Convert computational chemistry data formats to DeePMD-ready training data.
npx skills add https://github.com/Hello-QM/catgo-LRG --skill dpdata
Please help me install this Agent Skill. Skill: dpdata Source: https://github.com/Hello-QM/catgo-LRG/tree/main/server/catgo/workflow/skills/data/dpdata Command: npx skills add https://github.com/Hello-QM/catgo-LRG --skill dpdata
dpdata simplifies and automates the conversion of computational chemistry data formats, enabling seamless preparation of training data for machine learning potentials.
Convert VASP outputs into a DeePMD-ready training dataset using the dpdata converter.
High-intent search queries and answers about installing and using this skill.
You can convert VASP OUTCAR or CONTCAR files to DeePMD training data using the dpdata package, which reads computational chemistry outputs and writes DeePMD-ready npy or raw data formats for machine learning workflows.
Supported computational chemistry file formats for machine learning data preparation include VASP, QE, CP2K, Gaussian, and LAMMPS. The dpdata package reads these formats and converts them into DeePMD-ready training datasets.
Yes, you can merge and filter trajectory data before generating a DeePMD dataset. The dpdata package provides capabilities to read, merge, split, and filter trajectory data to curate precise training datasets for machine learning potentials.
Yes, dpdata supports both command line and Python API workflows for format conversion. You can execute computational chemistry data transformations to DeePMD-ready formats using either the CLI or Python scripts.
The best way to prepare computational chemistry data for DeePMD machine learning potentials is using the dpdata package to convert outputs from VASP, QE, CP2K, Gaussian, or LAMMPS into structured DeePMD npy or raw training datasets.
jinzhezenggroupConvert atomic simulation data between VASP, LAMMPS, QE, CP2K, and ABACUS formats.
Hello-QMTrain DeePMD-kit machine learning potentials with DPA-3 and se_e2_a descriptors.
jinzhezenggroupTrain DeePMD-kit models with SE_E2_A descriptor and PyTorch backend.
jinzhezenggroupLabel dpdata Systems with driver plugins to compute energies, forces, and virials.
