That code from early 2020 would need some updates. GPU-based MNIST Neural ODE Classifier · DiffEqFlux.jl might be a better starting point?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Struggling to train a UDE model with a GPU | 3 | 406 | February 14, 2023 | |
| GPU simulation time of neural ODE program | 8 | 916 | August 16, 2019 | |
| DiffEqFlux neural_ode used with Flux.Train! is slower on GPU than CPU | 0 | 738 | December 31, 2019 | |
| MethodError: no method matching getrf! when I try to use GPU to training a universal differential equation with ForwardDiff.jl | 5 | 335 | December 6, 2022 | |
| Calculate gradient of UDE model (without DiffEqFlux.sciml_train) | 3 | 727 | November 6, 2020 |