Quickstart guide, keyframing, small updates

[ifranda]

Adding two new notebooks

[github-actions]

👋 Thanks for opening this PR! The Cookbook will be automatically built with GitHub Actions. To see the status of your deployment, click below.
🔍 Git commit SHA: 46053bb
✅ Deployment Preview URL: https://ProjectPythia.github.io/vapor-python-cookbook/_preview/14

[ifranda]

Any suggestions are appreciated!

For downloading the datasets and session files, I included links to the files in Google Drive. Let me know if there is a better way!

[NihanthCW]

Thanks @ifranda. Looks good. A few suggestions:

• In the quickstart and other new notebooks. Can downloading the dataset be done as a cell int he notebook itself to prevent them from having to download it separately? What do you think?

[ifranda]

I think we'd be able to! Is the dataset too big to host on GitHub? That would probably be the easiest because then they'd have it when they clone the repository. It's about 170 MB. Otherwise I think we could find another way!

[NihanthCW]

RDA would be a good option. I've moved VAPOR sample data to RDA. We can move Katrina data to the same place?

[NihanthCW]

Awesome! Thanks Ian.
Two suggestions:

• The download is working great! It might be good to check if the file already exists so that people are not re-downloading the file every time they run the notebook.
• How do you feel about including one cell to allow people rotate and look around in the scene just before exporting the image? Something similar to viz = VaporVisualizerWidget(ses)

[ifranda]

Good ideas! Another thought -- if users are following along by copying and pasting into their own workspace instead of cloning the whole repository, they won't have access to the get_data.py script. Another idea I had was to have a cell for command lines, like this:

!curl -o Katrina.zip "https://data.rda.ucar.edu/ds897.7/Katrina.zip"
!unzip -o Katrina.zip -d ./data
!rm Katrina.zip

[NihanthCW]

Agreed.

[ifranda]

Ok, I've made some updates and I think it runs better!

[ifranda]

I got the image renderer working with a fresh conda build! But it looks like Natural Earth is the only builtin map at the moment, and I think that one only renders land, unless I'm wrong about that? Do you know if there's a way to get the oceans to render? Otherwise maybe we could try adding BigBlueMarble to the builtin maps

This is what it looks like at the moment:

Using the filepath to BigBlueMarble:

[NihanthCW]

You are right. The oceans are black in the dataset. Could this be another file that we could download? Alternatively, the black ocean down't look bad too...
I believe we intentionally donot include blue marble to avoid making the maps package too large.
Let's chat tomorrow.