This package contains various IPython demos based on Jupyter notebooks. The purpose is to illustrate their capabilities. Currently available:
- simulation of various dynamical systems defined by ordinary differential equations (sub-package
odesimu) - multi-zoom exploration of fractals (sub-package
fractals)
- Requires Python 3. There are dependencies to numpy, scipy, matplotlib (recent versions). There exist all-in-one software packages which simplify the management of these requirements: Miniconda, Python(x,y).
- A Jupyter server capable of launching a Python 3 kernel should be up and running, or the standalone Jupyter lab desktop client.
That should be pretty all. Should work on both Linux, MacOS and Windows.
Download the zip archive and expand it in some folder on your machine, e.g. your Download folder. Then create a symbolic link to the path ~/Download/IPYDEMO/src with name ipyshow (the name used in the demos) somewhere among the sys.path of your python install (e.g. by including it in the PYTHONPATH environment variable).
The demos are given as jupyter notebooks in folder ipyshow/demo. Most use matplotlib animations, and are configured with the widget backend which displays the animations inline. For better performances, it is advised to use the qt5 backend or similar (rendering in a separate window). Don't use the inline backend since, although it is inline, it does not support animations. To change the backend in a demo, modify the first line to be one of the following (and restart the kernel):
%matplotlib qt5
You can use sub-packages odesimu and fractals to build your own demos (don't hesitate to contact me on github if you have nice demos to add). The documentation of these modules can be generated using Sphinx: go to folder doc, copy file Makefile-in into Makefile and configure the latter (instructions inside), then run make html (or make pdf etc.).