TorchFort
TorchFort copied to clipboard
NVIDIA
An Online Deep Learning Interface for HPC programs on NVIDIA GPUs
This PR refactors the RL code for more flexibility. This was necessary for adding on-policy routines, which is also done with PPO along with all C and Fortran wrapper routines....
This PR introduces more flexible training/inference functions, `torchfort_train_multiarg` and `torchfort_inference_multiarg`, designed to support models that require multiple input/output tensor arguments. To facilitate passing multiple tensors from Fortran/C, we introduce a...
Hi ! My goal is to build a Multi-agent Reinforcement Learning model for a fluid simulation (https://github.com/ExtremeFLOW/neko) in Fortran. For this I have 3 state vectors (s1, s2, s3) and...
Hi ! Is it possible to run multiple simulations in parallel, all updating the same policy / critic network using TorchFort? If so can you please share some insights into...
This MR adds gradient clipping to all training procedures.
I am encountering an issue when attempting to run a simulation on the GPU after successfully building the Docker image using the provided Dockerfile_gnu. ``` root@8ef7710fa182:/opt/torchfort/bin/examples/fortran/simulation# ./train OpenACC support required...
Firstly, is there an alternative forum for general support? Calling the below a code development "issue" sounds like a category error to me, but I don't see anywhere else to...
