spectralbrain.diffusion_distance

spectralbrain.diffusion_distance(decomp, t, *, indices=None)

Diffusion distance — multi-scale intrinsic metric.

\[D_t^2(x, y) = \sum_{i=1}^{k} e^{-2\lambda_i t}\, (\varphi_i(x) - \varphi_i(y))^2\]

Small t ≈ geodesic; large t ≈ global diffusion.

Parameters:

• decomp (SpectralDecomposition)

• t (float) – Diffusion time scale.

• indices (ndarray of int, optional)

Returns:

ndarray, shape (M, N) or (N, N)

Return type:

ndarray[tuple[Any, …], dtype[floating]]

References

Coifman RR, Lafon S. Diffusion maps. Applied and Computational Harmonic Analysis 21(1):5–30, 2006.