Jupyter

From miki
Revision as of 08:30, 10 February 2016 by Mip (talk | contribs) (→‎Sagemath kernel)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

The Jupyter Notebook is a web application that allows you to create and share documents that contain live code, equations, visualizations and explanatory text.

References

Examples are available on the Jupyter viewer:

Jupyter is available as standard Python packages, but also distributed via Anaconda:

Modern open source analytics platform powered by Python

Install

First install Anaconda:

bash Anaconda2-2.5.0-Linux-x86_64.sh

To start the server:

jupyter notebook

Sagemath kernel

We can register sage to have it available as a kernel in Jupyter. Here the recipe I followed using

  • Ubuntu 14.04 LTS
  • Jupyter 4.10 — installed locally in user home ~/.ipython / ~/.jupyter.
  • Sage 6.10 — installed in /sage.

First starts sage and register the kernel using SageKernelSpec [1]

/sage/sage
sage: from sage.repl.ipython_kernel.install import SageKernelSpec
sage: spec = SageKernelSpec()
sage: spec.update()  # not tested
  • This does nothing, but apparently registration requires a file named kernel.json [2]. We find this file in /sage/local/share/jupyter/kernels/sagemath/
$ find /sage -name kernel.json
/sage/local/share/jupyter/kernels/sagemath/kernel.json
In the same tree we also find jsmol and mathjax extensions.
  • From iPython/Jupyter documentation [3], we understand we must add the links ourselves in user's home:
mkdir ~/.ipython/kernels
ln -sf /sage/local/share/jupyter/kernels/sagemath ~/.ipython/kernels
ln -sf /sage/local/share/jsmol                    ~/.ipython/nbextensions
ln -sf /sage/local/share/mathjax                  ~/.ipython/nbextensions

From now on sage is available in Jupyter notebook. Note that we don't need to start sage separately.

Update — More information on ipython-jupyter migration page. Can we use the following instead:

jupyter nbextension install [--user] EXTENSION
jupyter kernelspec install [--user] KERNEL