SunGeng Blog

练得身形似鹤形，千株松下两函经

高性能计算基础知识

High performance computing basics

随缘整理，其实还有很多知识需要记录，持续更新...

Posted by     SunGeng on Sunday, April 28, 2024

深入理解Pytorch源码中的CUDA算子01-BlockReduceSum

探究下pytorch是如何高效的对一个block中的数据进行规约的

Posted by     SunGeng on Saturday, April 27, 2024

avatar

High-performance computing engineer, CUDA operator developer, AI beginner.

FEATURED TAGS
cuda pytorch twikoo 建站
FRIENDS