Research data for "Intermediates of Forming Transition Metal Dichalcogenides Heterostructures Revealed by Machine Learning Simulations"

<p>This dataset supports the paper "Intermediates of Forming Transition Metal Dichalcogenides Heterostructures Revealed by Machine Learning Simulations".&nbsp;</p> <p><strong>Included Files:</strong></p> <ul> <li><strong>ocp_active.zip</strong>: Modified version of ocp (https://github.com/Open-Catalyst-Project/ocp)&nbsp;tailored for active learning applications.</li> <li><strong>deployed.pth</strong>: Pre-trained model used in the experiments.</li> <li><strong>chemiscopy_run.py</strong>: Script integrating the chemiscopy and nequip modules, designed for dataset visualization.</li> <li><strong>new_energy.py</strong>: Modified version of the nequip module, featuring a repulsive potential function.</li> <li><strong>test_datasets.extxyz</strong> &amp; <strong>train_datasets.extxyz</strong>: The test and training datasets in extxyz format.</li> </ul> <p>How to use the modified version of the nequip module:</p> <p>To train this version of the potential function, it is recommended to use nequip&lt;=0.5.6 (on Linux). The NequIP training files need to be updated as follows:</p> <pre><code>model_builders: - new_energy.EnergyModel - StressForceOutput min_bond_len: 1.8</code></pre> <p>Then run:</p> <p><code>export PYTHONPATH=${PYTHONPATH}:$PWD</code><br><code>nequip-train config.yml # Train the potential function</code><br><code>nequip-deploy build --train-dir nequipresultsdir build.pth # Deploy the trained model</code></p>