Jupyter Notebooks with Almond

Just like Python, ScalaPy can be used inside Jupyter notebooks to build experiments in an interactive environment. With Almond, a Scala kernel implementation, using ScalaPy only requires a little configuration to be used!

If you want to get set up in a hosted environment, take a look at the example notebook on Google Colab. To set up a local instance, read on!

Configuring Python Support in Jupyter#

First, we must modify the default Scala kernel definition to load Python native libraries.

In your jupyter/kernels/scala/kernel.json file, replace the contents with

{
  "language" : "scala",
  "display_name" : "Scala",
  "argv" : [
    "bash",
    "-c",
    "env SCALAPY_PYTHON_LIBRARY=python3.7m java -Djna.library.path=/usr/lib/x86_64-linux-gnu/ -jar /usr/local/share/jupyter/kernels/scala/launcher.jar --connection-file {connection_file}"
  ]
}"

Make sure to replace /usr/lib/x86_64-linux-gnu/libpython3.6m.so with the path to your Python native library and ajust the SCALAPY_PYTHON_LIBRARY appropriately if you want to use a different version of Python 3. python3-config --prefix will give you the folder containing your installation's native libraries.

Loading ScalaPy in a notebook#

ScalaPy on the JVM contains with all the logic to load native libraries built-in, so you can import ScalaPy and start using it as usual!

import $ivy.`me.shadaj::scalapy-core:0.5.2`

import me.shadaj.scalapy.py