projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5163aa4) | patch
[ hgemm ] Consider tiny gemm case
authorskykongkong8 <ss.kong@samsung.com>
Thu, 9 May 2024 10:50:59 +0000 (19:50 +0900)
committerMyungJoo Ham <myungjoo.ham@samsung.com>
Mon, 10 Jun 2024 09:43:30 +0000 (18:43 +0900)
commit2715af16f36a351efdf0fbea4d8db7febeddebad
tree42d7204540ad7256375b94a7a13db4556906f915tree | snapshot
parent5163aa4b09e4b70e0526d6caa637a74d5294c7a2commit | diff
[ hgemm ] Consider tiny gemm case

- Current hard-coded lenght kernel hgemm, matrix length that are smaller than unit vector length might fail.
- Toss such case to fallback. Such small sized hgemm will take a few nanosecond latency, therefore negligible.

**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/hgemm/hgemm.cpp diff | blob | history
Domain: Machine Learning / ML Framework;