I found there’s an update pipe for parallel computing using GPU, use this to improve it,
Original code:
pipe.to(“gpu”)
Then replace with:
pipe.to(device=“cuda”, dtype=torch.float16)
I found there’s an update pipe for parallel computing using GPU, use this to improve it,
Original code:
pipe.to(“gpu”)
Then replace with:
pipe.to(device=“cuda”, dtype=torch.float16)