aenet_env

Python tools for ænet ANN interatomic potentials, including reference-structure preparation, VASP integration, training-data collection, model training, and computational workflow and environment setup.

Documentation

• Installation and environment
  • aenet_env
  • Build and install ænet
  • External requirements
  • Build the bilingual documentation locally
• v0.0.3 workflow scripts
  • b00 — b00_make_poscar.py
  • b01 — b01_stage_structures.py
  • b02 — b02_prepare_vasp.py
  • b03 — b03_submit_vasp_jobs.py
  • c00 — c00_collect_references.py
  • c01 — c01_train_ann.py
  • Stage boundaries
• ANN training parameters and physical meaning
  • Call chain and models
  • Training backends
  • Validation and warnings
• DFT reference structures from RMC configurations
  • Research workflow
  • Local environments and periodic cells
  • Sampling reference candidates
  • Calculation and validation
• aenet_env
  • Layout and responsibilities (v0.0.3)
  • Processing stages
  • Four training interfaces
  • ANN training parameters: physical meaning and configuration
  • Installation
  • Validation boundary
• API reference
  • aenet_env