diff options
| author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2019-07-17 15:59:32 +0100 |
|---|---|---|
| committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2019-07-17 16:22:42 +0000 |
| commit | ff2719299ea76a95f20a35a7900875a8152e293a (patch) | |
| tree | e7183168ea86d92e1a244cea78cac8a375e97ad3 /arm_compute/core/CL | |
| parent | ca1f460ec33e84b9df84e29de3c3b733e6042b9c (diff) | |
| download | ComputeLibrary-ff2719299ea76a95f20a35a7900875a8152e293a.tar.gz | |
COMPMID-2336: Fix validation for quantized NEDepthwiseConvolutionLayer
Assertions from calculate_quantized_multiplier_less_than_one were not
captured resulting in incorrect computation being performed on invalid
inputs.
This patch also fixes a discrepancy between the interfaces of validate
and configure methods of NEDirectConvolutionLayerOutputStageKernel.
Change-Id: I011822f63b4062b5b3346ef047adc684e0bffa3d
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1552
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/CL')
0 files changed, 0 insertions, 0 deletions