diff options
| author | Moritz Pflanzer <moritz.pflanzer@arm.com> | 2017-07-28 09:09:16 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:16:42 +0100 |
| commit | f856c74fc4fd0524cf9f6594d9bcdd0a7f699095 (patch) | |
| tree | 094742359ff14213a636491611a96239ca7f4197 /tests/validation/CL/ConvolutionLayer.cpp | |
| parent | ec2de0f0784ece788e6d9aa9d2af1f9a8c4175b1 (diff) | |
| download | ComputeLibrary-f856c74fc4fd0524cf9f6594d9bcdd0a7f699095.tar.gz | |
COMPMID-417: Change rounding mode of half reference
Change-Id: Ib72bd0ce1cc0ea5fb474cb6f1b09a030a4162bb8
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/81922
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Diffstat (limited to 'tests/validation/CL/ConvolutionLayer.cpp')
| -rw-r--r-- | tests/validation/CL/ConvolutionLayer.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tests/validation/CL/ConvolutionLayer.cpp b/tests/validation/CL/ConvolutionLayer.cpp index 6082ab4924..570077120e 100644 --- a/tests/validation/CL/ConvolutionLayer.cpp +++ b/tests/validation/CL/ConvolutionLayer.cpp @@ -45,7 +45,7 @@ using namespace arm_compute::test::validation; namespace { -const float tolerance_f16 = 1.1f; /**< Tolerance value for comparing reference's output against implementation's output for DataType::F16 */ +const float tolerance_f16 = 0.1f; /**< Tolerance value for comparing reference's output against implementation's output for DataType::F16 */ const float tolerance_f32 = 1e-03f; /**< Tolerance value for comparing reference's output against implementation's output for DataType::F32 */ const float tolerance_q = 1.0f; /**< Tolerance value for comparing reference's output against implementation's output for fixed point data types */ |