MLACS

Machine-Learning Assisted Canonical Sampling

MLACS is a tool to sample a canonical distribution of an atomic system with high accuracy and low computational cost. The method is based on a variationnally constructed Machine-Learning Interatomic Potential (MLIP) which is optimized in order to provide the best approximation of the real distribution. The method has been implemented in the form of a python library, linked with the Large-scale Atomic/Molecular Massively Parallel Simulation code.

Contents:

• Background
  • Theory
  • Implementation
• Installation
  • Interfaces
  • Install
  • Python Packages
• Contributing
  • Github repository
  • Code review
  • Coding Rules
• Tutorials
  • Machine-Learning Interatomic Potential
  • Machine-Learning Assisted Sampling
  • Thermodynamic integration

Modules

• User interface
  • Machine-Learning assisted simulations
  • Calc Manager
  • Machine-Learning Interatomic Potential Manager
  • State sampling methods
  • Thermodynamic Integration
  • Properties
  • Utilities

Indices and tables

• Index

• Module Index

• Search Page