diff options
| author | Jakub Sujak <jakub.sujak@arm.com> | 2023-04-27 09:24:05 +0100 |
|---|---|---|
| committer | Jakub Sujak <jakub.sujak@arm.com> | 2023-05-12 09:00:33 +0000 |
| commit | 56fabbae2309856f74151c0bc909d15d84951a2c (patch) | |
| tree | 47da25ab4c124e146d3fb83fa923921cba74b06a /src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp | |
| parent | 7997603de02e3d9d901b80988c044d1184b2c069 (diff) | |
| download | ComputeLibrary-56fabbae2309856f74151c0bc909d15d84951a2c.tar.gz | |
Fix performance regression in FP16 Deconvolution
The previous heuristic for selecting the Deconvolution method with FP32 input data introduced a performance regression for FP16. A simple fix ensures the previous heuristic applies to FP32 types only.
Resolves: COMPMID-6027
Change-Id: I77ca6c9c72534057a3967db58924a972b0efb09f
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9616
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp')
0 files changed, 0 insertions, 0 deletions