SciML / DiffEqFlux.jl

Pre-built implicit layer architectures with O(1) backprop, GPUs, and stiff+non-stiff DE solvers, demonstrating scientific machine learning (SciML) and physics-informed machine learning methods
https://docs.sciml.ai/DiffEqFlux/stable
MIT License
871 stars 157 forks source link

reconstruct_Foo instead of reconstruct NeuralODE #264

Closed metanoid closed 4 years ago

metanoid commented 4 years ago

Maybe I just don't understand functors, in which case please close this issue.

But it seems odd to me that on this line, the name of the returned object is reconstruct_Foo rather than reconstruct_NeuralODE

https://github.com/SciML/DiffEqFlux.jl/blob/dceff374e5fff15c732685df3c69aa2c27102260/src/neural_de.jl#L69

Logging an issue in case it needs review

ChrisRackauckas commented 4 years ago

Oh, those functor definitions can be completely removed. They are overridden by https://github.com/SciML/DiffEqFlux.jl/blob/master/src/neural_de.jl#L3 , so Zygote doesn't end up caring about our functor definitions.