Installation¶

Temporal Mapper 2 is a flat folder of MATLAB functions — there is no build step, package manager, or installer. You "install" it simply by putting the tmapper_tools/ folder on your MATLAB path.

Requirements¶

Requirement	Notes
MATLAB	Tested on R2024a. Relies on built-in `digraph`, `distances`, `conncomp`, and `adjacency`.
Statistics and Machine Learning Toolbox	Provides `pdist2`, `tiedrank`, `rescale`, `prctile`, `zscore`, and `mode`, which the pipeline uses.

Checking your toolboxes

Run ver at the MATLAB prompt to list installed toolboxes, or license('test','Statistics_Toolbox') to check specifically for the Statistics and Machine Learning Toolbox (returns 1 if available).

Get the code¶

Clone the repository (or download it as a ZIP from GitHub):

git clone https://github.com/Multiscale-Complex-Systems-Lab/tmapper2.git

Add it to your MATLAB path¶

From the repository root, add the toolbox folder to the path. This is the one line every script needs:

addpath("tmapper_tools/")   % (1)!

This is exactly what tmapper_demo.m does at the top. The toolbox is unpackaged (no +namespace), so a plain addpath is all that is required.

To make the toolbox available in every MATLAB session without re-running this, add the same line to your startup.m, using an absolute path to wherever you cloned the repo:

addpath("/path/to/tmapper2/tmapper_tools/")

Verify the install¶

The quickest check is to run the bundled demo from the repository root:

tmapper_demo   % builds a network from the sample data and plots it

If a figure window opens showing a network graph next to a recurrence plot, you are ready to go. Head to the Quickstart for a step-by-step walkthrough of what that demo actually does.

Optional: run the smoke test

A lightweight, dependency-free test of the core pipeline lives in tests/test_core_pipeline.m. Run it from the repository root with:

matlab -batch "run('tests/test_core_pipeline.m')"

It prints All tests passed. on success.