How can I call multiple GPUs during training

autonomousvision / convolutional_occupancy_networks

[ECCV'20] Convolutional Occupancy Networks

https://pengsongyou.github.io/conv_onet

MIT License

830 stars 113 forks source link

How can I call multiple GPUs during training #43

Closed miaowang525 closed 2 years ago

miaowang525 commented 2 years ago

hello Your work is great and i am very interested in your article, and I have some questions to ask you. I have five GPUs, how to call multiple GPUs to run code during training @pengsongyou

pengsongyou commented 2 years ago

Hi @miaowang525 , I did not try to train with multiple GPUs for this project, but it should be straightforward with DataParallel in PyTorch (simply do net = torch.nn.DataParallel(model, device_ids=[0, 1, 2]), or DistributedDataParallel (recommended by PyTorch, but required more modifications).

Hope it helps.

Best, Songyou