Skip to content

bkataru-experiments/adagrad

Repository files navigation

adagrad

autograd in ada, by hand, as instructed and guided by deepseek-v4-pro

a complete, from-scratch reverse-mode automatic differentiation ("autograd") engine written in Ada, together with optimizers, layer abstractions, and a small training demo

getting started

install gnat either by itself or via alire, then

$ gnatmake -P grad.gpr

About

autograd in ada, by hand, as instructed and guided by deepseek-v4-pro

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages