A tutorial on pin_memory and non_blocking usage #2585

Job	Run time
pytorch_tutorial_build_worker (1, 15, linux.g5.12xlarge.nvidia.gpu)	15m 3s
pytorch_tutorial_build_worker (2, 15, linux.g5.4xlarge.nvidia.gpu)	51m 17s
pytorch_tutorial_build_worker (3, 15, linux.g5.4xlarge.nvidia.gpu)	18m 19s
pytorch_tutorial_build_worker (4, 15, linux.g5.4xlarge.nvidia.gpu)	25m 42s
pytorch_tutorial_build_worker (5, 15, linux.g5.4xlarge.nvidia.gpu)	19m 12s
pytorch_tutorial_build_worker (6, 15, linux.g5.4xlarge.nvidia.gpu)	16m 21s
pytorch_tutorial_build_worker (7, 15, linux.4xlarge.nvidia.gpu)	20m 32s
pytorch_tutorial_build_worker (8, 15, linux.4xlarge.nvidia.gpu)	19m 34s
pytorch_tutorial_build_worker (9, 15, linux.4xlarge.nvidia.gpu)	37m 8s
pytorch_tutorial_build_worker (10, 15, linux.4xlarge.nvidia.gpu)	42m 25s
pytorch_tutorial_build_worker (11, 15, linux.4xlarge.nvidia.gpu)	31m 43s
pytorch_tutorial_build_worker (12, 15, linux.4xlarge.nvidia.gpu)	19m 46s
pytorch_tutorial_build_worker (13, 15, linux.4xlarge.nvidia.gpu)	22m 29s
pytorch_tutorial_build_worker (14, 15, linux.4xlarge.nvidia.gpu)	14m 34s
pytorch_tutorial_build_worker (15, 15, linux.4xlarge.nvidia.gpu)	23m 34s
pytorch_tutorial_build_manager	24m 13s
	6h 41m 52s

Provide feedback