FluxML / Optimisers.jl

Optimisers.jl defines many standard optimisers and utilities for learning loops.
https://fluxml.ai/Optimisers.jl
MIT License
72 stars 20 forks source link

Document `destructure` handling shared parameters differently to ComponentArrays.jl #161

Open mcabbott opened 11 months ago

mcabbott commented 11 months ago

This came up here:

https://discourse.julialang.org/t/on-the-future-of-flux-destructure-and-sciml-integration/104760/4

Maybe the idea that ComponentArrays.jl treats shared arrays as independent should be mentioned as an important difference, perhaps at the end of this section:

https://fluxml.ai/Optimisers.jl/dev/#Obtaining-a-flat-parameter-vector

Edit, example here:

https://github.com/FluxML/Flux.jl/issues/2405#issuecomment-2015447460