Installation & Extras

Fyron keeps heavy scientific dependencies optional. Start small, then install extras as your workflow grows.

The base installation is enough for many integration workflows. Add extras only when you need the corresponding analysis surface: survival modeling, tabular ML, Excel files, or BOA visualization. This keeps deployment environments smaller and makes optional dependency errors easier to diagnose.

Base Package

bash

uv add fyron

The base package covers FHIR REST, FHIR JSON builders/write-back helpers, SQL helpers, DICOM client wrappers, documents, environment loading, and DataFrame utilities.

All Features

For local workstations, tutorials, full documentation examples, and exploratory clinical research notebooks, install every optional Fyron feature with:

bash

uv add "fyron[all]"

This includes the heavier scientific extras such as survival ML, DICOM SEG, BOA visualization, radiomics, explainability, feature selection, Excel, and S3 support. For servers, CI jobs, and smaller deployments, prefer the targeted extras below.

Extras

Extra	Install	Adds
All features	`uv add "fyron[all]"`	Full local research environment with every optional dependency
Excel	`uv add "fyron[excel]"`	`openpyxl` for `.xlsx` I/O
S3	`uv add "fyron[s3]"`	`boto3` for AWS S3 and S3-compatible object storage
Survival	`uv add "fyron[survival]"`	`lifelines`, `matplotlib`, `scipy`
Survival ML	`uv add "fyron[survival-ml]"`	`scikit-survival` for gradient survival boosting
Statistics	`uv add "fyron[statistics]"`	`numpy`, `scipy` for tests and bootstrap intervals
Descriptive	`uv add "fyron[descriptive]"`	NumPy/SciPy descriptive analysis helpers
DICOM SEG	`uv add "fyron[dicom-seg]"`	`highdicom` for standards-compliant DICOM SEG export
ML	`uv add "fyron[ml]"`	`numpy`, `scikit-learn`, `xgboost`, `matplotlib`, `joblib`
Feature Selection	`uv add "fyron[feature-selection]"`	statistical/model feature selection, including Boruta
Validation	`uv add "fyron[validation]"`	NumPy/scikit-learn validation helpers
Explainability	`uv add "fyron[explainability]"`	sklearn model inspection and SHAP tables
Reporting	`uv add "fyron[reporting]"`	reporting helper alias; no extra dependency
Audit	`uv add "fyron[audit]"`	provenance helper alias; no extra dependency
Boruta	`uv add "fyron[ml,boruta]"`	Boruta feature selection
Visualization	`uv add "fyron[visualization]"`	OpenCV for BOA collages
Radiomics	`uv add "fyron[radiomics]"`	PyRadiomics for BOA mask feature extraction

Recommended Analysis Install

bash

uv add "fyron[all]"

Working From A Clone

bash

git clone https://github.com/bitsandflames/fyron.git
cd fyron
uv sync --all-extras --dev
uv run pytest

This is mainly useful if you want to contribute, run the test suite, or inspect examples from the source tree.

Preview Documentation Locally

Fyron's docs are built from the Markdown files in docs/. If you are editing the docs, preview them locally with:

bash

uv run python scripts/build_docs.py
uv run python -m http.server 8001 --directory site