Skip to main content

smooth

smooth(
    lr,
    fwhm=None,
    sigma=None,
    space='onavg-ico32',
    mask=None,
    keep_sum=False,
)

Get a smoothing matrix.

Parameters

Name Type Description Default
lr str Hemisphere(s), either ‘l’, ‘r’, or ‘lr’. required
fwhm float Full-width at half-maximum of the Gaussian kernel. None
sigma float Standard deviation of the Gaussian kernel. None
space str Surface space where the data is in. 'onavg-ico32'
mask ndarray or bool or None Mask to apply to the smoothing matrix. If None or False, no mask is applied. If True, the standard cortical mask is used. If an ndarray, it is used as the mask. None
keep_sum bool If True, keep the sum of the data. Useful for transforming area, volume, etc., where the total area/volume is preserved when keep_sum=True. False

Returns

Name Type Description
M sparse matrix Smoothing matrix. Can be applied to data matrix X as X @ M.