get_padded_borders

seas.domains.get_padded_borders(domain_ROIs: numpy.ndarray, blur: int, rois: dict, n_roi_rotations: int = 0, bounding_box: Optional[bool] = None)[source]

Since a gaussian blur between a value with a number and a nan border will result in an increasingly large nan border, get this new border roi boundary. Rois are used to separate hemispheres rather than blur together.

Parameters

  • domain_ROIs – The domain map built by get_domain_map.

  • blur – True if background values should be nan, otherwise background values are 0.

  • rois – The roi dict used to generate the roimask. This should be saved in components[‘expmeta’] during decomposition.

  • n_roi_rotations – The number of times the rois were rotated when loading.

  • bounding_box – The bounding box applied before ICA decomposition.

Returns

ROI_timecourses:

A np array of shape (n_domains, t), with the mean of a given domain, i, under ROI_timecourses[i].

mean_filtered:

The frame mean, of shape (t), filtered by the default filtering method.

Return type

output, a dictionary with the following keys