Function test_XdawnTransformer

mne/preprocessing/tests/test_xdawn.py:239–311

Test XdawnTransformer.

()

def test_XdawnTransformer():
    """Test XdawnTransformer."""
    pytest.importorskip("sklearn")
    # Get data
    raw, events, picks = _get_data()
    raw.del_proj()
    epochs = Epochs(
        raw,
        events,
        event_id,
        tmin,
        tmax,
        picks=picks,
        preload=True,
        baseline=None,
        verbose=False,
    )
    X = epochs._data
    y = epochs.events[:, -1]
    # Fit
    xdt = XdawnTransformer()
    xdt.fit(X, y)
    pytest.raises(ValueError, xdt.fit, X, y[1:])
    pytest.raises(ValueError, xdt.fit, "foo")

    # Provide covariance object
    signal_cov = compute_raw_covariance(raw, picks=picks)
    xdt = XdawnTransformer(signal_cov=signal_cov)
    xdt.fit(X, y)
    # Provide ndarray
    signal_cov = np.eye(len(picks))
    xdt = XdawnTransformer(signal_cov=signal_cov)
    xdt.fit(X, y)
    # Provide ndarray of bad shape
    signal_cov = np.eye(len(picks) - 1)
    xdt = XdawnTransformer(signal_cov=signal_cov)
    pytest.raises(ValueError, xdt.fit, X, y)
    # Provide another type
    signal_cov = 42
    xdt = XdawnTransformer(signal_cov=signal_cov)
    pytest.raises(ValueError, xdt.fit, X, y)

    # Fit with y as None
    xdt = XdawnTransformer()
    xdt.fit(X)

    # Compare xdawn and XdawnTransformer
    xd = Xdawn(correct_overlap=False)
    xd.fit(epochs)

    xdt = XdawnTransformer()
    xdt.fit(X, y)
    # Subset filters
    xdt_filters = xdt._subset_multi_components()
    assert_array_almost_equal(
        xd.filters_["cond2"][:2, :], xdt_filters.reshape(2, 2, 8)[0]
    )