NeuralOperators.jl
NeuralOperators.jl copied to clipboard
SciML
DeepONets, (Fourier) Neural Operators, Physics-Informed Neural Operators, and more in Julia
* [chebyshev](https://github.com/CliMA/OperatorFlux.jl/blob/main/src/transform_chebyshev.jl) * [legendre](https://github.com/CliMA/OperatorFlux.jl/blob/main/src/transform_legendre.jl) * [multiwavelet](https://arxiv.org/abs/2109.13459) [Ref](https://github.com/CliMA/OperatorFlux.jl/tree/main/src)
It would be nice to have more benchmarks as they discuss here - https://arxiv.org/pdf/2111.05512.pdf - https://arxiv.org/pdf/2203.13181.pdf
In https://github.com/SciML/NeuralOperators.jl/blob/master/src/DeepONet.jl#L122-L124, multi-dimensional input grid is handled using an Argument error. Is there a reason this was done this way? Also, if it is a matter of contracting AbstractArray(s), we...
The neural operator is surprisingly absent in its fully continuous form. It can be implemented using Integrals.jl since that is differentiable. Unlike some of the other forms which kind of...
This would make them a lot more visible, able to show visuals, and would make them tested as part of the doctests. We'd just need to set `strict` mode, i.e....
- [ ] `sparseKernel1d` operation - [x] `sparseKernelFT1d` operation, which is covered by `SpectralConv` - [ ] `MWT_CZ1d` layer - [ ] `MWT1d` layer - [ ] their 2d versions...