Documentation points to unavailable gcc version

[StevenCHowell]

Expected Behavior

The documented gcc version, 7.6.0, is not available from conda-forge.

I was able to find gcc_linux-64=7.5.0 through the conda-forge channel. With that TIGRE successfully built, so I recommend changing the documentation to reflect that. The prior version I tried, 14.3.0, the version I got without specifying a version in my conda environment.yml file led to errors in the build process.

Actual Behavior

Code to reproduce the problem (If applicable)

This is the contents of the environment.yml file for the conda environment that enabled me to successfully install the python version of TIGRE.

name: base
channels:
  - nvidia/label/cuda-12.8.1
  - conda-forge
  - defaults
dependencies:
  - python>=3.7,<3.12
  - cuda
  - cudnn
  - cuda-nvcc
  - gcc_linux-64=7.5.0
  - gxx_linux-64=7.5.0
  - h5py
  - matplotlib
  - networkx
  - numpy
  - pyyaml
  - scipy
  - trimesh
  - tqdm
  - pip
  - pip:
    - python-fcl

Specifications

• MATLAB/python version: 3.11.13
• OS: Ubuntu Linux via the following docker container: mambaorg/micromamba:ubuntu24.04
• CUDA version: 12.8.1

[AnderBiguri]

@StevenCHowell true, that is very out of date, the documentation. Any gcc should work. I try to keep TIGRE such that the latest version of everything works, and I currently have no reason to think it won't.

Thanks for the heads up, I should make it more clear in the docs.