leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 28 Packages Projects Releases Wiki Activity

CUTLASS 3.4.0 (#1286)

Browse Source 
* CUTLASS 3.4.0

* Update CHANGELOG.md

---------

Co-authored-by: Pradeep Ramani <prramani@nvidia.com>
This commit is contained in:
Pradeep Ramani
2023-12-29 15:21:31 -05:00
committed by GitHub
co-authored by Pradeep Ramani
parent b7508e3379
commit 8236f30675
211 changed files with 11409 additions and 2763 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/cutlass/layout/tensor.h
+8 -3
View File
@@ -60,9 +60,6 @@ namespace layout {
//
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag used for 3-D NWC tensors for 1D conv, only used in 3.x API
class TensorNWC {};
/// Mapping function for 4-D NHWC tensors.
class TensorNHWC {
public:
@@ -632,6 +629,14 @@ public:
}
};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag used for linearized tensors with shape (NW, C) for 1D conv, only used in 3.x API
class TensorLinearizedNWC {};
/// Tag used for linearized tensors with shape (NHW, C) for 2D conv, only used in 3.x API
class TensorLinearizedNHWC : public TensorNHWC {};
/// Tag used for linearized tensors with shape (NDHW, C) for 3D conv, only used in 3.x API
class TensorLinearizedNDHWC : public TensorNDHWC {};
/////////////////////////////////////////////////////////////////////////////////////////////////