pescap
pescap
Work in progress. Add the hvd support, to use multiple GPUs. Sounds to be working properly (tested with 1 up to 8 GPUs) and is non-intrusive! Later, we could add...
### Discussed in https://github.com/lululxvi/deepxde/discussions/700 Originally posted by **pescap** May 25, 2022 Hi, https://github.com/lululxvi/deepxde/blob/4cd2178dfc70e06c629c858870a5f64f622982b6/examples/pinn_forward/Poisson_Lshape.py#L25 freezes when using `tensorflow` as a backend, in CPU and GPU modes. Example with 5000 ADAM epochs:...
Hi @stevenliuyi, It sounds like the formula for R0 gives liminf(R0)=1. Isn't there another formula giving a better estimation of the reproduction number? (with 0
Hi, Countries are: Germany, Spain, Brasil, Chile amongst others. Any idea of how to solve it? Best wishes, 
Hi, I am currently working with @estebaniglesias towards adding the "confirmedCount" for "Comunas" in Chile. Comunas are equivalent to subRegions. Data are available at: https://github.com/ivanMSC/COVID19_Chile/blob/master/historial_comunas.csv with: Date | Region |...
I propose to implement the [weighted GMRES(m)](http://eprints.maths.manchester.ac.uk/2084/), as this variant can lead to mesh independent convergence results for FEM/BEM simulations (see e.g. [this article](https://www.sciencedirect.com/science/article/abs/pii/S0898122121003606)). The weighted version requires only a...
Work in progress!
- [ ] Weak scaling for DeepONets - `Triple` - `TripleCartesianProd` - [ ] Strong scaling for DeepONets - `Triple` - `TripleCartesianProd`