Sovereign 8s.is mirror of standardgalactic/PCNtoolkit
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-03-28 16:05:16 +01:00
.github cos - remove redundant comment 2026-03-28 15:46:57 +01:00
doc fix - small fix 2026-03-18 16:33:22 +01:00
examples fix: correct harmonize kwarg to harmonize_data in example notebooks 2026-02-11 21:46:53 +05:30
pcntoolkit fix - typo in import 2026-03-28 14:21:49 +01:00
test fix - clip before fitting 2026-03-05 17:09:47 +01:00
.gitattributes 🎊 Added .gitattributes & .gitignore files 2016-08-18 13:11:43 +02:00
.gitignore Re-render citation file and docs 2025-11-14 15:53:28 +01:00
.readthedocs.yaml Updated the docs 2025-06-30 15:33:59 +02:00
CHANGES enh - prepare release 2026-03-28 16:02:43 +01:00
CITATION.cff enh - prepare release 2026-03-28 16:02:43 +01:00
citation.cff.in enh - update CITATION.cff with all the contributors 2026-03-24 14:53:46 +01:00
CONTRIBUTING.md Added new version 2025-06-30 14:53:23 +02:00
LICENSE Added new version 2025-06-30 14:53:23 +02:00
Makefile fix - specify Python version in dev-setup for conda environment 2026-02-24 11:54:26 +01:00
pyproject.toml enh - prepare release 2026-03-28 16:02:43 +01:00
pytest.ini Added new version 2025-06-30 14:53:23 +02:00
README.md Updated the docs 2025-06-30 15:33:59 +02:00
render_citation.py enh - render CITATION.cff with all the contributors 2026-03-24 14:50:52 +01:00
warning_banner.png Added new version 2025-06-30 14:53:23 +02:00

Predictive Clinical Neuroscience Toolkit

Predictive Clinical Neuroscience software toolkit (formerly nispat).

A Python package for normative modelling, spatial statistics and pattern recognition.

IMPORTANT

Deprecation warning

This is PCNtoolkit version 1.X.X, released originally in June 2025. Any scripts, models, and results created with version 0.X.X are not compatible with this and future versions of the toolkit.

To use the models created with versions 0.35 and earlier, please install the appropriate version using pip install pcntoolkit==0.35, or replace 0.35 with your desired version. The old version of the toolbox is also still available on GitHub.

Installation

pip install pcntoolkit

Documentation

See the documentation for more details.

Documentation for the earlier version of the toolbox is available here

Example usage

from pcntoolkit import {load_fcon, BLR, NormativeModel}

fcon1000 = load_fcon()

train, test = fcon1000.train_test_split()

# Create a BLR model with heteroskedastic noise
model = NormativeModel(BLR(heteroskedastic=True), 
                       inscaler='standardize', 
                       outscaler='standardize')

model.fit_predict(train, test)