NeuralPDE.jl
NeuralPDE.jl copied to clipboard
SciML
Physics-Informed Neural Networks (PINN) Solvers of (Partial) Differential Equations for Scientific Machine Learning (SciML) accelerated simulation
The docs claim that when doing parameter estimation problems w/ additional loss functions, the params should go to the end of $\theta$  By copying the Lorenz attractor [example](https://neuralpde.sciml.ai/stable/tutorials/param_estim/), I...
[Neural Kernel Tangent Adaptive Loss](https://github.com/SciML/NeuralPDE.jl/issues/501) - [Paper (Section 3) and algorithm 1](https://arxiv.org/pdf/2007.14527.pdf) [Inverse Dirichlet Adaptive Loss](https://github.com/SciML/NeuralPDE.jl/issues/500) - [Paper (Section 3.2)](https://iopscience.iop.org/article/10.1088/2632-2153/ac3712/pdf)
Hey folks, I am running into another problem related to integrals. I can't seem to get GPU support working correctly with them. The script I wrote runs correctly when the...
Hey folks, I am moving the issue here from the Julia Slack: I am running into a problem where the loss function generated from PDEs that have an integral in...
Hi, I noticed that there has been a major update from version 4 to version 5, there are some changes to the functions and no more documentation on low level...
I am currently trying to implement the additional loss function using Lux so that I can run using GPU. I have used the following settings: function additional_loss(phi, θ , p)...
Hi, I was about to solve a multi-dimensional Stochastic PDE like the one that used to be in the documentation. No wonder why you removed it, the old code does...
Before merging #578 (commit 15fece6), the grid training strategy was generating PDE training sets that did not share coordinates with the boundary training sets. Since this PR was merged, this...
## Checklist - [ ] Appropriate tests were added - [ ] Any code changes were done in a way that does not break public API - [ ] All...
Added in StochasticTraining strategy for generate_loss function ## Checklist - [ ] Appropriate tests were added - [ ] Any code changes were done in a way that does not...
