Back-to-School Part 3 – Class Setup for Professors with “Courses” in ODA and VFL
Welcome, welcome Academic friends – to Part 3 of our Back-to-School series! Oh the places we’ve been!
Part 1 focused on how professors and students can upload their data to SAS Viya for Learners (VFL). Part 2 encouraged the academics using SAS Enterprise Miner on SAS OnDemand for Academics (ODA) to consider switching to predictive modeling and machine learning tools in SAS Viya for Learners (VFL) for a better, more integrated user-experience.
This post is all about course setup and data sharing.
With hundreds of thousands of users worldwide, SAS OnDemand for Academics is – put modestly – a great tool for educators and students. Professors love it for a variety of reasons, including access to SAS Studio and SAS Enterprise Miner – as well as SAS + Python integration in Jupyter.
One feature that many professors utilize is the ability to share data through “Courses” in ODA. Courses are found in the right pane of the ODA landing page, with that fun little graduation cap icon:
Courses allow the professor to share data directly with students. Have a bunch of data sets that your students will work through over the course of a semester? Well, then this is the place for you. And you’ll have up to 3G of cloud space to share with the analytical minds that you’re molding. More details on creating a course from instructors can be found in the SAS OnDemand for Academics support page.
When instructors look to transition from ODA to VFL, they often ask: How do I set up a course in VFL? Hmmm. Sorry. That’s not a thing in VFL3.5… and won’t be a thing when we transition to VFL4.
But all is not lost! The general essence of the question is: how do I share course data in VFL? And the simple answer is GitHub!
I use GitHub in two main places in SAS Viya for Leaners. The first is in SAS Studio. In all this whitespace, do you see the beacon of light on the left pane?
Yes! That right there is a Git Repositories button. It will allow you + your students to pull from a GitHub file location, after you set up the connection. Here is an illustrative example. Start by clicking on that Git Repositories icon. Then select the More options button and then Manage Git Connections:
As noted, it’s now time to Manage Git Connections. Which means that you’ll need to set up a Git profile first – so, technically, there’s a step before this one. But, for our purposes, you’ll just need to set up a basic connection, which includes a Profile name, User name, and email address. All of those can be linked to your Git Profile. The two main clicks to get you there:
With the main information required to Add a Profile:
After all that Git’n setup is done, you now have a profile to pull from. The next step is to Clone a repository… which simply means to copy data. Like you’re a professor sharing data with a student. As in this example 😊
Find the button towards the top of the Git Repositories box:
The Clone button reveals the Clone a Repository window:
In my example, I’ll pull some data for a SAS On-the-Job that I’m currently developing. What’s a SAS On-the-Job? Well, that’s a cool story for another time. But the repo is here with this pretty screenshot from the webpage:
To connect to this repo, the rest of the details are something like:
And know that we can create new folders under our shared space, such as SAS_OTJ_HHS, to better organize our project.
When you’ve got it set up properly, click that Clone button! Next, return to Explorer and expand your newly cloned folder, here it is SAS_OTJ_HHS, so that you can know the connection is a good one.
Success!
The second place that I like to use GitHub data is via direct Jupyter Notebook calls. Because the call is so elegant, this section will be very short. But it’s a powerful tool, nonetheless.
Within the SAS_OTJ_HHS_LFP_Part1.ipynb file above, there is a single line of code that shows the magic. And that trick is to go out to GitHub, grab a single file, import it, and then carry on with the analysis. And, within a Python kernel, using SASpy, it looks like this:
That single line of code does oh so much. And please check out the rest of my GitHub page if you’re interested in more about the illustrative example. I think it’s fun.
One last caveat before wrapping up. Currently, you’re not able to push change to files to Git – for security reasons – but this could change in the future. Then again, you may not want students overwriting your files. And I get that.
If you’ve somehow made it to the end of this article and are thinking, “what the heck is he talking about?!” then thank you for your perseverance. And reading skills. Here are a few YouTube videos to help you better understand our free software for teaching and learning: (1) SAS Tutorial | SAS OnDemand for Academics vs. SAS Viya for Learners (2) SAS Tutorial | Getting Started with SAS OnDemand for Academics, and (3) SAS Tutorial | Getting Started with SAS Viya for Learners
Happy teaching and book learnin’!
