Function _execute_csd_function

mne/time_frequency/csd.py:1302–1384

Estimate cross-spectral density with a given function. This function will apply the given CSD function in parallel across epochs. Parameters ---------- X : array-like, shape (n_epochs, n_channels, n_times) The time series data consisting of n_epochs separate observations

(
    X,
    times,
    frequencies,
    csd_function,
    params,
    n_fft,
    ch_names=None,
    projs=None,
    n_jobs=None,
    *,
    verbose=None,
)

@verbose
def _execute_csd_function(
    X,
    times,
    frequencies,
    csd_function,
    params,
    n_fft,
    ch_names=None,
    projs=None,
    n_jobs=None,
    *,
    verbose=None,
):
    """Estimate cross-spectral density with a given function.

    This function will apply the given CSD function in parallel across epochs.

    Parameters
    ----------
    X : array-like, shape (n_epochs, n_channels, n_times)
        The time series data consisting of n_epochs separate observations
        of signals with n_channels time-series of length n_times.
    times : float
        Timestamps for each sample.
    frequencies : list of float
        The frequencies of interest for which the CSD is going to be computed.
    csd_function : function
        Function that performs the actual CSD computation
    params : list
        List of parameters to pass the CSD function.
    n_fft : int
        Number of FFT points. This is stored in the CSD object.
    ch_names : list of str | None
        A name for each time series. If ``None`` (the default), the series will
        be named 'SERIES###'.
    projs : list of Projection | None
        List of projectors to store in the CSD object. Defaults to ``None``,
        which means the projectors defined in the Epochs object will be copied.
    %(n_jobs)s
    %(verbose)s

    Returns
    -------
    csd : instance of CrossSpectralDensity
        The computed cross-spectral density.
    """
    n_epochs, n_channels, _ = X.shape

    logger.info("Computing cross-spectral density from epochs...")

    n_freqs = len(frequencies)
    csds_mean = np.zeros(
        (n_channels * (n_channels + 1) // 2, n_freqs), dtype=np.complex128
    )

    # Prepare the function that does the actual CSD computation for parallel
    # execution.
    parallel, my_csd, n_jobs = parallel_func(csd_function, n_jobs, verbose=verbose)