Jake VanderPlas - Exploratory Data Visualization with Vega, Vega-Lite, and Altair - PyCon 2018

Speaker: Jake VanderPlas
Exploring a new dataset visually can provide quick intuition into the relationships within the data. There are a few well-developed visualization packages in Python, but they often have very imperative APIs that force the user to focus on the mechanics of the visualization - tick locations, axis limits, legends, etc. - rather than the salient relationships within the data.
This tutorial will introduce data visualization with [Altair](altair-viz.gith...), a package designed for exploratory visualization in Python that features a declarative API, allowing data scientists to focus more on the data than the incidental details. Altair is based on the [Vega](vega.github.io/) and [Vega-Lite](vega.github.io...) visualization grammars, and thus automatically incorporates best practices drawn from recent research in effective data visualization.
The tutorial will provide an introduction to the Altair package and its API, but more importantly will dive into the core concepts of effective data visualization that can be applied using any visualization package or tool.
Slides can be found at: speakerdeck.co... and github.com/PyC...

[@Alex-om6zq]

3:00:30-3:02:25 that's hilarious.

[@penguinmagehello]

Came to the visualization party late. Really great tutorial.

[@fernandoirarrazaval]

Here you can find the notebooks: github.com/altair-viz/altair-tutorial

[@jeebogareeb]

32:02 Fantastic stuff!

[@petrmitev4769]

Wow thanks for linking that. It's like Power BI, but for Jupyter!

[@ivanovmg]

This is a great tutorial! I haven't expected Altair to be so differentiating library. Now I definitely have to use it. Selections are just amazing and seem to be much clearer than the ones in bokeh library. Thank you very much!

[@jonasherseth5306]

Awesome stuff. Jake did a great job with this presentation.

[@harveysummers3175]

LOVE! This is exactly what I've been looking for! Thanks Jake!

[@saurabhdeshpande5973]

This is incredible ! Thanks !

[@10e999]

The api is incredible. wow.

[@alibek24galiyev]

because I love ggplot, I was using R. Now there's no reason to in R anymore (except dplyr, maybe)

[@tgwashdc]

This is an incredible visualization toolkit, if we shift our thinking to the "grammar of graphics".

[@anneest]

This is fantastic. I started using Altair yesterday, did quite a bit, got stuck, and then Jake gave me the reference to his Pycon 2018 workshop. This could not be better! :-)

[@drkarmakid]

Jake is awesome... his Pycon videos on visualization are pure magic

[@seanys]

How do I download the tutorial files? All the notebook links on altair-viz.github.io/altair-tutorial/README.html appear to be to non-existant wiki pages, i.e. in red.

[@seanys]

Also, the slides do not appear in the linked Speakerdeck collection, but can be found at:
speakerdeck.com/jakevdp/altair-tutorial-intro-pycon-2018
The GitHub link just 404's. Ugh.

[@dmitrysysoev4254]

Great progress in the library of over the past six months. Thank you, Jake!

[@EitanLees]

The notebooks and slides are available at github.com/altair-viz/altair-tutorial

[@mohsin-ashraf]

How we can get slides github link is not working???

[@baitoeysb7015]

Thank you so much sir

[@deepakpratap3792]

Hey Jake..Great tutorial....Loved it..

[@seanmortazyt]

fantastic tutorial.

[@digipak07]

Greetings from Pakistan. Supercool. Great job

[@smellypunks]

5 cylinders are in a row. Six and eight cylinders are side-by-side so need to be pairs.

[@timbledum]

The notebooks can be found at github.com/altair-viz/altair-tutorial

[@pekkalehtonen6075]

The notebooks github.com/altair-viz/altair-tutorial

[@immanuelhaffner6915]

Both links to the slides are broken. Could you please fix this? Thanks.
Great tutorial!

[@DavidNehme]

github.com/altair-viz/altair-tutorial

[@notbobbobby]

Holy. Shit. This is AMAZING

[@lliwnomolos]

11:55 + 17:00 conda ftw!

[@armynyus9123]

yes !
Absolutely missing in Pycon (again). Seeing Kenneth's Pipenv presentation and when he admits, yeah its slow, dependency calc is hard and we have to download everything... aaargh...

[@alex-craft]

So, basically Altair has two parts API and Translator (Altair Python API -> Altair Translator -> Vega LIte Grammar). Maybe it would be better to develop Translator in JS. Python API will be exactly the same, but Translator could be reused and contributed not only by Python community but by both Python + JS.

[@nutelina]

Jake explains in the video how he made the API automagically from Vega Lite.

[@navjotsingh2251]

Nah ain't nobody wanna touch JS 😆

[@MrSnasr20]

can we export this grapghs i want to use in IDLE

[@KhalilKHAMLICHI]

The ultimate tutorial.

[@kipishism]

where can I get the notebooks?

[@usf5914]

break time :1:40:55 to 2:01:00

[@DSMPerfi]

This is the greatest and best visualization library in the world.
Possibly.

[@nickspirit3]

ggplot is pretty cool too

[@davisalesbarreira2884]

F**king amazing!

[@Nadhiantara]

Watch it with VLC media player to maximize the volume up to 200%.

[@nutelina]

An invert and zoom...