Learn R Programming

riemtan (version 0.2.5)

Riemannian Metrics for Symmetric Positive Definite Matrices

Description

Implements various Riemannian metrics for symmetric positive definite matrices, including AIRM (Affine Invariant Riemannian Metric, ), Log-Euclidean (), Euclidean, Log-Cholesky (), and Bures-Wasserstein metrics (). Provides functions for computing logarithmic and exponential maps, vectorization, and statistical operations on the manifold of positive definite matrices.

Copy Link

Version

Install

install.packages('riemtan')

Monthly Downloads

279

Version

0.2.5

License

MIT + file LICENSE

Maintainer

Nicolas Escobar

Last Published

November 10th, 2025

Functions in riemtan (0.2.5)

Compute the Inverse Vectorization (AIRM)

Compute the AIRM Exponential

ListBackend Class

DataBackend Abstract Class

CSuperSample Class

Compute the AIRM Vectorization of Tangent Space

Compute the AIRM Logarithm

ParquetBackend Class

TangentImageHandler

TangentImageHandler Class

bures_wasserstein_log

Compute the Bures-Wasserstein Logarithm

bures_wasserstein_exp

Compute the Bures-Wasserstein Exponential

compute_frechet_mean

Compute the Frechet Mean

configure_progress

Configure Progress Handlers

create_progressor

Create a Progress Reporter for Iterative Operations

create_parquet_backend

Create ParquetBackend from Directory

is_progress_available

Check if Progress Reporting is Available

euclidean_unvec

Compute the Inverse Vectorization (Euclidean)

Compute the Euclidean Logarithm

is_parallel_enabled

Check if Parallel Processing is Enabled

Differential of Matrix Logarithm Map

Default reference point

Compute the Euclidean Exponential

Differential of Matrix Exponential Map

Create an Identity Matrix

half_underscore

Half-underscore operation for use in the log-Cholesky metric

log_cholesky_unvec

Compute the Log-Cholesky Inverse Vectorization

log_cholesky_vec

Compute the Log-Cholesky Vectorization

log_euclidean_unvec

Compute the Inverse Vectorization (Euclidean)

log_euclidean_vec

Vectorize at Identity Matrix (Euclidean)

Vectorize at Identity Matrix (Euclidean)

bures_wasserstein_vec

Compute the Bures-Wasserstein Vectorization

bures_wasserstein_unvec

Compute the Bures-Wasserstein Inverse Vectorization

Relocate Tangent Representations to a New Reference Point

reset_parallel_plan

Reset Parallel Plan to Sequential

validate_log_args

Validate arguments for Riemannian logarithms

log_cholesky_exp

Compute the Log-Cholesky Exponential

log_cholesky_log

Compute the Log-Cholesky Logarithm

validate_metric

Validate Metric

spd_isometry_from_identity

Reverse isometry from tangent space at identity to tangent space at P

should_parallelize

Decide Whether to Use Parallel Processing

Metric Object Constructor

Pre-configured Riemannian metrics for SPD matrices

Progress Reporting Utilities for riemtan

validate_tan_imgs

Validate Tangent Images

validate_unvec_args

Validate arguments for inverse vectorization

parallel_config

Parallel Processing Configuration for riemtan

write_connectomes_to_parquet

Write Connectomes to Parquet Files

with_progress_signal

Execute Function with Progress Reporting for Each Item

set_parallel_plan

Set Parallel Processing Plan

Wrapper for the matrix logarithm

Vectorize at Identity Matrix

validate_vec_args

Validate arguments for vectorization

Validate Connections

validate_vec_imgs

Validate Vector Images

validate_exp_args

Validate arguments for Riemannian logarithms

validate_parquet_dir

Validate Parquet Directory Structure

validate_parquet_directory

Validate Parquet Directory

Execute Expression with Progress Reporting

log_euclidean_exp

Compute the Log-Euclidean Exponential

Get Current Number of Parallel Workers

validate_backend

Validate Backend Object

spd_isometry_to_identity

Isometry from tangent space at P to tangent space at identity

riemtan-package

riemtan: Riemannian Metrics for Symmetric Positive Definite Matrices

log_euclidean_log

Compute the Log-Euclidean Logarithm

Generate Random Samples from a Riemannian Normal Distribution