Function test_sample_conditional

tests/gpflow/conditionals/test_multioutput.py:264–320

(whiten: bool, full_cov: bool, full_output_cov: bool)

def test_sample_conditional(whiten: bool, full_cov: bool, full_output_cov: bool) -> None:
    if full_cov and full_output_cov:
        return

    q_mu = tf.random.uniform((Data.M, Data.P), dtype=tf.float64)  # [M, P]
    q_sqrt = tf.convert_to_tensor(
        [np.tril(tf.random.uniform((Data.M, Data.M), dtype=tf.float64)) for _ in range(Data.P)]
    )  # [P, M, M]

    Z = Data.X[: Data.M, ...]  # [M, D]
    Xs: AnyNDArray = np.ones((Data.N, Data.D), dtype=float_type)

    inducing_variable = InducingPoints(Z)
    kernel = SquaredExponential()

    # Path 1
    value_f, mean_f, var_f = sample_conditional(
        Xs,
        inducing_variable,
        kernel,
        q_mu,
        q_sqrt=q_sqrt,
        white=whiten,
        full_cov=full_cov,
        full_output_cov=full_output_cov,
        num_samples=int(1e5),
    )
    value_f = value_f.numpy().reshape((-1,) + value_f.numpy().shape[2:])

    # Path 2
    if full_output_cov:
        pytest.skip(
            "sample_conditional with X instead of inducing_variable does not support full_output_cov"
        )

    value_x, mean_x, var_x = sample_conditional(
        Xs,
        Z,
        kernel,
        q_mu,
        q_sqrt=q_sqrt,
        white=whiten,
        full_cov=full_cov,
        full_output_cov=full_output_cov,
        num_samples=int(1e5),
    )
    value_x = value_x.numpy().reshape((-1,) + value_x.numpy().shape[2:])

    # check if mean and covariance of samples are similar
    np.testing.assert_array_almost_equal(
        np.mean(value_x, axis=0), np.mean(value_f, axis=0), decimal=1
    )
    np.testing.assert_array_almost_equal(
        np.cov(value_x, rowvar=False), np.cov(value_f, rowvar=False), decimal=1
    )
    np.testing.assert_allclose(mean_x, mean_f)
    np.testing.assert_allclose(var_x, var_f)