Multiple GPU inferencing

#13
by cjj2003 - opened

Is it possible to run this on multiple GPUs? I tried the demo on 4 A6000s and it didn't only seemed to use the first GPU.

Hey @cjj2003 you can use this model to inference on multiple GPU's. Use this:

if num_gpus > 1:
    model = torch.nn.DataParallel(model, device_ids=list(range(num_gpus)))
model.to(device)
amanrangapur changed discussion status to closed

Sign up or log in to comment