Jupyter Notebooks Gallery | Hacker News Reader

Jupyter Notebooks Gallery

notebook.community

101 pointsjeffnotebook26 comments

This is not a valid Show HN, so we've taken that out of the title. Please see the rules: https://news.ycombinator.com/showhn.html. Note the word 'lists'.

Lists tend not to make good HN submissions because the only thing one can really discuss about them is the lowest common denominator (or greatest common factor?) of the list elements. It would be better to pick one of the most interesting elements and submit that instead.

HN is itself a list, and a pointer to a pointer to a pointer is too much indirection.

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...

Jupyter/Jupyter > Wiki > "A gallery of interesting Jupyter Notebooks" lists hundreds of notebooks: https://github.com/jupyter/jupyter/wiki/A-gallery-of-interes...

The mybinder.org Grafana dashboard lists the most popular notebook repos in the last hour: https://grafana.mybinder.org/

Jupyter/Nbviewer > FAQ > "How do you choose the notebooks featured on the nbviewer.jupyter.org homepage?" :

> We originally selected notebooks that we found and liked. We are currently soliciting links to refresh the home page using a Google Form. You may also open an issue with your suggestion.

https://nbviewer.jupyter.org/faq#how-do-you-choose-the-noteb...

Google Form: https://docs.google.com/forms/d/e/1FAIpQLSd6AlVvC7KagENypGTc...

Here's the Nbviewer source code. AMP (And https://schema.org/ScholarlyArticle / Book / CreativeWork metadata) could be useful. https://github.com/jupyter/nbviewer

Unrelated to the content itself, but it takes 5-10 seconds for the text to be displayed on the homepage if 3rd party JS is blocked. 10s+ to finish loading when using the "Slow 3G" speed profile on Chrome.

I'm not sure if AMP is a good choice for a simple site like this. It's slower than regular HTML/CSS.

I remember posting about this on HN some time ago [1]. I didn't even have the patience to wait for the content to show up on this site since it wouldn't load in a reasonable time without access to the ampproject.com servers.

1: https://news.ycombinator.com/item?id=18893808

Jupyter notebooks are almost a huge step forward for programming. The instant feedback, saving of partial program state, and visualizations mixed in are amazing and fun to work with. The fact that it's taking place in a browser, you don't have modal editing with Vim, you don't get plugin support (like MyPy), and you don't have intellisense is a huge step backwards.

I'm hoping that someone addresses these issues, I know VSCode is trying, and Scala "Metals" worksheets are another similar concept (though different, it puts variable values in comments next to code as you update the code).

I feel like all of these are a step closer than what the Lighttable/Eve teams were shooting for, but not quite the next generation way of coding that it could be.

I'm a huge fan of rmarkdown since it addresses many of these shortcomings by way of being much simpler than jupyter notebooks. The files are simply text files, and you can write python too if you like, they aren't just for R.

Jupytext:

> Jupyter Notebooks as Markdown Documents [MyST Markdown, R Markdown], Julia, Python or R scripts

https://github.com/mwouts/jupytext

We (Paperspace YCW15) have something similar at https://ml-showcase.paperspace.com/ where you can launch and run some popular Jupyter notebooks and run for free :)

Disclaimer: I’m the CEO/co-founder

How about this one:

The Super Duper NLP Repo (262 notebooks)

https://notebooks.quantumstat.com/

You could also check out all the notebooks on https://DataCrayon.com

e.g.

https://datacrayon.com/posts/search-and-optimisation/practic...

Disclaimer: our site

If you're taking submissions, I have a repo of blog-posts-as-notebooks, e.g.:

Lindermayer systems from the ground up: https://github.com/paulgb/notebooks/blob/master/source/l-sys...

"Did the White House doctor a video?" (basic forensic video analysis): https://github.com/paulgb/notebooks/blob/master/source/white...

Reminds me of Peter Norvig's pytudes[0].

On another register, if you are active in machine learning and you use notebooks a lot, we're working on our platform which you might find useful. No setup notebook, parameter/model/metrics tracking, model deployment, live editing and scheduling for notebooks[1].

- [0]: https://github.com/norvig/pytudes#pytudes-index-of-jupyter-i...

- [1]: https://iko.ai

Co-creator here of notebook.community here!

I wanted to add an extra value by:

1. Improving the way notebooks are indexed by Google, and make quality content more discoverable (hence AMP).

2. Implementing recommendations, again, to aid finding similar quality notebooks. These are currently a TF-IDF of notebook contents.

It's an experiment whether this helps people navigate among data science work. There's about 100k notebooks in total...

Interestingly, by looking at search console, so far it mostly helps students find solutions to their assignments, and others some specific usage of python libraries.

Good luck with your project.

A few people here have noted the lack of links to the original source - are you going to implement those? You clearly state they are open source notebooks.

They're there already - at the very bottom there's a link to content source, and a link to GitHub repo.

The path to file within the repo can be inferred from the URL, but perhaps I can also add a direct link too, makes sense :)

Thanks for the suggestion

You are right, I missed that, apologies and thanks.

The Similar notebooks at the end are pretty cool.

What would be even cooler is some kind of analysis of what jupyter notebooks usually contain, how big they are, what python libraries they use, what plotting library they use, etc.

I am a fan of the approach. However, it looks to me that the mathematical notation is not being rendered properly for some of the exercise pages.

Any suggestion for notebooks/examples that use interesting datasets? Yet another Twitter sentiment analysis its not exciting stuff for me.

I've been wanting to learn Jupyter and self host a Jupyter server for local network use. Anyone have a favorite tutorial?

I like the idea however I would prefer to at least have the option to download the .ipynb files

It would be convenient to have links included, however they seem to be discoverable by replacing the domain with "github.com" and then removing extraneous paths.

https://github.com/nealcaren/workshop_2014/ https://github.com/dibgerge/ml-coursera-python-assignments/

Yeah - it clearly says they are open source but it's not obvious where the originals are hosted...

Anybody else find the teal font color nigh unreadable?

You mean 10 interesting notebooks? This probably could be a nice blog post.

and it's not clear in the title that they are ml/data science related notebooks

Reply on news.ycombinator.com