Show HN: Jixp, a Lisp DSL for describing Jax neural nets

A developer released Jixp, a Lisp DSL for describing Jax neural nets, as a side project for learning Jax. The compiler reads Racket-style DSL and outputs Python, enabling simple model specification with hidden implementation details. The developer trained a 4m parameter model on their Obsidian vault for recall, but does not recommend using Jixp for anything besides learning.

This is a side project I've been working on while learning Jax. I noticed that a bunch of the neural net math looked like it would work well in a lisp syntax because most of the data flows through layers in a "functional" manner. Data is threaded through one layer at a time, each layer composing with the previous layer. Jixp is designed as a learning tool for myself while learning Jax to toy around with different shapes/styles of models. I used this to train a 4m parameter model on my Obsidian vault to see if I could use it for recall which partially worked. lisp let-dim d 256 heads 8 define transformer-block chain residual layernorm d attention d heads :causal residual layernorm d mlp d 4d d :bias The Jixp compiler reads the racket style DSL and outputs python which is then consumed via your normal python training stack. The advantage here is you get a pretty simple interface to describe your model and the implementation details are largely hidden in the generated python. I definitely do not recommend using this for anything besides learning. One day I can see models being specified in terms of a DSL like this which would allow different implementations of inference/training to load the same model. You could write one model definition, then vLLM, PyTorch, and your custom inference stack could all use the same definition. Sort of like a .safetensor file for the model. Comments URL: https://news.ycombinator.com/item?id=49037725 https://news.ycombinator.com/item?id=49037725 Points: 1 Comments: 0