DiffEqFlux.jl
DiffEqFlux.jl copied to clipboard
SciML
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
Mirror of [this issue](https://github.com/SciFracX/FractionalDiffEq.jl/issues/63). I am interested in exploring the possibility of using this package together with [FractionalDiffEq.jl](https://github.com/SciFracX/FractionalDiffEq.jl). In particular, I would like to setup neural differential equation problem using...
@prbzrg I saw your https://github.com/impICNF/ICNF.jl which seems to have a lot of FFJORD things, should FFJORD and all just be a whole separate repo and pull that into SciML? If...
Here's the current implementation -- https://github.com/avik-pal/LocalRegNeuralDE.jl, which is not super user-friendly (and is unlikely that I will register it). All we need to do is: - Implement perform step for...
https://github.com/SciML/DiffEqFlux.jl/blob/01173be0cbc56bc862336d26d85cb7a0e601dfca/.github/workflows/CI.yml#L41
Is there any interest in the implementation of algorithms outlined in the following paper: [Matching Normalizing Flows and Probability Paths on Manifolds](https://arxiv.org/pdf/2207.04711.pdf) As an alternative to FFJORD? I would be...
The Nested AD used in DiffEqFlux is not ideal because it calls `ForwardDiff.gradient/jacobian` and Zygote overrides them to compute the Hessian before doing the HVP. [BatchedRoutines.jl](https://github.com/LuxDL/BatchedRoutines.jl/blob/main/src/api.jl) has routines to do...
This pull request changes the compat entry for the `ADTypes` package from `0.2` to `0.2, 1`. This keeps the compat entries for earlier versions. Note: I have not tested your...
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "github-actions[bot] avatar"){kind=avatar}
Current version ```julia @concrete struct NeuralODE{M
