That is very cool, and probably useful for beginners to visualize what is going on, at least on a macroscopic level.
In my first Springer-Verlag Common Lisp book (late 1980s), I had similar visualizations for the backpropagation neural network examples, and I got a lot of nice feedback that visualization is helpful.
I quickly looked at the ManimML source code. Interesting that everything is implemented from scratch just using numpy.
Hey Mark, Thanks so much! I'm glad you like the library. I actually built it on top of the https://www.manim.community/ library.
I checked out some of your writing, it's so cool! I have recently been going through Make a Lisp to write a Lisp interpreter in Rust.
I'm working hard to eventually bridge the gap from the library being just a tool that creates interesting/pretty graphics to something that could actually be used as a pedagogical tool.
That is very cool, and probably useful for beginners to visualize what is going on, at least on a macroscopic level.
In my first Springer-Verlag Common Lisp book (late 1980s), I had similar visualizations for the backpropagation neural network examples, and I got a lot of nice feedback that visualization is helpful.
I quickly looked at the ManimML source code. Interesting that everything is implemented from scratch just using numpy.
Hey Mark, Thanks so much! I'm glad you like the library. I actually built it on top of the https://www.manim.community/ library.
I checked out some of your writing, it's so cool! I have recently been going through Make a Lisp to write a Lisp interpreter in Rust.
I'm working hard to eventually bridge the gap from the library being just a tool that creates interesting/pretty graphics to something that could actually be used as a pedagogical tool.
How does it visualize transformers?
Coming soon!