Predict energies, forces, and stresses for atomic structures with DeePMD-kit.
npx skills add https://github.com/Hello-QM/catgo-LRG --skill deepmd-inference
Please help me install this Agent Skill. Skill: deepmd-inference Source: https://github.com/Hello-QM/catgo-LRG/tree/main/.claude/skills/deepmd-inference Command: npx skills add https://github.com/Hello-QM/catgo-LRG --skill deepmd-inference
Process atomic structures by performing DeePMD-kit inference to predict energies, forces, and stresses using a trained DP model, enabling rapid evaluation and deployment.
Run inference on a given structure using the trained DP model to obtain energies, forces, and stresses, and compare them to reference data.
High-intent search queries and answers about installing and using this skill.
You can perform DeePMD-kit inference to predict energies, forces, and stresses for atomic structures using a frozen DP model. This requires an installed deepmd-kit and input data in dpdata format or ASE-readable structures.
Yes, you can evaluate a DP model against DFT reference data to assess prediction accuracy. The DeePMD-kit inference process compares predicted energies, forces, and stresses with the provided reference structures.
Yes, DeePMD-kit inference requires a frozen DP model in .pb or .savedmodel format. Additionally, your input atomic structures must be provided in dpdata format or as ASE-readable structures.
Yes, you can use a frozen DP model as an ASE calculator for optimization or NEB workflows. The DeePMD-kit inference provides the necessary energies, forces, and stresses to drive these ASE simulations.
DeePMD-kit inference calculates energies, forces, and stresses for new atomic structures. These predictions are generated by processing the input structures through a frozen DP model.
jinzhezenggroupTrain DeePMD-kit models with SE_E2_A descriptor and PyTorch backend.
learningmatter-mitCalculate NEB activation energy barriers for atomic migration and reactions.
s-choungGenerate ASE Python scripts for atomistic simulation workflows.
Hello-QMCompute phonon spectra, DOS, and thermodynamic data from DFT forces.
