3. Reference documentation: all fmralign functions

This is the class and function reference of fmralign. Please refer to the full user guide for further details, as the class and function raw specifications may not be enough to give full guidelines on their uses.

3.1. fmralign.pairwise_alignment: Pairwise Alignment

Module for pairwise functional alignment

Classes:

PairwiseAlignment(alignment_method[, ...])

Decompose the source and target images into regions and align corresponding regions independently.

3.2. fmralign.template_alignment: Template Alignment

Module for functional template inference. Uses functional alignment on Niimgs and predicts new subjects’ unseen images.

Classes:

TemplateAlignment([alignment_method, ...])

Decompose the source images into regions and summarize subjects information in a template, then use pairwise alignment to predict new contrast for target subject.

3.3. fmralign.alignment_methods: Alignment Methods

Module implementing alignment estimators on ndarrays.

Classes:

Identity()

Compute no alignment, used as baseline for benchmarks : RX = X.

DiagonalAlignment([n_jobs, parallel_backend])

Compute the voxelwise projection factor between X and Y.

ScaledOrthogonalAlignment([scaling])

Compute a orthogonal mixing matrix R and a scaling sc.

RidgeAlignment([alphas, cv])

Compute a scikit-estimator R using a mixing matrix M s.t Frobenius norm || XM - Y ||^2 + alpha * ||M||^2 is minimized with cross-validation

Hungarian()

Compute the optimal permutation matrix of X toward Y

OptimalTransportAlignment([metric, reg, ...])

Compute the optimal coupling between X and Y with entropic regularization using a OTT-JAX as a backend for acceleration.