ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-12 01:10:17 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
39abb4704aa829321bd109c4d71b86061d947daf
composable_kernel/example
History
Shaojie WANG 40942b9098 Optimization for gridwise group norm (#453) 
* use another instance to check the efficiency

* optimize group layer norm

* 1. coalesce load/store data for gridwise layer norm welford. 2. move a sqrt and divison into a outer static loop

* add more instances to layernorm

* add 2 more test cases

* remove ignore in generating tuple of vector

Co-authored-by: Chao Liu <chao.liu2@amd.com>
2022-10-06 21:24:13 -05:00
..
01_gemm
add more datatype to gemm+gemm and conv+conv example (#397)
2022-09-01 09:31:17 -05:00
02_gemm_bilinear
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
03_gemm_bias_relu
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
04_gemm_add_add_fastgelu
Add example of Gemm + AddAddFastGelu (data type: int4) (#369)
2022-08-23 10:38:41 -05:00
09_convnd_fwd
add more datatype to gemm+gemm and conv+conv example (#397)
2022-09-01 09:31:17 -05:00
10_convnd_fwd_multiple_d_multiple_reduce
Add examples of Conv + reduction (data type: int4, int8, bf16, fp16, fp32) (#380)
2022-08-31 16:32:17 -05:00
12_reduce
Add int4 reduction examples (#372)
2022-08-25 16:58:48 -05:00
13_pool2d_fwd
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
14_gemm_xdl_requant_relu_requant
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
15_grouped_gemm
GEMM batched/splitK/cgemm/grouped int4 examples (#383)
2022-08-25 17:19:15 -05:00
16_gemm_multi_d_multi_reduces
Gemm reduce examples int4/int8/fp32/bf16 (#368)
2022-08-30 11:38:26 -05:00
17_convnd_bwd_data
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
18_batched_gemm_reduce
Fused attention (#345)
2022-08-13 00:16:14 -05:00
19_binary_elementwise
Batchnorm-forward and Batchnorm-infer Implemented using generic kernels (#320)
2022-08-15 10:11:02 -05:00
20_convnd_bwd_weight
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
21_gemm_layernorm
Batchnorm-forward and Batchnorm-infer Implemented using generic kernels (#320)
2022-08-15 10:11:02 -05:00
22_cgemm
GEMM batched/splitK/cgemm/grouped int4 examples (#383)
2022-08-25 17:19:15 -05:00
23_softmax
Softmax client example (#396)
2022-09-06 12:22:48 -05:00
24_batched_gemm
GEMM batched/splitK/cgemm/grouped int4 examples (#383)
2022-08-25 17:19:15 -05:00
25_gemm_bias_e_permute
add g; fixed strides (#355)
2022-08-12 15:22:39 -05:00
26_contraction
Clean up conv example, Instances, profiler and test (#324)
2022-07-29 18:19:25 -05:00
27_layernorm
Group norm (#417)
2022-09-19 22:30:46 -05:00
28_grouped_gemm_bias_e_permute
Add batched/grouped_gemm contraction deviceOps (#349)
2022-08-10 12:20:29 -05:00
29_batched_gemm_bias_e_permute
Add batched/grouped_gemm contraction deviceOps (#349)
2022-08-10 12:20:29 -05:00
30_grouped_convnd_fwd_bias_relu_add
Conv bwd data multiple d (#404)
2022-09-19 11:25:28 -05:00
31_batched_gemm_gemm
add more datatype to gemm+gemm and conv+conv example (#397)
2022-09-01 09:31:17 -05:00
32_batched_gemm_scale_softmax_gemm
MNKO padding support on bmm+masking+scale+softmax+bmm+premute (#425)
2022-09-20 12:43:53 -05:00
33_multiple_reduce
Batchnorm-forward and Batchnorm-infer Implemented using generic kernels (#320)
2022-08-15 10:11:02 -05:00
34_batchnorm
Batchnorm-forward and Batchnorm-infer Implemented using generic kernels (#320)
2022-08-15 10:11:02 -05:00
35_splitK_gemm
GEMM batched/splitK/cgemm/grouped int4 examples (#383)
2022-08-25 17:19:15 -05:00
36_sparse_embedding
embedding fuse layernorm (#405)
2022-09-09 10:41:15 -05:00
37_batched_gemm_add_add_relu_gemm_add
batched_gemm + multiple_d + gemm + multiple_d (#394)
2022-09-14 17:54:18 -05:00
38_grouped_conv_bwd_data_bias_relu
Conv bwd data multiple d (#404)
2022-09-19 11:25:28 -05:00
39_permute
Add 'Permute' device op & example (#408)
2022-09-19 21:30:25 -05:00
41_grouped_conv_conv_fwd
add more datatype to gemm+gemm and conv+conv example (#397)
2022-09-01 09:31:17 -05:00
42_groupnorm
Optimization for gridwise group norm (#453)
2022-10-06 21:24:13 -05:00
CMakeLists.txt
Conv bwd data multiple d (#404)
2022-09-19 11:25:28 -05:00