Example Data

Two public example heads you can download into any open project, so you can learn TI-Toolbox and test every page before using your own data. They come from the SimNIBS example dataset (GPL-3.0) and are hosted in idossha/ti-toolbox-example-data; every file is checksum-verified before it is written into your project.

Datasets and parts

Each dataset is one head. Each of its parts is downloaded, and detected, on its own: taking the head model does not oblige you to take the raw MRIs, and deleting the raw MRIs later does not make the head model report itself as missing.

Dataset Part Id Size What lands Meaning
Ernie — SimNIBS example subject Raw MRI ernie/nifti 35 MB sub-ernie/anat/sub-ernie_T1w.nii.gz, _T2w.nii.gz needs pre-processing
  Head model ernie/headmodel 592 MB derivatives/SimNIBS/sub-ernie/m2m_ernie/ ready for optimizer, simulator, analyzer
MNI152 template Raw MRI mni152/nifti 15 MB sub-MNI152/anat/sub-MNI152_T1w.nii.gz needs pre-processing
  Head model mni152/headmodel 462 MB derivatives/SimNIBS/sub-MNI152/m2m_MNI152/ ready for optimizer, simulator, analyzer

A Raw MRI part is what a real study starts from: you build its head model yourself on the Pre-processing page (charm takes 1–2 h). A Head model part is a finished charm run, so the Optimizer, Simulator, Analyzer and the example notebook work the moment it lands — you do not need that subject’s raw MRIs for any of them.

Where to get it

  • First time you open a new project the app asks Add example data? and lists the parts under their dataset, with ernie/headmodel pre-ticked. Download selected queues the ticked parts one after another. The answer (either way) is recorded in the project’s project_status.json, so you are asked once per project, not once per machine.
  • Help ▸ Example data has the same grouping at any time, with a Download button per part, a slim progress bar while one runs, and Installed ✓ (plus a quiet re-download) for what this project already holds. Overview’s Add example data button opens that tab.

  • From a shell or a notebook:

    python -m tit.examples --project /path/to/project --list
python -m tit.examples --project /path/to/project ernie/headmodel mni152/nifti
python -m tit.examples --project /path/to/project ernie          # a bare dataset: all its parts

    from tit.examples import fetch, fetch_ernie
fetch("mni152", "headmodel", PROJECT)   # or fetch("mni152/headmodel", PROJECT)
fetch_ernie(PROJECT)                    # both ernie parts, the shorthand the notebook uses

Every route runs the same tit.examples.fetch and is idempotent: a part already in place is skipped unless you pass --force. Downloading example data is not a job, so it does not appear in the jobs rail, and any project directory, new or long-established, can take a part. In the app, parts download one at a time.