Jupyter

From ArchWiki

Jupyter is a project which produces browser-based interactive environments for programming, mathematics, and data science. It supports a number of languages via plugins ("kernels"), such as Python, Ruby, Haskell, R, Scala, Julia and Kotlin.

JupyterLab is "Jupyter’s Next-Generation Notebook Interface", while Jupyter Notebook is the original. See the Jupyter website for a comparison.

Installation

To install third-party Jupyter Notebook extensions for the current user, use the --user option while executing jupyter nbextension install. To do the same for installation of JupyterLab extensions, set the following environment variable:

JUPYTERLAB_DIR=$HOME/.local/share/jupyter/lab

and verify it by running jupyter lab paths. Then onwards follow usual installation instructions.

Running

To start JupyterLab run:

$ jupyter lab

To start Jupyter Notebook run:

$ jupyter notebook

Navigate to the URL given on the standard output if a web browser does not automatically open.

To start JupyterLab without launching browser and listening on port 9999 run

$ jupyter lab --no-browser --port 9999

To change the default behavior edit

~/.jupyter/jupyter_lab_config.py
c.ExtensionApp.open_browser = False
c.ServerApp.port = 9999

Kernels

C++

Install the cling-jupyter-gitAUR package.

Haskell

Install the ihaskell-gitAUR package. Then run ihaskell install.

Julia

Install the julia package and run julia to get a REPL prompt. Then run:

using Pkg
Pkg.add("IJulia")

See the Julia manual for more details on package management.

Python

Python 3 kernel is used by default via python-ipykernel.

Perl

Install kernel and run interactive perl shell at least once:

cpanm Devel::IPerl
iperl

Then press Ctrl+d. Now if you run jupyter you will see perl there.

R

Follow the installation instructions in IR Kernel.

Rust

Install the evcxr_jupyterAUR package.

SageMath

Install the sagemath package.

Octave

Install the jupyter-octave_kernelAUR package.

Maxima

Install the maxima-jupyter-gitAUR package.

Cadabra

Install the cadabra2AUR package.

Kotlin

There is Kotlin Jupyter integration project.

The factual accuracy of this article or section is disputed.

Reason: pip is discouraged outside a Python environment, see Python#Package management. (Discuss in Talk:Jupyter)

Install Kotlin kernel with pip:

pip install kotlin-jupyter-kernel

The kernel is automatically bind to jupyter.

Interactive widgets in JupyterLab

The factual accuracy of this article or section is disputed.

Reason: The linked issue is specific to python-matplotlib, the title of this section is overly general. The linked comment says to use %matplotlib ipympl, while the following instruction here contains %matplotlib widget. It would be much better to link to a proper documentation on matplotlib's magic commands. (Discuss in Talk:Jupyter)

In order to enable interactive widgets in Jupyter Lab install python-ipympl and jupyterlab-widgets according to this github issue. Afterwards, in your notebook use:

%matplotlib widget

Don't forget to restart you JupyterLab instance after installing extensions.

It also might be helpful to RMB->Clear Outputs of All Cells after your extension manipulations

See also