TiledTensor / TiledCUDA

TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.
MIT License
157 stars 10 forks source link

feat(cell): Add a Column-Major reduce implementation. #105

Closed KuangjuX closed 3 months ago

KuangjuX commented 3 months ago

Resolved #99 .