ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 06:49:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

add batched_transpose implement (#1660)

Browse Source 
* add batched_transpose implement

---------

Co-authored-by: root <root@ctr-ubbsmc16.amd.com>
Co-authored-by: ThruptiRajLakshmanaGowda <tlakshma@amd.com>
Co-authored-by: ThomasNing <thomas.ning@amd.com>
This commit is contained in:
fangche123
2025-01-29 08:22:02 +08:00
committed by GitHub
parent d6a4605e1c
commit c5fff071e5
14 changed files with 760 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
example/ck_tile/35_batched_transpose/script/smoke_test.sh Executable file
View File

@@ -0,0 +1,11 @@
#!/bin/sh
EXE=./build/bin/tile_example_batched_transpose
for pr in "fp32" "fp16" "int8" ; do
$EXE -pr=$pr -N=1 -C=32 -H=1 -W=32 -layout_in='NCHW' -layout_out='NHWC'
$EXE -pr=$pr -N=2 -C=12 -H=1 -W=32 -layout_in='NHWC' -layout_out='NCHW'
$EXE -pr=$pr -N=3 -C=1334 -H=1 -W=37 -layout_in='NHWC' -layout_out='NCHW'
$EXE -pr=$pr -N=4 -C=27 -H=1 -W=32 -layout_in='NCHW' -layout_out='NHWC'
$EXE -pr=$pr -N=5 -C=1234 -H=1 -W=12 -layout_in='NCHW' -layout_out='NHWC'
done