lls¶

lls is a linear least-squares toolkit for fitting 2-tissue compartment models to dynamic PET data with PyTorch.

Start here¶

• If you already have TAC, AIF, and time data in your own files, start with Use your own TAC, AIF, and time data.
• If you want the API reference and model choices, read Fitting models.
• If you are validating solver changes, use Simulation pipeline and Prediction and benchmarking.

What is in the repository¶

• core fitting functions for 2-tissue compartment models
• a NumPy-friendly entrypoint for regular use
• low-level Torch solvers for direct control
• simulation and prediction scripts for validation work

Documentation map¶

• Use your own TAC, AIF, and time data is the shortest path for collaborators who already have data.
• Installation explains package install options and how to preview the docs locally.
• Fitting models covers the main API, model variants, and common usage patterns.
• Simulation pipeline documents how to generate validation datasets.
• Prediction and benchmarking explains how to fit those datasets and summarize results.
• Self-hosting the docs shows a Compose and Caddy-friendly deployment pattern.