Exemplo n.º 1
0
def test_continuous_y():
    for inference_method in get_installed(["lp", "ad3"]):
        X, Y = generate_blocks(n_samples=1)
        x, y = X[0], Y[0]
        w = np.array([1, 0, 0, 1, 0, -4, 0])  # unary  # pairwise

        crf = GridCRF(inference_method=inference_method)
        crf.initialize(X, Y)
        joint_feature = crf.joint_feature(x, y)
        y_cont = np.zeros_like(x)
        gx, gy = np.indices(x.shape[:-1])
        y_cont[gx, gy, y] = 1
        # need to generate edge marginals
        vert = np.dot(y_cont[1:, :, :].reshape(-1, 2).T, y_cont[:-1, :, :].reshape(-1, 2))
        # horizontal edges
        horz = np.dot(y_cont[:, 1:, :].reshape(-1, 2).T, y_cont[:, :-1, :].reshape(-1, 2))
        pw = vert + horz

        joint_feature_cont = crf.joint_feature(x, (y_cont, pw))
        assert_array_almost_equal(joint_feature, joint_feature_cont)

        const = find_constraint(crf, x, y, w, relaxed=False)
        const_cont = find_constraint(crf, x, y, w, relaxed=True)

        # djoint_feature and loss are equal:
        assert_array_almost_equal(const[1], const_cont[1], 4)
        assert_almost_equal(const[2], const_cont[2], 4)

        # returned y_hat is one-hot version of other
        if isinstance(const_cont[0], tuple):
            assert_array_equal(const[0], np.argmax(const_cont[0][0], axis=-1))

            # test loss:
            assert_almost_equal(crf.loss(y, const[0]), crf.continuous_loss(y, const_cont[0][0]), 4)
Exemplo n.º 2
0
def test_continuous_y():
    for inference_method in get_installed(["lp", "ad3"]):
        X, Y = generate_blocks(n_samples=1)
        x, y = X[0], Y[0]
        w = np.array([
            1,
            0,  # unary
            0,
            1,
            0,  # pairwise
            -4,
            0
        ])

        crf = GridCRF(inference_method=inference_method)
        crf.initialize(X, Y)
        joint_feature = crf.joint_feature(x, y)
        y_cont = np.zeros_like(x)
        gx, gy = np.indices(x.shape[:-1])
        y_cont[gx, gy, y] = 1
        # need to generate edge marginals
        vert = np.dot(y_cont[1:, :, :].reshape(-1, 2).T,
                      y_cont[:-1, :, :].reshape(-1, 2))
        # horizontal edges
        horz = np.dot(y_cont[:, 1:, :].reshape(-1, 2).T,
                      y_cont[:, :-1, :].reshape(-1, 2))
        pw = vert + horz

        joint_feature_cont = crf.joint_feature(x, (y_cont, pw))
        assert_array_almost_equal(joint_feature, joint_feature_cont)

        const = find_constraint(crf, x, y, w, relaxed=False)
        const_cont = find_constraint(crf, x, y, w, relaxed=True)

        # djoint_feature and loss are equal:
        assert_array_almost_equal(const[1], const_cont[1], 4)
        assert_almost_equal(const[2], const_cont[2], 4)

        # returned y_hat is one-hot version of other
        if isinstance(const_cont[0], tuple):
            assert_array_equal(const[0], np.argmax(const_cont[0][0], axis=-1))

            # test loss:
            assert_almost_equal(crf.loss(y, const[0]),
                                crf.continuous_loss(y, const_cont[0][0]), 4)
Exemplo n.º 3
0
def test_loss_augmentation():
    X, Y = generate_blocks(n_samples=1)
    x, y = X[0], Y[0]
    w = np.array([1, 0, 0, 1, 0, -4, 0])  # unary  # pairwise
    crf = GridCRF()
    crf.initialize(X, Y)
    y_hat, energy = crf.loss_augmented_inference(x, y, w, return_energy=True)

    assert_almost_equal(energy + crf.loss(y, y_hat), -np.dot(w, crf.joint_feature(x, y_hat)))
Exemplo n.º 4
0
def test_energy_lp():
    # make sure that energy as computed by ssvm is the same as by lp
    np.random.seed(0)
    found_fractional = False
    for inference_method in get_installed(["lp", "ad3"]):
        crf = GridCRF(n_states=3, n_features=4, inference_method=inference_method)
        while not found_fractional:
            x = np.random.normal(size=(2, 2, 4))
            w = np.random.uniform(size=crf.size_joint_feature)
            inf_res, energy_lp = crf.inference(x, w, relaxed=True, return_energy=True)
            assert_almost_equal(energy_lp, -np.dot(w, crf.joint_feature(x, inf_res)))
            found_fractional = np.any(np.max(inf_res[0], axis=-1) != 1)
Exemplo n.º 5
0
def test_loss_augmentation():
    X, Y = generate_blocks(n_samples=1)
    x, y = X[0], Y[0]
    w = np.array([1, 0,  # unary
                  0, 1,
                  0,     # pairwise
                  -4, 0])
    crf = GridCRF()
    crf.initialize(X, Y)
    y_hat, energy = crf.loss_augmented_inference(x, y, w, return_energy=True)

    assert_almost_equal(energy + crf.loss(y, y_hat),
                        -np.dot(w, crf.joint_feature(x, y_hat)))
Exemplo n.º 6
0
def test_energy_lp():
    # make sure that energy as computed by ssvm is the same as by lp
    np.random.seed(0)
    found_fractional = False
    for inference_method in get_installed(["lp", "ad3"]):
        crf = GridCRF(n_states=3, n_features=4,
                      inference_method=inference_method)
        while not found_fractional:
            x = np.random.normal(size=(2, 2, 4))
            w = np.random.uniform(size=crf.size_joint_feature)
            inf_res, energy_lp = crf.inference(x, w, relaxed=True,
                                               return_energy=True)
            assert_almost_equal(energy_lp,
                                -np.dot(w, crf.joint_feature(x, inf_res)))
            found_fractional = np.any(np.max(inf_res[0], axis=-1) != 1)