API Reference

Generated from the source docstrings. The reference is intentionally terse — the why lives in Learn, the how in Tutorials and the Gallery.

Everything below is importable from the top-level spectralbrain namespace (e.g. sb.compute_hks), with the full module paths shown for reference.

Subpackages

spectralbrain.core	SpectralBrain core geometric objects — base ops, meshes, point clouds.
spectralbrain.io	SpectralBrain I/O — loaders, exporters, preprocessing, parcellation, and GPU pipeline.
spectralbrain.spectral	SpectralBrain spectral analysis — descriptors, distances, wavelets.
spectralbrain.statistics	SpectralBrain statistics — EDA, frequentist, Bayesian, normative, surrogates, clustering.
spectralbrain.viz	SpectralBrain visualization — 7 modules across 2D stats, 3D brain, geometry, and clustering.
spectralbrain.utils	SpectralBrain utilities — atlases, datasets, and helpers.
spectralbrain.backends	SpectralBrain compute backends — CPU (NumPy/SciPy) and GPU (CuPy/JAX).
spectralbrain.runtime	SpectralBrain runtime infrastructure.

Most-used entry points

Geometry

BrainMesh(vertices, faces[, metadata])	Triangle surface mesh for brain structures.
BrainPointCloud(points[, metadata])	Unstructured 3D point cloud for brain structures.
SpectralDecomposition(eigenvalues, ...[, ...])	Eigenvalues and eigenvectors of a Laplace–Beltrami operator.

Spectral descriptors

compute_shapedna(decomp, *[, normalize, ...])	ShapeDNA — the LBO eigenvalue fingerprint.
compute_hks(decomp[, t_values, n_times, ...])	Heat Kernel Signature — multi-scale per-vertex descriptor.
compute_si_hks(decomp, *[, n_times, ...])	Scale-Invariant HKS — removes scale dependence from HKS.
compute_wks(decomp[, e_values, n_energies, ...])	Wave Kernel Signature — band-pass per-vertex descriptor.
compute_gps(decomp, *[, skip_zero])	Global Point Signature — spectral embedding of the surface.
compute_bates_signatures(decomp[, t_values, ...])	Symmetric polynomial signatures — sign/ordering invariant.
compute_bks(decomp)	Biharmonic Kernel Signature — parameter-free per-vertex scalar.
compute_ibks(decomp, *[, ...])	Improved BKS with curvature-aware neighbourhood aggregation.
sgw_transform(L, scales, *[, signal, ...])	Spectral Graph Wavelet Transform via Chebyshev approximation.
sgw_descriptor(decomp[, scales, n_scales, ...])	Spectral Graph Wavelet descriptor from precomputed eigenpairs.
compute_all_descriptors(decomp, *[, ...])	Compute all 8 spectral descriptors from one decomposition.
anisotropic_laplacian(vertices, faces, *[, ...])	Build an anisotropic Laplacian weighted by curvature direction.

Correspondence & distances

compute_functional_map(decomp_a, decomp_b, *)	Estimate a functional map C between two shapes.
shape_difference_operator(C, *[, type])	Compute a shape-difference operator from a functional map.
wesd(dna_a, dna_b, *[, p, normalize])	Weighted Spectral Distance between two ShapeDNA vectors.
wesd_matrix(dna_collection, *[, p, normalize])	Pairwise WESD matrix for a collection of ShapeDNA vectors.
shapedna_distance(dna_a, dna_b, *[, metric, ...])	Simple distance between two ShapeDNA vectors.
biharmonic_distance(decomp, *[, indices])	Biharmonic distance — parameter-free intrinsic metric.
commute_time_distance(decomp, *[, indices, ...])	Commute-time distance from random walk theory.
diffusion_distance(decomp, t, *[, indices])	Diffusion distance — multi-scale intrinsic metric.
descriptor_distance(desc_a, desc_b, *[, method])	Distance between two descriptor distributions.
build_geometric_connectome(parcel_descriptors, *)	Build a ROI × ROI geometric connectome from parcel descriptors.

I/O

load(path, *[, fmt])	Auto-detect format and load a neuroimaging / geometry file.
load_freesurfer_surface(path)	Load a FreeSurfer surface file.
load_gifti_surface(path)	Load a GIfTI surface file.
load_nifti(path)	Load a NIfTI or MGZ volume.
load_group(files, *[, mode, loader, n_jobs, ...])	Load and stack a cohort for group statistics.
parcellate(*[, subjects_dir, subject_id, ...])	Parcellate a cortical hemisphere into atlas-defined regions.
apply_parcellation(vertices, faces, labels, *)	Split a surface into sub-meshes according to a parcellation.