Skip to content

Commit 9402062

Browse files
amd-ruitang3CopilotgyohuangxinvalarLip
authored
remove_iris_from_setup (ROCm#1644)
* remove_iris_from_setup * Update setup.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> * update * update * update --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Xin Huang <Xin.Huang@amd.com> Co-authored-by: Lingpeng Jin <103567126+valarLip@users.noreply.github.com>
1 parent b11fc9a commit 9402062

12 files changed

Lines changed: 30 additions & 38 deletions

aiter/configs/bf16_tuned_gemm.csv

Lines changed: 0 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1 @@
11
cu_num,M,N,K,bias,dtype,outdtype,scaleAB,bpreshuffle,libtype,solidx,splitK,us,kernelName,err_ratio,tflops,bw
2-
80,64,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,1,10,9.3967,_ZN5aiter30bf16gemm_fp32bf16_tn_32x64_pf3E,0.0,17.85,355.69
3-
80,80,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,2,10,9.7635,_ZN5aiter30bf16gemm_fp32bf16_tn_48x64_pf3E,0.0,21.48,360.79
4-
80,128,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,3,10,10.8196,_ZN5aiter30bf16gemm_fp32bf16_tn_64x64_pf3E,0.0,31.01,375.54
5-
80,150,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,2,5,13.2734,_ZN5aiter30bf16gemm_fp32bf16_tn_48x64_pf3E,0.0,29.62,324.79
6-
80,192,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,2,5,15.0013,_ZN5aiter30bf16gemm_fp32bf16_tn_48x64_pf3E,0.0,33.55,318.91
7-
80,220,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,3,5,14.6675,_ZN5aiter30bf16gemm_fp32bf16_tn_64x64_pf3E,0.0,39.32,347.67
8-
80,256,256,5120,False,torch.bfloat16,torch.float32,False,False,asm,3,5,16.2047,_ZN5aiter30bf16gemm_fp32bf16_tn_64x64_pf3E,0.0,41.41,339.72
-25.8 KB
Binary file not shown.
-29.7 KB
Binary file not shown.
-12 KB
Binary file not shown.
-16.1 KB
Binary file not shown.
-16 KB
Binary file not shown.
-18.1 KB
Binary file not shown.
-18 KB
Binary file not shown.
-20 KB
Binary file not shown.
-21.9 KB
Binary file not shown.

0 commit comments

Comments
 (0)