anvaka

I asked each model a simple prompt: "Most stereotypical woman in XXX. No text. " I didn't edit anything, just one attempt. I'd expect some variety too, but it does show some variety/biases in the model's image generation~

The source code is available here: https://github.com/anvaka/imaginary-faces

I have updated the map with the latest data. This one is based on 500M stars given to repositories from GitHub's start till end of April, 2025.

Repositories are clustered based on jaccard similarity.

The mao helps you find what other related projects are available. So if you know something good - you can easily find what else is there.

Thank you!

oh wow. Glad you liked it!

appreciate the feedback - I'll take a look

haha thank you!

I couldn't find a universal clustering algorithm yet: Frequently there is more than one way to group data that still makes sense, and as a result whichever final clustering option we choose - it will not be perfect.

Hm... unless maybe we do some sort of quantum clustering, which could be a fun project to explore!

It's a bit hazy now, but I remember trying hdbscan algorithm (hierarchical clustering), and on the graph of the GitHub size - I just couldn't fit it in memory.

I did end up using something similar to hierarchical clustering (mix of louvain/leiden/my own), and that's what we see in the final map.

haha! I love vim.

We shall not quit.

love it =)!

Yes...

Aiming to redo it some time in early 2025!

Jaccard similarity is not particularly good for "celebrity" projects.

They are similar because they are popular, not because there is semantic relationship.

It's the same problem I faced with the map of reddit (https://anvaka.github.io/map-of-reddit/ ) - all popular subreddits are just "similar" to each other.

Stil works great for smaller, non-celebrity projects :D

Thank you for your kind words!

That link you've shared - doesn't open for some reason

Thank you!

I tried quite a few various similarity metrics, and Jaccard was giving me the best results. This is all very subjective, of course.

Thank you =)

This is correct!

I agree. I'm not sure why they are together. Maybe all that tooling stuff that was happening in early days?

You are very kind and your compliment put smile on my face too :)

Thank you so much for your compliment!

thank you!

Thank you so much!

I'm puzzled by some countries in Frontartia's island too. I'm not sure why it was even pulled away, as if there is something I'm missing.

Indeed! A lot of amazing things happens there. A few are getting world wide adoption (echarts are awesome if you are into frontend viz, so is element-plus library for vue components). There are giant communities of ML too

True... Scheming Lispania :)?

thank you :)

Should I rename Schemaria to Lispania?

haha, I love the flags idea :). I wish I had the prowess to implement it in a way that is visually appealing and not obscuring the map.

I would also love to have a giant octocat hugging the archipelago, with some radial gradient emitting inside of it. Alas my design-gl foo is not there yet

True. Though I also expected to "Reactistan" somewhere more explicitly, I wonder if it is scattered in Fronterra continent?

I am also puzzled myself why the algorithm has separated Fronterra from another large island to the west, seemingly related to node and some other JS libraries. Decided to keep it all as is in case I'm missing some reason

Thank you!

The connections are inferred from stargazers. If they lead to different set of related projects it might be a sign that different group of people gave stars different things at that time.

Github of course has much more dimensions than a flat 2d surface can show