diff options
author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2020-10-08 10:25:49 +0100 |
---|---|---|
committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2020-10-19 12:18:07 +0000 |
commit | 7333e1f10f5da9dc67b511d326121a843771a107 (patch) | |
tree | 9b36047d0f13846406680bc01a2cb8df8660de66 /src/graph/mutators/NodeFusionMutator.cpp | |
parent | e4558b501bc4a8e4e731517916a29fb1594d2a78 (diff) | |
download | ComputeLibrary-7333e1f10f5da9dc67b511d326121a843771a107.tar.gz |
COMPMID-3732: Remove OpenCL padding from CLPoolingLayer
- Refactor pooling layer kernels on OpenCL (F32/F16/QASYMM8) to avoid
padding and improve performance
- Add test for checking zero padding requirement
- Fix issue with extracting the index. The issue was caused by the
padding passed at compile time
- auto_init indices tensor in CLPoolingLayerKernel
Change-Id: I1ae5a2ef8c4ce787c80dcd73e35c17bb34623cb5
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4188
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/graph/mutators/NodeFusionMutator.cpp')
0 files changed, 0 insertions, 0 deletions