About the transport method

Transport is a method for computing inflationary correlation functions. Rather than directly implement the traditional Feynman formulae, based on calculation of wavefunctions and vertex integrals, it evolves the correlation functions themselves. It can be applied to a large class of models, including those with noncanonical kinetic terms. For technical details, see the publications page.

This site collects three tools that implement the transport framework. The tools use different technology and have different use cases, but each one is automated in the sense that it can compute observables directly from a Lagrangian. The PyTransport and CppTransport platforms (written in Python and C++, respectively) are two- and three-point function integrators, currently for canonical—but otherwise arbitrary—inflationary models. mTransport is implemented in Mathematica and is a two-point function integrator for models with non-Euclidean field-space metric.

For more details on the individual codes, read our CppTransport, PyTransport and mTransport pages.


How to cite the transport tools

If you use one or more of these tools as part of a publication or presentation, we ask that you include a reference to the corresponding paper and optionally a link to the transportmethod.com website.

For CppTransport or PyTransport, the main journal article is

Numerical evaluation of the bispectrum in multiple field inflation
Mafalda Dias, Jonathan Frazer, David Mulryne and David Seery
arXiv:1609.00379 | INSPIRE

For mTransport, the main journal article is

Computing observables in curved multifield models of inflation — A guide (with code) to the transport method
Mafalda Dias, Jonathan Frazer and David Seery
arXiv:1502.03125 | INSPIRE
JCAP 1512 (2015) 12 030 | DOI 10.1088/1475-7516/2015/12/030


Licensing

mTransport is licensed under the 3-clause BSD license.

PyTransport is licensed under the GNU General Public License version 3.

CppTransport is licensed under the GNU General Public License version 2 or a later version.

We encourage other researchers to use or adapt these codes as they find appropriate.