mirror of
https://github.com/pytorch/pytorch.git
synced 2025-10-20 21:14:14 +08:00
Updated CUDA basics (markdown)
@ -31,5 +31,6 @@ TensorIterator kernels
|
||||
* Use pytorch built-in profiler (kineto) or nsys to get information on GPU utilization and most time-consuming kernels
|
||||
|
||||
See https://www.dropbox.com/s/3350s3qfy8rpm5a/CUDA_presentation.key?dl=0 for high level overview of CUDA programming and useful links
|
||||
|
||||
Goto N1023015 to do TensorIterator cuda perf lab
|
||||
|
||||
|
Reference in New Issue
Block a user