diff options
author | Michalis Spyrou <michalis.spyrou@arm.com> | 2020-07-31 15:28:18 +0100 |
---|---|---|
committer | Michalis Spyrou <michalis.spyrou@arm.com> | 2020-08-03 10:21:22 +0000 |
commit | ad411e077ef6662f1fa48542eacaa43552b12e65 (patch) | |
tree | a38ead8326c7b5aa7abeabb3e0971fee030da108 /src/core/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp | |
parent | 3b64e3e78c166773fea680afb4829d886e90552d (diff) | |
download | ComputeLibrary-ad411e077ef6662f1fa48542eacaa43552b12e65.tar.gz |
COMPMID-3526: LOGISTIC support for values bigger than [-40.f,40.f]
With this patch: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3178
we can support any range of values since we handle overflows by clamping.
This means that for large negative values we'll get 0 and for possitive inf
which aligns with math.h implementation.
Change-Id: I01e92010bb0c514c12b19da97e369a75d782cac7
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3639
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/GLES_COMPUTE/kernels/GCGEMMMatrixAdditionKernel.cpp')
0 files changed, 0 insertions, 0 deletions