micki 2.0.1

micki

A modular, extensible, robust object-oriented microkinetic modeling package written in Python.

DEPENDENCIES (installed automatically, see below):

• Python >= 3.12
• sundials4py (the official SUNDIALS Python interface; >= 7.9, beta)
• numpy >= 2.0
• sympy
• ase

No compiler, SUNDIALS build or LAPACK library is needed: sundials4py ships prebuilt wheels that include SUNDIALS.

Earlier versions that generate Fortran and compile it with f2py are kept on branches: fortran-sundials7 (SUNDIALS 7 Fortran 2003 interface) and sundials4 (SUNDIALS 4.X FCMIX interface).

Installation from PyPI (recommended):

#in an environment with Python >= 3.12 (e.g. a venv or a conda environment)
pip install micki

This installs Micki and all of its dependencies and is probably the simplest option for most users.

Installation with conda (for development):

#fetch Micki and create a conda environment with all dependencies
git clone https://github.com/jrschmidt2/micki.git
cd micki
conda env create -f environment.yml
conda activate micki

Python, numpy, sympy and ase come from conda-forge. sundials4py and Micki itself are not on conda-forge, so environment.yml installs them with pip inside the environment (Micki in editable mode from the clone, so changes take effect immediately).

Testing:

Run the test suite (about 15 seconds; includes a water-gas shift regression test) from the repository root:
python -m unittest discover -s tests -v

Releasing:

Releases are published to PyPI automatically by GitHub Actions (.github/workflows/publish.yml, PyPI trusted publishing):

1. Set the new version in pyproject.toml and commit it.
2. Create a GitHub release with tag v<version> (e.g. v2.0.0).

Publishing the release runs the tests, builds the package and uploads it to PyPI. The workflow refuses to publish if the tag does not match the version. To check the build and tests without publishing, start the workflow by hand ("Run workflow" under Actions -> Publish to PyPI).