Skip to main content

Create analytical Jacobian matrix source code for chemical kinetics

Project description


DOI Code of Conduct License PyPI Anaconda

This utility creates source code to calculate the Jacobian matrix analytically for a chemical reaction mechanism.


The full documentation for pyJac can be found at

User Group

Further support can be found at our user group, or by opening an issue on our github repo.


Detailed installation instructions can be found in the full documentation. The easiest way to install pyJac is via conda. You can install to your environment with

> conda install -c slackha pyjac

pyJac can also be installed from PyPI using pip:

pip install pyjac

or, using the downloaded source code, installed as a Python module:

> python install


pyJac can be run as a python module:

> python -m pyjac [options]

The generated source code is placed within the out (by default) directory, which is created if it doesn’t exist initially. See the documentation or use python pyjac -h for the full list of options.


Theory, derivations, validation and performance testing can be found in the paper fully describing version 1.0.2 of pyJac:, now published via and available openly via `arXiv:1605.03262 [physics.comp-ph] <>`__.


pyJac is released under the MIT license; see the LICENSE for details.

If you use this package as part of a scholarly publication, please see for the appropriate citation(s).


We welcome contributions to pyJac! Please see the guide to making contributions in the file.

Code of Conduct

In order to have a more open and welcoming community, pyJac adheres to a code of conduct adapted from the Contributor Covenant code of conduct.

Please adhere to this code of conduct in any interactions you have in the pyJac community. It is strictly enforced on all official pyJac repositories, websites, and resources. If you encounter someone violating these terms, please let a maintainer ([@kyleniemeyer]( or [@arghdos](, via email at know and we will address it as soon as possible.


Created by Kyle Niemeyer ( and Nicholas Curtis (

Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

[1.0.6] - 2018-02-21


  • DOI for 1.0.4


  • Syntax errors in
  • Conda install instructions in
  • Corrected TRange columns in parser
  • Minor documentation fixes


  • Add check to reactions to test that all species exist
  • Duplicate warning from falloff->chemically-activated TROE reactions for zero-parameters
  • Add handling of non-unity default third body efficiency


  • Bump internal version to 1.0.5.c

[1.0.5.b0] - 2017-06-02


  • Added usergroup info to README and documentation



  • Now strip whitespace from mechanism file lines prior to parsing keywords


  • Removed plotting scripts specific to first paper on pyJac

1.0.4 - 2017-04-18


  • Adds Travis config for automatic PyPI and conda builds
  • Adds minimal unittest test suite for module imports
  • Adds code of conduct


  • Changed README back to Markdown for simplicity
  • Updated citation instructions

1.0.3 - 2017-04-01


  • Fix for SRI Falloff functions with non-default third bodies (issue #12)
  • Fixed removal of jac/rate lists before libgen of functional_tester
  • Fixed pywrap module import


  • Issue warning in Cantera parsing if the installed version doesn’t have access to species thermo properties.


  • Added significantly more documentation and examples for data ordering, the state vector / Jacobian, and using the python interface

1.0.2 - 2017-01-18


  • Added documentation for libgen / pywrap features


  • Minor compilation fixes for including OpenMP
  • Updated github links to point to SLACKHA / Niemeyer Research Group


  • Shared library creation for CUDA disabled, as CUDA does not allow linkage of SO’s into another CUDA kernel


  • Explicitly conserve mass in PaSR
  • Minor path fixes
  • Division by zero in some TROE parameter cases

1.0.1 - 2016-05-25


  • Added GPU macros, e.g., THREAD_ID, GRID_SIZE


  • Much better handling of removal of files created during testing


  • Bugfix that generates data.bin files correctly from .npy files for performance testing (important)
  • Explicit setting of OpenMP # threads for performance testing

1.0 - 2016-05-07


  • pyJac is now a Python package
  • pyJac can now create a static/shared library for a mechanism (for external linkage)
  • Added documentation
  • Added examples


  • Handles CUDA compilation better via Cython
  • pointers are now restricted where appropriate
  • better Python3 compatibility


  • other minor bugfixes

0.9.1-beta - 2015-10-29


  • Implemented the strict mass conservation formulation
  • Updated CUDA implementation such that it is testable vs. pyJac c-version (and Cantera where applicable)
  • More robust build folder management
  • More robust mapping for strict mass conservation

0.9-beta - 2015-10-02


  • First working / tested version of pyJac

Citation of pyJac


If you use pyJac in a scholarly article, please cite it directly as

Kyle E. Niemeyer and Nicholas J. Curtis (2017). pyJac v1.0.4 [Software]. Zenodo.

A BibTeX entry for LaTeX users is

BibTeX entry:

    author = {Kyle E Niemeyer and Nicholas J Curtis},
    year = 2017,
    title = {{pyJac} v1.0.4},
    doi = {10.5281/zenodo.######},
    url = {},

In both cases, please update the entry with the version used. The DOI for the latest version can be found in the badge at the top. If you would like to cite a specific, older version, the DOIs for each release are:

Project details

Download files

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

Files for pyJac, version 1.0.6
Filename, size File type Python version Upload date Hashes
Filename, size pyjac-1.0.6.tar.gz (244.6 kB) File type Source Python version None Upload date Hashes View
Filename, size pyjac-1.0.6-py2.py3-none-any.whl (130.5 kB) File type Wheel Python version py2.py3 Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page