diff options
author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2019-08-15 15:00:37 +0100 |
---|---|---|
committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2019-08-20 17:02:40 +0100 |
commit | 7c2d92c477e893b797d8db0a1bb1beffe7c26a63 (patch) | |
tree | 61a3c42893c8366f60359792543f29bbe15a8b65 /src/core/NEON/kernels/NETileKernel.cpp | |
parent | d460c29db6fb51cafba7b5591cacc2ce4a7f4592 (diff) | |
download | ComputeLibrary-7c2d92c477e893b797d8db0a1bb1beffe7c26a63.tar.gz |
COMPMID-2590: Avg_Pooling2d Fails on CL NHWC FP16
For large tensors with large pooling sizes, accumulation on FP16 gets
saturated. This patch provides accumulation on F32 to overcome this
issue.
Change-Id: I5b2985c801970c8f2cb0053442296cc69f854186
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1762
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/NETileKernel.cpp')
0 files changed, 0 insertions, 0 deletions