97022b0733
* Initial commit * Undo master commit * Initial commit * Update CODEOWNERS * Add links to resolve PR comments Co-authored-by: Ward K Harold <wkh@google.com> Co-authored-by: Karl Weinmeister <11586922+kweinmeister@users.noreply.github.com>
2.1 KiB
Overview
Pluto is a programming environment for Julia, designed to be interactive and helpful. It provides a familiar notebook interface but it is not a Jupyter notebook. The biggest difference is that Pluto notebooks are reactive, changing a variable or function in one cell causes the cells that depend on that variable or function to be reevaluated. Pluto also provides useful interaction mechanisms that allow users to dynamically interact with the notebooks computation state.
The JuliaCon 2020 presentation: Interactive notebooks ~ Pluto.jl provides a good introduction to Pluto. The source is at fonsp/Pluto.jl
Install Pluto
Create a Vertex AI JupyterLab Instance
-
From the GCP console "hamburger menu"
select Vertex AI > Workbench
-
Click NEW NOTEBOOK
- Choose Python 3 if you won't be using a GPU
- Choose Python 3 (CUDA Toolkit xx.y) if you do want use a GPU
-
Give the notebook an appropriate name
-
Edit Notebook properties if you have special requirements otherwise accept the defaults and click CREATE
-
When the notebook instance is ready click OPEN JUPYTERLAB
Configure JupyterLab
- Open a terminal by clicking the Terminal icon.
- Install the plutoserver pip3 install git+https://github.com/fonsp/pluto-on-jupyterlab.git
- In a browser go to julialang.org/downloads
- In the Current stable release right click on the
Generic Linux on x86 / 64-bit (glibc)link Select copy link address - Back in the terminal switch to root via sudo -i
- Download the release to /opt and install julia in /usr/local/bin
cd /opt wget <paste the release link address> tar xf <name of the downloaded tar file> ln -s /opt/<julia-x.y.z>/bin/julia /usr/local/bin ^d - Add the Pluto package to Julia
julia julia> ]add Pluto julia> bksp julia> using Pluto julia> ^d - From the JupyterLab menu bar select File > Shut Down
Start Pluto
- Click OPEN JUPYTERLAB in the Workbench
- In the Notebook section of the Launcher click Pluto.jl
- The welcome to Pluto.jl screen should appear