airsspy Documentation#
Modern Python interface for AIRSS (Ab initio Random Structure Searching).
New to airsspy? Start here to learn installation and basic usage.
Step-by-step guides for common tasks and workflows.
Understand AIRSS concepts, seed files, and buildcell parameters.
Complete API documentation for all modules and classes.
Reference for the installed ap command and local AIRSS workflows.
Quick Example#
from airsspy import SeedAtoms
# Create a seed for silicon structure search
seed = SeedAtoms('Si8', cell=[5, 5, 5], pbc=True)
seed.gentags.minsep = 2.0
seed.gentags.varvol = 20
# Generate a random structure
random_atoms = seed.build_random_atoms()
What is airsspy?#
airsspy provides a modern, Pythonic interface to AIRSS by:
Extending ASE’s Atoms class with AIRSS-specific features
Managing buildcell subprocess communication
Handling AIRSS/CASTEP
.resfile I/O and lossless RES/extxyz conversionRunning local searches, relaxations, single-points, and packed-file ranking
Deploying jobflow searches and querying stored results
Features#
ASE Integration: Seamless integration with the Atomic Simulation Environment
Modern Python: Type hints, clean API, Python 3.9+ support
Structure Generation: Generate random structures with flexible constraints
RES File Support: Read and write CASTEP .res files
Buildcell Interface: Full access to AIRSS buildcell parameters
CLI Workflows:
ap run,ap rank,ap convert,ap pack, and jobflow database tools
Installation#
Install airsspy using pip:
pip install airsspy
For development installation with all dependencies:
git clone https://github.com/zhubonan/airsspy.git
cd airsspy
pip install -e ".[dev,docs]"
See the Installation Guide for more details.
Quick Links#
Quickstart Tutorial - Your first structure search
Creating Seeds - How to create structure templates
Command Line - Installed
apcommands and examplesAPI Reference - Complete API documentation
GitHub Repository - Source code and issues