Installation¶

SPM-Kit requires Python 3.11 or newer. The numerical Core is headless; Fathom adds PyQt6, pyqtgraph and plotting dependencies through the gui extra.

Choose a version deliberately¶

Source	Version on 29 July 2026	Command	Use when
PyPI	`0.1.2`	`python -m pip install spmkit`	You specifically need the older published wheel
GitHub release	`0.1.4`	`python -m pip install "spmkit @ git+https://github.com/kegouro/spmkit@v0.1.4"`	You need the latest tagged source release
Current development	`0.1.5.dev0`	`python -m pip install "spmkit[gui] @ git+https://github.com/kegouro/spmkit@main"`	You need the behavior documented on this development site

PyPI lags the GitHub release. A bare pip install spmkit does not currently produce the release or development versions documented here.

Core and CLI¶

python -m venv .venv
source .venv/bin/activate
python -m pip install "spmkit @ git+https://github.com/kegouro/spmkit@v0.1.4"
spmkit --version
spmkit --help

On Windows PowerShell activate with .venv\Scripts\Activate.ps1.

Fathom¶

python -m pip install "spmkit[gui] @ git+https://github.com/kegouro/spmkit@main"
python -c "from spmkit.gui.app_workspace import build_workspace; print('Fathom import OK')"
spmkit gui

Fathom requires a display. QT_QPA_PLATFORM=offscreen is for automated GUI tests, not a useful interactive session.

Optional extras¶

Extra	Adds
`gui`	Fathom: PyQt6, pyqtgraph and visualization support
`gwy`	`.gwy` read/write through `gwyfile`
`nanosurf`	`.nhf` support through NSFopen
`afm`	optional long-tail readers through `afmformats`
`jpk`	JPK TIFF support through `tifffile`
`hdf5`	HDF5 export/read dependencies
`grains`	SciPy-backed grain analysis
`viz`	publication figures and scientific colormaps
`report`	HTML/PDF reporting dependencies
`all`	all runtime extras listed in `pyproject.toml`

Editable development install¶

git clone https://github.com/kegouro/spmkit.git
cd spmkit
python -m pip install -e ".[dev,gui,hdf5,test-gui]"
spmkit --version
pytest --collect-only -q --no-cov

The companion repositories are separate packages. See the ecosystem installation matrix before attempting an evidence-discovery or validation workflow.

Verify the numerical boundary¶

python - <<'PY'
from spmkit.core.analysis.roughness import statistics
from spmkit.core.models import SPMChannel
import numpy as np

channel = SPMChannel("Height", np.array([[0.0, 1.0], [2.0, 3.0]]), "nm", 1e-6, 1e-6)
result = statistics(channel)
print(result.Sa, result.Sq, result.Sz, result.unit)
PY

This verifies import and a public numerical function without requiring an instrument file. It does not validate your instrument calibration or establish physical metrological traceability.

First analysis