Dataset for "AtomProNet: Data flow to and from machine learning interatomic potentials in materials science"

This dataset supports the paper “AtomProNet: Data Flow To and From Machine Learning Interatomic Potentials in Materials Science”.Access the paper here: https://arxiv.org/abs/2501.14039 The MACE folder includes: mace_tutorial.ipynb – A Jupyter Notebook with installation instructions and training workflow. Datasets/ – Contains training, validation, and testing datasets used for Al₂O₃. The Allegro folder includes: Allegro_model_Nov2024.ipynb – A Jupyter Notebook with Allegro installation details. tutorial.yaml – Allegro configuration file. MLIP_data_16k.extxyz – Dataset used for Al₂O₃ MLIP training, validation, and testing.本数据集支撑论文《AtomProNet：材料科学领域机器学习原子间势（Machine Learning Interatomic Potentials，MLIP）的数据流研究》，可通过以下链接获取论文：https://arxiv.org/abs/2501.14039 MACE 文件夹包含以下内容： mace_tutorial.ipynb：一份涵盖安装指南与训练流程的Jupyter笔记本。 Datasets/：存放用于Al₂O₃训练、验证与测试的数据集。 Allegro 文件夹包含以下内容： Allegro_model_Nov2024.ipynb：一份包含Allegro安装细节的Jupyter笔记本。 tutorial.yaml：Allegro配置文件。 MLIP_data_16k.extxyz：用于Al₂O₃的机器学习原子间势训练、验证与测试的数据集。