A tutorial on pin_memory and non_blocking usage #2599

Job	Run time
pytorch_tutorial_build_worker (1, 15, linux.g5.12xlarge.nvidia.gpu)	11m 57s
pytorch_tutorial_build_worker (2, 15, linux.g5.4xlarge.nvidia.gpu)	11m 56s
pytorch_tutorial_build_worker (3, 15, linux.g5.4xlarge.nvidia.gpu)	11m 56s
pytorch_tutorial_build_worker (4, 15, linux.g5.4xlarge.nvidia.gpu)	11m 56s
pytorch_tutorial_build_worker (5, 15, linux.g5.4xlarge.nvidia.gpu)	11m 56s
pytorch_tutorial_build_worker (6, 15, linux.g5.4xlarge.nvidia.gpu)	11m 56s
pytorch_tutorial_build_worker (7, 15, linux.4xlarge.nvidia.gpu)	11m 51s
pytorch_tutorial_build_worker (8, 15, linux.4xlarge.nvidia.gpu)	11m 51s
pytorch_tutorial_build_worker (9, 15, linux.4xlarge.nvidia.gpu)	11m 50s
pytorch_tutorial_build_worker (10, 15, linux.4xlarge.nvidia.gpu)	11m 53s
pytorch_tutorial_build_worker (11, 15, linux.4xlarge.nvidia.gpu)	11m 53s
pytorch_tutorial_build_worker (12, 15, linux.4xlarge.nvidia.gpu)	11m 53s
pytorch_tutorial_build_worker (13, 15, linux.4xlarge.nvidia.gpu)	11m 53s
pytorch_tutorial_build_worker (14, 15, linux.4xlarge.nvidia.gpu)	11m 52s
pytorch_tutorial_build_worker (15, 15, linux.4xlarge.nvidia.gpu)	11m 52s
pytorch_tutorial_build_manager	1s
	2h 58m 26s

Provide feedback