JupyterBook idea

[DPotoyan]

This is an incredible resource you guys have created! I applaud the big effort you put in.
I am a comp chem faculty who was looking for an appropriate resource with an optimal learning curve to onboard new students with no prior experience with scientific python or coding. This seems to fit the bill.

One interesting idea I want to toss out there is having a website build with Jupyterbook which allows for interactivity in cells https://jupyterbook.org/intro.html which besides Bokeh/Plotly graphs you can embed gifs and videos. I have build lecture notes using Jupiter notebooks and markdown and quite please with it.

I look forward to using your materials to train students and hopefully contributing or adopting some parts for boot camps.

[rsokl]

@DPotoyan thank you for the kind feedback and for taking the time to share your idea with me.

I have looked only briefly into jupyterbook, and it looks like a really great resource. There are a few things that I need to look into, e.g.

1. Does jupyterbook enable me to insert metadata into pages for search engine optimization (as does sphinx)
2. Can I convert my content nearly directly - and preserve things like the colored "call out boxes" that I use for things like reading comprehension questions
3. How do the load times and browser compatibility compare to the HTML produces by sphinx/RTD

Hopefully I will find some time to dig into these items. I have thought about pinging people from the jupyterbook project to see if they can lend some insight (or even a helping hand!) to this.

I look forward to using your materials to train students and hopefully contributing or adopting some parts for boot camps.

I am so glad that this will be a useful resource for you. And I would be more than happy to work with you to add more to PLYMI. The things that I most want to add to the site are:

1. a "cheat sheet" for the end of each module
2. (many) more practice problems, both short and long from, to put at the end of modules. I have a few at the end of modules 2 and 3, but it would be great to have many more

and please let me know if you have any suggestions or ideas, or material to contribute.

P.S. Just curious - how did you find the site? It seems like organic search gets a decent number of people here, but direct references produce more long-term readers. You, a university faculty, are certainly the kind of person I hope to reach more of.

[DPotoyan]

@rsokl

• I believe Jupyterbook allows all of that and more. The color boxes, as well as show/hide solution panels, exist there and are aesthetically quite pleasing. Probably some small scale tweaking will be needed on your part when converting. But note that Jupyter book is based on Sphinx and you can provide your custom sphinx configuration! See here. Asking Jupyterbook developers for advice on how to do conversion painlessly is a very good idea!

• Some of the things that are still missing from scientific python tutorials are some examples of interactive plotting (with plotly or holoviews), using widgets, and making dashboards for interactive exploration of data and deployment on the web. Matploltib while a good place to start for quick plots ultimately becomes inefficient when wanting to explore large and complex datasets. I am sure as someone working on machine learning you would appreciate these points :)

P.S. As with my most discoveries the site I discovered by simply googling for some beginner python tutorials for scientists. Though can't remember exactly what keyword combinations I used.

1. The cheat sheet is a great idea but you may also link to some cheat sheets that already exist on Github or other sides like this form Datacamp.

2. Agreed and this is one reason why I was searching for a source of material with good challenging problems relevant to the STEM field. When you frame the problem in a scientific context it becomes more understandable for students. E.g explaining OOP and classes using the hierarchical relationship between elementary particles, atoms, and molecules