A tutorial on pin_memory and non_blocking usage #2580

Job	Run time
pytorch_tutorial_build_worker (1, 15, linux.g5.12xlarge.nvidia.gpu)	59s
pytorch_tutorial_build_worker (2, 15, linux.g5.4xlarge.nvidia.gpu)	23s
pytorch_tutorial_build_worker (3, 15, linux.g5.4xlarge.nvidia.gpu)	22s
pytorch_tutorial_build_worker (4, 15, linux.g5.4xlarge.nvidia.gpu)	17s
pytorch_tutorial_build_worker (5, 15, linux.g5.4xlarge.nvidia.gpu)	1s
pytorch_tutorial_build_worker (6, 15, linux.g5.4xlarge.nvidia.gpu)	1s
pytorch_tutorial_build_worker (7, 15, linux.4xlarge.nvidia.gpu)	1m 10s
pytorch_tutorial_build_worker (8, 15, linux.4xlarge.nvidia.gpu)	1m 14s
pytorch_tutorial_build_worker (9, 15, linux.4xlarge.nvidia.gpu)	1m 9s
pytorch_tutorial_build_worker (10, 15, linux.4xlarge.nvidia.gpu)	1m 10s
pytorch_tutorial_build_worker (11, 15, linux.4xlarge.nvidia.gpu)	1m 15s
pytorch_tutorial_build_worker (12, 15, linux.4xlarge.nvidia.gpu)	1m 20s
pytorch_tutorial_build_worker (13, 15, linux.4xlarge.nvidia.gpu)	1m 14s
pytorch_tutorial_build_worker (14, 15, linux.4xlarge.nvidia.gpu)	1m 13s
pytorch_tutorial_build_worker (15, 15, linux.4xlarge.nvidia.gpu)	1m 14s
pytorch_tutorial_build_manager	1s
	13m 3s

Provide feedback