As far as I understand, this doesn't turn "regular" JavaScript into a NN, it's rather "write your logic in a DSL that creates a neural-net-shaped circuit"<p>the source code for the OS demo is not "regular" code, it's full of `emiRelu`... <a href="https://github.com/chenglou/handwired/blob/main/src/handwiredos/photo-gallery.ts" rel="nofollow">https://github.com/chenglou/handwired/blob/main/src/handwire...</a><p>The main site & readme seem to overclaim, while the docs say things like "With Handwired you write a numeric program" and "No tool converts existing JavaScript source into a program" - <a href="https://github.com/chenglou/handwired/tree/main/docs" rel="nofollow">https://github.com/chenglou/handwired/tree/main/docs</a><p>happy to be proven wrong, to have misunderstood this, etc.
It seems interesting but I am confused. What is the purpose of that landing page? It is not clear since it showcases several (not related) ideas:<p>- turning js code into something else, arguably high-level code for training neural networks<p>- a mini OS inside the browser<p>- a gallery<p>- document search
I think this is basically a compiler that turns arbitrary JS code into weights of a neural net that then behaves exactly like the JS program. And the stuff at the bottom are demo projects that are now running as a neural net.
the "DEMOS" label is rather small, I understand the confusion
I think for everyone interested in this project checking your OG twitter thread is also somewhat mandatory:<p><a href="https://x.com/_chenglou/status/2110557373651931267" rel="nofollow">https://x.com/_chenglou/status/2110557373651931267</a><p>It took me a bit to wrap my head around what you did.
Clever, I wonder what would happen if you did this in reverse? Like, if we froze the weights then added some layers trained on a bunch of weird edge cases then tried to convert those weights back into branches.<p>QA through entropy lol