Overview of the cythonic components

As we described in the conceptual overview, RidePy enables the users to run a simulation using either pure pythonic components, or using cythonic components. Using a dispatcher implemented in C++ and exposed to Python using Cython can dramatically speed up the simulation runs.

In order to make it happen, RidePy contains C++ implementations of all its core components, i.e.

• data_structures

• spaces

• and dispatchers,

as well as Cython wrappers thereof.

The module structure of the cythonic components mimics exactly that of their pythonic counterparts, so that in most cases, a user wishing to switch to cythonic components can simply change their import statements from

from ridepy.data_structures import Stop

to

from ridepy.data_structures_cython import Stop

A concrete example of how to transition from a pure python simulation to a cythonic simulation is provided in tutorial 2.