Still one of the best explanations of decision trees I've seen. The
scroll-driven animation that builds the tree split by split, while
simultaneously showing where each data point lands, does in 30 seconds
what most textbook diagrams fail to do in three pages
fwiw I work on data ingestion pipelines and I've found that starting with just boxes-and-arrows in something like Excalidraw gets you 80% of the way to knowing what you actually want. The gap between "I can picture it" and "I can build it on a webpage" is mostly a d3 learning curve problem, not a design problem.
xyflow that the creator mentioned is probably the right call for pipeline DAGs though -- we use it internally for visualizing our scraping workflows and it was surprisingly painless to get running
- A previous comment by me about my list of absolutely gorgeous, interactive, animated, high dynamic learning resources classified as S TIER
- S-TIER blogs are those that are animated, visual, interactive and absolutely blow your mind off
- A-TIER are highly informative and you ll learn something
- opinion blogs at the absolute bottom of the tier list because everyone everywhere ll always have an opinion about everything and my life is too short to be reading all that
- these are the BEST of the BEST, you ll be blown away opening each page is how good they are. i am thinking of creating a bookmark manager that uses my criteria above and runs across every damn blog link ever posted on HN to categorize them as S-TIER, A-TIER, opinion and so on
So amazing, wish there were more articles like this. I love visual learning.
Also reminds me of another blog post: https://pomb.us/build-your-own-react/ , probably not directly the same, but similar-ish written blog posts, easy to stay on track and follow. It is so easy to learn with this kind of blog post.
The balls-from-the-sky sieve-style animation* showing classifications literally falling out of the decision tree is my favorite part. I haven't seen this anywhere else (yet); this visualization technique deserves more percolation (pun intended). (#1)
Not even to mention the fact that the animation is controlled by scrolling, which gives an intuitive control over play, pause, rewind, fast-forward, etc. Elegant and brilliant. (#2)
Stunningly good also in the sense that it advances the story so people don't just drool at the pretty animation and stop engaging. Thus putting the "dark arts" in the service of learning. (#3)
All three ideas warrant emulation in other contexts!
* Find it towards the bottom under the "Making predictions" heading.
2015 was about the last year you could get away with publishing an interactive graphic with a fixed width — this made it harder do really creative/original work.
Still one of the best explanations of decision trees I've seen. The scroll-driven animation that builds the tree split by split, while simultaneously showing where each data point lands, does in 30 seconds what most textbook diagrams fail to do in three pages
One of the creators of R2D3 here. Funny to wake up to this today! Happy to answer questions here or on bsky
If I would like to build a visualization like this, but for a data ingestion pipeline, any tips on where to start?
I have it visually in my head, but it feels overwhelming getting it into a website.
fwiw I work on data ingestion pipelines and I've found that starting with just boxes-and-arrows in something like Excalidraw gets you 80% of the way to knowing what you actually want. The gap between "I can picture it" and "I can build it on a webpage" is mostly a d3 learning curve problem, not a design problem.
xyflow that the creator mentioned is probably the right call for pipeline DAGs though -- we use it internally for visualizing our scraping workflows and it was surprisingly painless to get running
Sort of like this? https://docs.tecton.ai/docs/introduction/interactive-tour I used https://github.com/xyflow/xyflow for this, with css animations for the edges. It’s probably easier now with coding agents and what not
Any plans for more articles, 10 years later?
It is a masterpiece! Each time I give an introduction to machine learning, I use this explorable explanation.
There is a collection of a few more here: https://p.migdal.pl/interactive-machine-learning-list/
Nice list! I remember HN talking about https://students.brown.edu/seeing-theory/ when it came out but sadly it seems like this website was discomissioned.
Added an entry for my data visualisation tool here: https://github.com/stared/interactive-machine-learning-list/....
Edit: found an updated link for seeing theory so I fixed it in the PR above. Feel free to cherry-pick if #24 is not relevant.
- A previous comment by me about my list of absolutely gorgeous, interactive, animated, high dynamic learning resources classified as S TIER
- S-TIER blogs are those that are animated, visual, interactive and absolutely blow your mind off
- A-TIER are highly informative and you ll learn something
- opinion blogs at the absolute bottom of the tier list because everyone everywhere ll always have an opinion about everything and my life is too short to be reading all that
- these are the S-TIER ones on my system
- https://growingswe.com/blog
- https://ciechanow.ski/archives/
- https://mlu-explain.github.io/
- https://seeing-theory.brown.edu/index.html#firstPage
- https://svg-tutorial.com/
- https://www.lumafield.com/scan-of-the-month/health-wearables
- these are the BEST of the BEST, you ll be blown away opening each page is how good they are. i am thinking of creating a bookmark manager that uses my criteria above and runs across every damn blog link ever posted on HN to categorize them as S-TIER, A-TIER, opinion and so on
https://visxai.io/ has a bunch more too — see the Hall of Fame section at the bottom for some of the highlights.
I also made a dozen of these a couple years ago, my two favorites:
- https://pair.withgoogle.com/explorables/fill-in-the-blank/
- https://pair.withgoogle.com/explorables/grokking/
This is from 2015. Both technically and conceptually it was ahead of its time.
It's a pity there seems not to be new (or other) material from Tony Hschu and Stephanie Jyee.
(Or can anybody find something more?)
So amazing, wish there were more articles like this. I love visual learning. Also reminds me of another blog post: https://pomb.us/build-your-own-react/ , probably not directly the same, but similar-ish written blog posts, easy to stay on track and follow. It is so easy to learn with this kind of blog post.
The interactive explanations here are still some of the best examples of how visualization can make ML concepts intuitive.
I wish more technical articles took this approach instead of starting with equations.
has anyone come across an r2d3-style explainer for something as high-dimensional as a Transformer's attention mechanism?
Not quite, but these help
https://poloclub.github.io/transformer-explainer/
https://youtu.be/wjZofJX0v4M?si=gT8Zlz1IY14KV_ju
This is still great after more than a decade.
R2D3 did an amazing job here. It’s rare to see statistical learning concepts explained visually this clearly.
Where's the rest of it?
Part 2: https://r2d3.us/visual-intro-to-machine-learning-part-2/
The balls-from-the-sky sieve-style animation* showing classifications literally falling out of the decision tree is my favorite part. I haven't seen this anywhere else (yet); this visualization technique deserves more percolation (pun intended). (#1)
Not even to mention the fact that the animation is controlled by scrolling, which gives an intuitive control over play, pause, rewind, fast-forward, etc. Elegant and brilliant. (#2)
Stunningly good also in the sense that it advances the story so people don't just drool at the pretty animation and stop engaging. Thus putting the "dark arts" in the service of learning. (#3)
All three ideas warrant emulation in other contexts!
* Find it towards the bottom under the "Making predictions" heading.
nice
Did they not have mobile responsive sites in 2015? Lol
2015 was about the last year you could get away with publishing an interactive graphic with a fixed width — this made it harder do really creative/original work.
Josh Starmers books are very visual as well, probably the best source I'd recommend to learn ML
https://www.youtube.com/c/joshstarmer https://statquest.org/