Skip to main content

Simulation in Python.

Project description

Scipy Simulator provides a concurrent way of modelling and simulating heterogeneous systems in Python using scipy. You might find it most useful for tasks involving embedded systems or signal processing.

Scipysim models are created in pure Python code, by instantiating various actors representing different components of a systems, and connecting the actors to each other through channels. Here’s a simple example of what a model looks like:

#!/usr/bin/env python

from scipysim.actors.signal import Ramp
from scipysim.actors.display import Plotter
from scipysim.actors import Channel, Model

class RampPlot( Model ):
    def __init__( self ):
            super( RampPlot, self ).__init__()
            connection = Channel()
            src = Ramp( connection )
            dst = Plotter( connection )
        self.components = [src, dst]


You can find a number of other examples of models in the ‘models’ directory.

The scipysim project is inspired by the UC Berkeley Ptolemy project, but we are taking a slightly different approach to implementing the simulation engine. Our approach is based on implementing the simulator as a Kahn network of actors that communicate via tagged-signals. Each of these actors run in their own thread, and communicate via dedicated Channels - which are based on the thread safe FIFO queue implementation in the Python standard library. These base level actors can be composed together to create models, which are also actors in their own right - running in their own thread with all communication occurring through input and output channels.

Scipysim is still very much under active development, and contains a number of experimental or prototype components. The structure of the simulator is in a state of flux, so there are no guarantees that models developed to work with a particular release will still work with the next release.

Testing Scipy Simulator

Scipy Simulator comes with a large collection of unit tests. All the tests can be run as a suite using nosetests:


A helper script called has been placed in the scipysim module to launch nosetests:


If you downloaded from the repository the tests can be run with setuptools:

python test

The tests can also be found in the module hierarchy and run individually:

python ./scipysim/actors/io/

Installing Scipy Simulator

You can install scipysim to your main site-packages folder with:

sudo python install

on Linux or Mac OS X; and:

python install

on Windows. To install in a more sandboxed “development” environment substitute develop for install, e.g.:

sudo python develop

This installs an egg at the current directory and links to the package in your site-packages folder.

Creating Binary Installers

Firstly to clean the obsolete .pyc or .pyo files use:

python clean --all

Generate a built distribution like so:

python bdist

On Windows, to make a nice pretty GUI installer:

python bdist --format wininst

Similarly a source distribution can be created with:

python sdist


This project was initiated in the Department of Electrical & Computer Engineering at the University of Canterbury ( by:

  • Brian Thorne (brian dot thorne at canterbury dot ac dot nz)

  • Allan McInnes (allan dot mcinnes at canterbury dot ac dot nz)

Project Site

The main development occurs on Google Code at

Contribute to scipysim

First get the source code with mercurial:

hg clone scipy-sim

And send us a patch by creating a new issue

Project details

Download files

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

Source Distribution

ScipySim-0.1.2.tar.gz (46.9 kB view hashes)

Uploaded Source

Built Distribution

ScipySim-0.1.2-py2.6.egg (189.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page