JOSS review page Test status Link to conda-forge page

A tiny package to compute the dynamics of stochastic and molecular simulations.

License:BSD 3-clause
Author:Pierre de Buyl


  • performs the computation of mean-square displacements and correlation functions.
  • accepts as input NumPy arrays storing the positions and velocities of particles.
  • implements the so-called Fast Correlation Algorithm proposed by Kneller and others [KKKS95] for the nMOLDYN analysis program.
  • depends only Python and NumPy.

For a quick jump into tidynamics, have a look at the examples.

Goals and plans:

  • Minimal dependencies.
  • Serve as a reference implementation for common algorithms that are useful for molecular and stochastic simulations.
  • Provide later a bit more flexibility to handle cross correlations and many-body systems.


It is necessary to have Python and NumPy to install and use tidynamics.

tidynamics can be installed with pip:

pip install --user tidynamics

or with conda (via conda-forge):

conda install -c conda-forge tidynamics

It is also possible to download the source code and execute the setup.py file.

I ran the tests with Python 2.7, 3.5 and 3.6 and NumPy 1.11 and 1.13. If you encounter any issue, let me know (see Contact, support, and contribution information below).


We use pytest for testing:

python -m pytest

Installing tidynamics does not install the tests. It is necessary to download tidynamics’ source and to install pytest to run the tests.

Contact, support, and contribution information

To contact the author about tidynamics, you can either write an email to Pierre de Buyl or use the issue tracker of the GitHub project.

Bug reports are welcome. If you consider proposing a feature, please keep in mind the goals and plans exposed above.

Contributors are listed in Contributors.