diff options
| author | Viet-Hoa Do <viet-hoa.do@arm.com> | 2023-03-01 15:46:10 +0000 |
|---|---|---|
| committer | Viet-Hoa Do <viet-hoa.do@arm.com> | 2023-03-02 09:46:28 +0000 |
| commit | e2e6d745c940cdfd8c3340fd1227dbef1badfb3c (patch) | |
| tree | 9af6bc878aa397025524d2eca303e0d4a5a1693a /src/runtime/NEON/functions/NEDeconvolutionLayer.cpp | |
| parent | bbf2e7477be984702e1a51f2a23910ee8349b867 (diff) | |
| download | ComputeLibrary-e2e6d745c940cdfd8c3340fd1227dbef1badfb3c.tar.gz | |
Fix direct conv2d in dynamic fusion
* Put input and output tensor shape value directly to the CL code.
* Use texture for weights when it is possible.
Resolves: COMPMID-5938
Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Change-Id: Ib53b310a80ce857eac36564b352136fdde55b131
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9249
Reviewed-by: SiCong Li <sicong.li@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/NEON/functions/NEDeconvolutionLayer.cpp')
0 files changed, 0 insertions, 0 deletions