Method init_parallel_env

lightx2v_platform/base/ppu_cuda.py:30–32 · view source on GitHub ↗

()

Source from the content-addressed store, hash-verified

28
29	@staticmethod
30	def init_parallel_env():
31	dist.init_process_group(backend="nccl")
32	torch.cuda.set_device(dist.get_rank())

nothing calls this directly

init_process_groupMethod · 0.80

set_deviceMethod · 0.80

no test coverage detected