vega / altair

Declarative statistical visualization library for Python
https://altair-viz.github.io/
BSD 3-Clause "New" or "Revised" License
9.21k stars 783 forks source link

Call for Altair projects and initiatives to showcase in the docs #2697

Open joelostblom opened 1 year ago

joelostblom commented 1 year ago

I think it would be helpful to have a page in the docs that includes links to inspiring Altair-based projects and initiatives to highlights some of the impressive things that package is used for. This could include projects that use Altair to tackle a real world problem, a demonstration of how to use some of the more advanced/complex functionality in Altair, or show how to use Altair together with some other package that is outside the scope of the official Altair documentation. This can also include talks, workshops, and lecture materials for Altair (which we already started to add to the resource section of the docs).

I have seen this done for other projects such as Bokeh and scikit-image in the past and found it quite helpful and encouraging myself. If anyone is aware of a project that fall under either of these categories, please comment with a link to it here together with a 1-2 sentences describing what it does (and optionally a link to the source code if your project allows it).

(Later edit) Maybe we should also include a list of interesting articles, starting one here for now:

daniel-finnan commented 1 year ago

Not particularly advanced or complex, but I recently enjoyed using Altair to chart French wind speed data, linking it to an interactive map created with Folium: https://github.com/daniel-finnan/zephyr Thanks for the library Altair team.

joelostblom commented 1 year ago

Thank you for contributing @daniel-finnan ! I haven't seen Altair used with Folium before (and didn't even know folium had vega lite support), that's really cool!

jbloom commented 1 year ago

We have been using altair to create a SARS-CoV-2 "antibody escape calculator" which enables interactive prediction of the effect of mutations to the virus's receptor-binding domain on antibody escape. This calculator is now widely used in the field to help assess the possible impacts of mutations: https://jbloomlab.github.io/SARS2_RBD_Ab_escape_maps/escape-calc/

sderuiter commented 1 year ago

Using Altair to run a nightly batch on data, which generates json files, stored in a GitHub repo, which then get served on Concordium Explorer.

pierre-haessig commented 1 year ago

Back in 2019, I've created dot plots to show yearly statistics on the French power system. It's mostly in Matplotlib, because I'm more proficient with it, but in the last iteration, I wanted an interactive version which made me jump to Altair

betaigeuze commented 1 year ago

I just recently finished my master thesis, where I used altair quite extensively together with streamlit. You can check it out here (it's a little bit slow, since I'm using the free tier of streamlit cloud).

It's a proof of concept to investigate random forests. On the "Dashboard" page you can see all of the altair-related work. You can also find the repo here. Would be happy about any feedback 😀

palewire commented 1 year ago

I would nominate "First Python Notebook," a fully-scripted training I've been running for journalists for years now. It features a lengthy introduction to Altair that showcases how you can use it to explore a real news dataset.

mcp292 commented 1 year ago

I don't think academic papers fit this category. Is there a place to post and browse academic papers that use Altair for their graphs?

joelostblom commented 1 year ago

@mcp292 I don't think there is any manually curated repository of papers that use Altair. In Google Scholar you can click on "cited by" for the original Altair paper to see some papers that either use or talk about Altair (we could add this link to the docs when we add the projects mentioned here).

mcp292 commented 1 year ago

I was thinking a place for people who use Altair to post their papers so that other users can read (and cite) papers with pleasing visuals and intelligent authors.