Skip to main content

Lighthweight but caffeinated Python implementation of computational methods for statistical mechanical calculations of configurational states in crystalline material systems.

Project description

Statistical Mechanics on Lattices

test Codacy Badge status pypi version Static Badge Binder status

Lightweight but caffeinated Python implementation of computational methods for statistical mechanical calculations of configurational states in crystalline materials.

smol is a minimal implementation of computational methods to calculate statistical mechanical and thermodynamic properties of crystalline material systems based on the cluster expansion method from alloy theory and related methods. Although smol is intentionally lightweight---in terms of dependencies and built-in functionality---it has a modular design that closely follows underlying mathematical formalism and provides useful abstractions to easily extend existing methods or implement and test new ones.


smol currently includes the following functionality:

  • Defining cluster expansion functions for a given disordered structure using a variety of available site basis functions with and without explicit redundancy.

  • Option to include explicit electrostatics in expansions using the Ewald summation method.

  • Computing correlation vectors for a set of training structures with a variety of functionality to inspect the resulting feature matrix.

  • Defining fitted cluster expansions for subsequent property prediction.

  • Fast evaluation of correlation vectors and differences in correlation vectors from local updates in order to quickly compute properties and changes in properties for specified supercell sizes.

  • Flexible toolset to sample cluster expansions using Monte Carlo with canonical, semigrand canonical, and charge neutral semigrand canonical ensembles using a Metropolis or a Wang-Landau sampler.

  • Special quasi-random structure generation based on either correlation vectors or cluster interaction vectors.

  • Solving for periodic ground-states of any given cluster expansion with or without electrostatics over a given supercell.

smol is built on top of pymatgen so any pre/post structure analysis can be done seamlessly using the various functionality supported there.


From pypi:

pip install smol

From source:

Clone the repository. The latest tag in the main branch is the stable version of the code. The main branch has the newest tested features, but may have more lingering bugs. From the top level directory

pip install .

Although smol is not tested on Windows platforms, it should still run on Windows since there aren't any platform specific dependencies. The only known installation issue is building pymatgen dependencies. If simply running pip install smol fails, try installing pymatgen with conda first:

conda install -c conda-forge pymatgen
pip install smol

You can also simply use the environment.yml file in the repository to install smol:

conda env create -f environment.yml
source activate smol-env


Refer to the documentation for details on using smol. Going through the example notebooks will also help you get started. You can run the example notebooks interactively in binder.


If you use smol in your research, please give the repo a star :star: and refer to the citing page in the documentation for formal citation information.


We welcome all your contributions with open arms! Please fork and pull request any contributions. See the developing page in the documentation for how to contribute.


The most recent changes are detailed in the change log.

Copyright Notice

Statistical Mechanics on Lattices (smol) Copyright (c) 2022, The Regents
of the University of California, through Lawrence Berkeley National
Laboratory (subject to receipt of any required approvals from the U.S.
Dept. of Energy) and the University of California, Berkeley. All rights reserved.

If you have questions about your rights to use or distribute this software,
please contact Berkeley Lab's Intellectual Property Office at

NOTICE.  This Software was developed under funding from the U.S. Department
of Energy and the U.S. Government consequently retains certain rights.  As
such, the U.S. Government has been granted for itself and others acting on
its behalf a paid-up, nonexclusive, irrevocable, worldwide license in the
Software to reproduce, distribute copies to the public, prepare derivative
works, and perform publicly and display publicly, and to permit others to do so.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

smol-0.5.3.tar.gz (10.2 MB view hashes)

Uploaded source

Built Distributions

smol-0.5.3-cp312-cp312-win_amd64.whl (11.0 MB view hashes)

Uploaded cp312

smol-0.5.3-cp312-cp312-win32.whl (10.9 MB view hashes)

Uploaded cp312

smol-0.5.3-cp311-cp311-win_amd64.whl (11.0 MB view hashes)

Uploaded cp311

smol-0.5.3-cp311-cp311-win32.whl (10.9 MB view hashes)

Uploaded cp311

smol-0.5.3-cp311-cp311-macosx_12_0_arm64.whl (11.2 MB view hashes)

Uploaded cp311

smol-0.5.3-cp310-cp310-win_amd64.whl (11.0 MB view hashes)

Uploaded cp310

smol-0.5.3-cp310-cp310-win32.whl (10.9 MB view hashes)

Uploaded cp310

smol-0.5.3-cp310-cp310-macosx_12_0_arm64.whl (11.2 MB view hashes)

Uploaded cp310

smol-0.5.3-cp39-cp39-win_amd64.whl (11.0 MB view hashes)

Uploaded cp39

smol-0.5.3-cp39-cp39-win32.whl (10.9 MB view hashes)

Uploaded cp39

smol-0.5.3-cp39-cp39-macosx_12_0_arm64.whl (11.2 MB view hashes)

Uploaded cp39

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page