projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5465688) | patch
[neon] Optimzie sgemm_transB
authorskykongkong8 <ss.kong@samsung.com>
Wed, 11 Oct 2023 06:29:46 +0000 (15:29 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Wed, 11 Oct 2023 22:31:50 +0000 (07:31 +0900)
commit20e0c12e3a9d2fbdbc95962eb2f29629f108f8b4
tree154be5a9017b0c763ef0a09d45bdba5db3f9b9e3tree | snapshot
parent54656886335b6e05e081de50bf6fb55e6c9308eccommit | diff
[neon] Optimzie sgemm_transB

- We can reduce the number of function calls by re-using the register variable
- This optimization is valid especially for large scale Tensor

**Self evaluation:**
1. Build test:     [X]Passed [ ]Failed [ ]Skipped
2. Run test:     [X]Passed [ ]Failed [ ]Skipped

Signed-off-by: skykongkong8 <ss.kong@samsung.com>
nntrainer/tensor/blas_neon.cpp diff | blob | history
Domain: Machine Learning / ML Framework;