Installing XANESNET

Download

XANESNET can be cloned from GitHub or GitLab:

git clone https://github.com/NewcastleRSE/XANESNET.git

or

git clone https://gitlab.com/team-xnet/xanesnet.git

The repository contains source code, example configs under configs/, and toy data under data/.

Training datasets for first-row transition-metal X-ray absorption and emission can be obtained separately:

git clone https://gitlab.com/team-xnet/training-sets.git

Requirements

  • Linux (tested on Ubuntu)

  • Python 3.12 and above

  • A C/C++ build toolchain if pip needs to compile native PyTorch Geometric extensions.

  • Optional: NVIDIA GPU and CUDA-compatible PyTorch wheels for device: cuda.

The Lightning trainer is an optional extra. Install it with python -m pip install -e ".[lightning]" when using trainer_type: lightning. The tested Lightning environment uses Python 3.12.9 and PyTorch 2.5.1 with CUDA 12.4. CPU-only training does not require CUDA. MPS devices are currently unsupported and untested.

Installation

Install the package from the XANESNET repository root:

python3.12 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip setuptools wheel
python -m pip install -e .

For a non-editable install:

python -m pip install .

To use the optional Lightning trainer, include its extra:

python -m pip install -e ".[lightning]"

Verify the installation:

xanesnet --help

Continue with Running XANESNET for command-line workflows and Input files for configuration files.

For the work-in-progress browser-based config editor, see Configuration editor.

PyTorch and PyTorch Geometric

XANESNET depends on PyTorch and PyTorch Geometric. On Linux with CPU-only PyTorch, a normal pip install is often enough.

For CUDA, or if packages such as torch-scatter, torch-sparse, or torch-cluster fail to install, install matching wheels for your Python, PyTorch, and CUDA versions first, then install XANESNET.

Example for PyTorch 2.5 with CUDA 12.4:

python -m pip install torch --index-url https://download.pytorch.org/whl/cu124
python -m pip install torch-scatter torch-sparse torch-cluster -f https://data.pyg.org/whl/torch-2.5.1+cu124.html
python -m pip install -e .

Adjust the wheel URLs for your target platform. The PyTorch and PyTorch Geometric installation guides are the source of truth for the current compatibility matrix.