MIC-DKFZ / MedNeXt

[MICCAI 2023] MedNeXt is a fully ConvNeXt architecture for 3D medical image segmentation.
https://arxiv.org/pdf/2303.09975
Apache License 2.0
345 stars 26 forks source link

How to use RTX 4090 *2 running train #35

Open Fucapisun opened 5 days ago

Fucapisun commented 5 days ago

When I train with two GPUs using CUDA, I get a memory error message, torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 416.00 MiB (GPU 0; 23.64 GiB total capacity; 21.07 GiB already allocated; 405.69 MiB free; 22.77 GiB reserved in total by PyTorch)indicating that only one GPU is being called. Is there any way to train with two GPUs simultaneously?