[hgemm] hgemm noTrans with 1x4 kernel
authorDebadri Samaddar <s.debadri@samsung.com>
Tue, 23 Apr 2024 06:30:16 +0000 (12:00 +0530)
committerJijoong Moon <jijoong.moon@samsung.com>
Thu, 25 Apr 2024 23:07:39 +0000 (08:07 +0900)
commitf9a4cd48336dac92e1f0ee357c0a6b4719b21c42
treeac8a343ad0ffb6aa60fe6b45c514d737bd2706e5
parent467c21c8b075fab03bac4e1e81aad242b1903469
[hgemm] hgemm noTrans with 1x4 kernel

Added hgemm_kernel_1x4
Added hgemm_noTrans_1x4 calls
Added unittest dot_gemm_50_768_516

Signed-off-by: Debadri Samaddar <s.debadri@samsung.com>
nntrainer/tensor/hgemm/hgemm.cpp
nntrainer/tensor/hgemm/hgemm.h
nntrainer/tensor/hgemm/hgemm_kernel_1x4.h [new file with mode: 0644]
test/unittest/unittest_nntrainer_tensor_neon_fp16.cpp