Introduction
XANESNET Overview
XANESNET is an open-source platform for the rapid and automated analysis and prediction of X-ray spectroscopy data. The platform provides machine learning solution for both forward and reverse mapping problems, linking material properties or structures with their corresponding X-ray Absorption Near Edge Structure (XANES) spectra.
The forward mapping approach is similar to the approach used by computational researchers in the sense that an input structure is used to generate a spectral observable. In this area the objective of XANESNET is to supplement and support analysis provided by first principles quantum mechanical simulations. The reverse mapping problem, on the other hand, focuses on interpreting the material property from XANES spectra. This is perhaps the more natural of the two, as it has a clear connection to the problem that X-ray spectroscopists face day-to-day in their work: how can a measurement/observable be interpreted?
XANESNET approaches this problem by employing machine learning models, including a variety of deep neural network architectures to perform inverse analysis. By training these models on large datasets, XANESNET can accurately predict the characteristics of unknown samples. The software is designed in a modular, plugin-based framework, enabling users to customise their workflows by selecting appropriate models, structural descriptors, and other components based on their specific requirements.
XANESNET features
GPLv3 licensed open-source distribution
Automated data processing: Fourier transform, Gaussian transform
Feature extraction: wACSF, RDC, pDOS, MACE
Neural network architecture: MLP, CNN, GNN, LSTM, Autoencoder, AE-GAN, Multihead, Transformer, EnvEmbed
Learning scheme: standard, K-fold, ensemble learning, bootstrapping
Experiment tracking and visualisation: MLFlow, TensorBoard
Learning rate scheduler
Custom ML workflow components and run via input file
Easy to extend with new components
Web interface
XANESNET development team
XANESNET is developed by the Penfold Group and the Research Software Engineering (RSE) team at Newcastle University.
Publications
XANESNET:
A Deep Neural Network for the Rapid Prediction of X-ray Absorption Spectra - C. D. Rankine, M. M. M. Madkhali, and T. J. Penfold, J. Phys. Chem. A, 2020, 124, 4263-4270.
Accurate, affordable, and generalizable machine learning simulations of transition metal x-ray absorption spectra using the XANESNET deep neural network - C. D. Rankine, and T. J. Penfold, J. Chem. Phys., 2022, 156, 164102.
Extension to X-ray Emission:
A deep neural network for valence-to-core X-ray emission spectroscopy - T. J. Penfold, and C. D. Rankine, Mol. Phys., 2022, e2123406.
The Applications:
On the Analysis of X-ray Absorption Spectra for Polyoxometallates - E. Falbo, C. D. Rankine, and T. J. Penfold, Chem. Phys. Lett., 2021, 780, 138893.
Enhancing the Anaysis of Disorder in X-ray Absorption Spectra: Application of Deep Neural Networks to T-Jump-X-ray Probe Experiments - M. M. M. Madkhali, C. D. Rankine, and T. J. Penfold, Phys. Chem. Chem. Phys., 2021, 23, 9259-9269.
Miscellaneous:
The Role of Structural Representation in the Performance of a Deep Neural Network for X-ray Spectroscopy - M. M. M. Madkhali, C. D. Rankine, and T. J. Penfold, Molecules, 2020, 25, 2715.
Citing XANESNET
@software{xanesnet,
author = {Penfold Group, Newcastle University},
title = {XANESNET},
url = {https://gitlab.com/team-xnet/xanesnet},
date = {2023},
}