4.1 Environment Setup (ODL/TIGRE)

Goal: get a minimal, reproducible environment to run the Chapter 4 case studies.

---

1) Create a clean Python environment

conda create -n medimg python=3.10
conda activate medimg
pip install numpy scipy matplotlib scikit-image

---

2) Install ODL (recommended)

pip install odl

Quick check:

import odl
space = odl.uniform_discr([-1, -1], [1, 1], (64, 64))
print(space.one().shape)

---

3) TIGRE (optional)

TIGRE installation varies across platforms/CUDA versions. If you don’t want to deal with CUDA yet, you can skip TIGRE and still complete the case studies with CPU-friendly tooling.