DiffEqGPU.jl icon indicating copy to clipboard operation
DiffEqGPU.jl copied to clipboard

GPU-acceleration routines for DifferentialEquations.jl and the broader SciML scientific machine learning ecosystem

Results 46 DiffEqGPU.jl issues
Sort by recently updated
recently updated
newest added

The documentation says `CuArray` `u0` are supported. But only one dimensional arrays are supported because of lines like: https://github.com/SciML/DiffEqGPU.jl/blob/8489a3e74f9179aafddb38912bedd9321fadd68b/src/DiffEqGPU.jl#L345 To generalize this to N dimensions the `u0` should concatenate across...

https://github.com/SciML/DiffEqGPU.jl/pull/170 The latest profile, while solving from `EnsembleGPUKernel`, raises some questions: ![](https://user-images.githubusercontent.com/37050056/182220489-9e82dcf8-2b3c-41d0-a876-e413d96468e5.png) Some overheads are discussed here for potential improvements `EnsembleGPUKernel` for `Tsit5`. 1. Converting the solution back to CuArrays....

Hello all! This is related to [this post](https://discourse.julialang.org/t/passing-jacobian-generated-from-modelingtoolkit-to-odefunction-for-ensemblegpuarray/84198) where I was attempting to recreate the lorentz equations example in the DiffEqGPU.jl but instead of providing numerical functions I wanted to...

Hi, for my application I would like to simulate a large number of trajectories (1000 to 20000) of the same system with different parameters but also with different time spans...

Hello i am trying to solve an ODE system for different parameter vectors. However, some of the parameters lead to imaginary or unstable solutions. Is it possible to just skip...

upstream

Hi! Using the Lorenz example in the README, EnsembleGPUArray seems to be running quite a bit slower than all other methods, including EnsembleSerial. On my machine I get: ``` using...

Currently being prototyped in gpu_ode.jl with SimpleDiffEq.jl. Most of the issues that need to be worked out are in GPUifyLoops.jl

```julia using OrdinaryDiffEq using DiffEqGPU function f1(out,du,u,p,t) out[1] = p[1]*(u[2]-u[1]) - du[1] out[2] = u[1]*(p[2]-u[3]) - u[2] - du[2] out[3] = u[1]*u[2] - p[3]*u[3] - du[3] end u₀= Float32[1.0;0.0;0.0] du₀...

Based on the DiffEq documentation, I've written this minimal example. If I set `trajectories = 1`, it compiles properly and everything is fine. But when `trajectories = 32`, say, I...

Hello, I'm trying to solve and ODE with my GPU. I tested the example code present in the [DiffEqGPU.js](https://github.com/SciML/DiffEqGPU.jl) repository, and it worked fine. I have the following function to...

upstream