Lectures

The class material may include Jupyter-based notebooks.

You can view them here, or open them to interact. You can use any environment for your local development environment, or use the SDSU’s JupyterHub on the Instructional Cluster to experiment and develop without a local install. If you have never logged-in before, check SDSU’s Research & Cyberinfrastructure resources for students. Once you are on the Instructional Cluster page, simply click on the Launch JupyterHub button.

If you decide to use SDSU’s Instructional Cluster’s JupyterHub, you can download each individual notebook from the class website, under the lectures section on your machine, and once opened Jupyter on SDSU’s JupyterHub you can select Upload Files and upload the desired notebook in the web app to interact with.

Environment

This explains how I configure my environment so you can experiment with the lectures locally, or adapt these tools for your own use.

Install Required Packages

First of all, for Jupyter Notebooks involving the Julia programming language, download and install Julia.

To interact with the notebooks, you want to install JupyterLab (new, richer ecosystem), via

pip install jupyterlab

Once installed, launch JupyterLab with:

jupyter lab

and selcet the Notebook app.

You can also only install the classic Jupyter Notebook standalone app via

pip install notebook

and to run the notebook type:

jupyter notebook

And for these (using Julia in Jupyter) you can run

julia -e 'import Pkg; Pkg.add("IJulia")'

from your terminal, or first start a Julia session with

julia

and then type:

]add IJulia

go back to your Julia REPL using backspace and type

julia> using IJulia

and then

julia> notebook()

to run the notebook.

Best practices

• When saving, use Kernel -> Restart & Clear Output to keep the stored data and diffs in the notebook small.

• For your images, several formats will do. Just keep in mind that PDF files are not web-friendly. PDF images can be converted to SVG using pdf2svg, and the result will still look sharp (unlike PNG) no matter the scale or zoom.

Online resources:

• For the first part of our course, an online MOOC: LAFF-On Programming for High Performance (LAFF-On-PfHP) will be particularly useful.

I encourage you to watch the videos before class. When watching the videos, really you should be watching for the big picture and we will revisit most of the material in lecture as well, mainly you will have a better learning experience if the in class lectures are the first time you’ve seen some of the material.