Bartłomiej Kocot a860c20099 Add NGCHW bf16 grouped conv fwd instances (#1783) ...

* Add NGCHW bf16 grouped conv fwd instances

* add missed cmake

[ROCm/composable_kernel commit: 159fa31946]

2025-01-01 18:00:06 +01:00

..

batched_gemm

…

batched_gemm_gemm

…

batched_gemm_multi_d

…

batched_gemm_reduce

…

batched_gemm_softmax_gemm

…

batched_gemm_softmax_gemm_permute

…

batchnorm

…

block_swizzle_test

…

block_to_ctile_map

…

ck_tile

[CK TILE] GEMM and Batched GEMM SplitK support (#1724)

2024-12-28 14:40:17 +01:00

contraction

…

conv_tensor_rearrange

…

conv_util

…

convnd_bwd_data

…

convnd_fwd

…

data_type

Add copy assignment op test (#1718)

2024-12-06 09:56:27 -06:00

elementwise_normalization

…

gemm

…

gemm_add

…

gemm_layernorm

…

gemm_reduce

…

gemm_split_k

…

gemm_universal

…

grouped_convnd_bwd_data

Support large batch tensors in grouped conv bwd data (#1711)

2024-12-06 10:55:23 +01:00

grouped_convnd_bwd_weight

…

grouped_convnd_fwd

Add NGCHW bf16 grouped conv fwd instances (#1783)

2025-01-01 18:00:06 +01:00

grouped_gemm

…

magic_number_division

…

normalization_bwd_data

…

normalization_bwd_gamma_beta

…

normalization_fwd

…

permute_scale

…

pool

OCP FP8 support for gfx12. (#1710)

2024-12-03 08:42:55 -08:00

position_embedding

…

reduce

…

reference_conv_fwd

…

scatter_gather

…

smfmac_op

…

softmax

…

space_filling_curve

…

transpose

…

wmma_op

…

wrapper

…

CMakeLists.txt

…