x, y)
# check acquisition without noise
acq_noise_var = 0
t = 1
acquisition_method = acquisition.LCBSC(target_model, noise_var=acq_noise_var)
new = acquisition_method.acquire(n2, t=t)
assert np.allclose(new[1:, 0], new[0, 0])
assert np.allclose(new[1:, 1], new[0, 1])
# check acquisition with scalar noise
acq_noise_var = 2
t = 1
acquisition_method = acquisition.LCBSC(target_model, noise_var=acq_noise_var)
new = acquisition_method.acquire(n2, t=t)
assert new.shape == (n2, n_params)
assert np.all((new[:, 0] >= bounds['a'][0]) & (new[:, 0] <= bounds['a'][1]))
assert np.all((new[:, 1] >= bounds['b'][0]) & (new[:, 1] <= bounds['b'][1]))
# check acquisition with separate variance for dimensions
acq_noise_var = np.random.uniform(0, 5, size=2)
t = 1
acquisition_method = acquisition.LCBSC(target_model, noise_var=acq_noise_var)
new = acquisition_method.acquire(n2, t=t)
assert new.shape == (n2, n_params)
assert np.all((new[:, 0] >= bounds['a'][0]) & (new[:, 0] <= bounds['a'][1]))
assert np.all((new[:, 1] >= bounds['b'][0]) & (new[:, 1] <= bounds['b'][1]))
# check acquisition with arbitrary covariance matrix
acq_noise_cov = np.random.rand(n_params, n_params) * 0.5
acq_noise_cov += acq_noise_cov.T
acq_noise_cov += n_params * np.eye(n_params)
t = 1
with pytest.raises(ValueError):
acquisition.LCBSC(target_model, noise_var=acq_noise_cov)
# test Uniform Acquisition
t = 1
acquisition_method = acquisition.UniformAcquisition(target_model, noise_var=acq_noise_var)
new = acquisition_method.acquire(n2, t=t)
assert new.shape == (n2, n_params)
assert np.all((new[:, 0] >= bounds['a'][0]) & (new[:, 0] <= bounds['a'][1]))
assert np.all((new[:, 1] >= bounds['b'][0]) & (new[:, 1] <= bounds['b'][1]))
def _get_dependencies_acq_fn():
"""Provide the requirements for the MaxVar-based acquisition function initialisation.
Returns
-------
(GPy.model.GPRegression, elfi.methods.utils.ModelPrior)
Tuple containing a fit gp and a prior.
"""
mean = [4, 4]
cov_matrix = [[1, .5], [.5, 1]]
names_param = ['mu_0', 'mu_1']
eps_prior = 5 # The prior's range indicator used in the Gaussian noise model.
bounds_param = {
'mu_0': (mean[0] - eps_prior, mean[0] + eps_prior),
'mu_1': (mean[1] - eps_prior, mean[1] + eps_prior)
}
# Initialising the prior.
gm_2d = elfi.examples.gauss.get_model(true_params=mean,
nd_mean=True,
cov_matrix=cov_matrix)
prior = ModelPrior(gm_2d)
# Generating the coordinates and the values of the fitting data.
n_pts_fit = 10
x1 = np.random.uniform(*bounds_param['mu_0'], n_pts_fit)
x2 = np.random.uniform(*bounds_param['mu_1'], n_pts_fit)
x = np.column_stack((x1, x2))
y = np.random.rand(n_pts_fit)
# Fitting the gp with the generated points.
gp = GPyRegression(names_param, bounds=bounds_param)
gp.update(x, y)
return gp, prior