Setup the FUSE environment

Julia installation

We highly recommend using the Juliaup manager to install Julia

  • Mac & Linux: curl -fsSL https://install.julialang.org | sh
  • Windows: winget install julia -s msstore

Once installed, restart your termninal to pick-up the julia executable.

FUSE installation

FUSE and related packages are registered at the FuseRegistry. For installation start your Julia interpreter by typing julia at the terminal, then:

  1. Add the FuseRegistry and the FUSE package as you would for any other julia package (for a fresh install this can take 20+ mins):

    using Pkg
    Pkg.Registry.add(RegistrySpec(url="https://github.com/ProjectTorreyPines/FuseRegistry.jl.git"))
    Pkg.Registry.add("General")
    Pkg.add("FUSE")
  2. Now you should be able to import the FUSE package:

    using FUSE
  3. Install the fusebot utility to simplify install/updates later on. Now fusebot should be a command that you can type anywhere from the terminal.

    FUSE.install_fusebot()
  4. Run the regression tests (optional, this can take 1h+)

    julia ] test FUSE

  5. Exit julia and clone FUSE examples in the current working directory. To see/run those .ipynb files, you'll need to use Jupyter-Lab or VScode.

    fusebot install_examples

Keeping FUSE up-to-date

  1. Get notified of new FUSE releases by "watching" the FUSE repo on GitHub

  2. FUSE is updated like any other Julia package:

    julia ] up

Tip

Become familiar with how managing Julia packages works.

Install Jupyter-Lab and add the Julia kernel to it

  1. You will need to install jupyter-lab if that's not already available on your system

  2. Install the IJulia package by running:

    fusebot install_IJulia
    Note

    Run fusebot install_IJulia every time a new julia version is installed.

    This will setup the single- and multi-thread julia kernels in Jupyter.

    The number of threads of the multi-threaded julia kernels can be set via the JULIA_NUM_THREADS environmental variable.

  3. Start a new Jupyter-lab session (this should open a web-browser page with Jupyter running)

    jupyter-lab
  4. Now you can browse the examples in the FuseExamples folder that you have cloned, and take a tour of the example Jupyter notebooks there.