Machine-Learning Force Field Data for Janus PtSTe Thermal Conductivity Study

Code and data accompany a published study on the thermal conductivity of Janus PtSTe van der Waals heterostructures. The package includes a neural network force field code, first-principles DFT training data in ASE JSON format, and trained model parameters. Author Jesús Carrete from TU Wien produced this research artifact.

Use Cases

• Training machine-learning force fields based on the provided first-principles DFT data.
• Predicting thermal conductivity of Janus PtSTe heterostructures using the supplied trained model parameters.
• Benchmarking neural-network force field uncertainty methods like ensembles vs committees.
• Applying the differentiable force-field code to other ionic liquid or material systems.

Strengths

• Includes the specific force-field code version used in the published manuscript.
• Contains first-principles DFT data stored in the structured ASE JSON database format.
• Provides saved parameters for a fully trained force field model.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

First-principles Density Functional Theory (DFT) calculations.