diff options
| author | Gunes Bayir <gunes.bayir@arm.com> | 2024-03-19 17:37:10 +0000 |
|---|---|---|
| committer | Gunes Bayir <gunes.bayir@arm.com> | 2024-03-20 10:10:38 +0000 |
| commit | 1618e956130ddb0cb69acd56595d1a959e1db513 (patch) | |
| tree | 08873bc28388dde453e652a01eb597420de1f49f | |
| parent | 43ba0dd3ea46d85267914c3b25cb8f9fdd719789 (diff) | |
| download | ComputeLibrary-1618e956130ddb0cb69acd56595d1a959e1db513.tar.gz | |
Increase tolerance_num of Cpu RNNLayer tests
Instead of increasing the tolerance amount, we increase the number of elements we tolerate to 0.02 % of the whole tensor. This ensures we do not affect the tolerance for smaller tests.
This amount is set according to the number of elements above the threshold. it was 1 over 512 elements, 1/512 ~ 0.02 %.
Resolves: COMPMID-6932
Change-Id: I9d3ce29a3972aa8b9daea5288005a0a41a266328
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11321
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
| -rw-r--r-- | tests/validation/NEON/RNNLayer.cpp | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/tests/validation/NEON/RNNLayer.cpp b/tests/validation/NEON/RNNLayer.cpp index 14d9a5d14e..979aa0f2c5 100644 --- a/tests/validation/NEON/RNNLayer.cpp +++ b/tests/validation/NEON/RNNLayer.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2018-2021 Arm Limited. + * Copyright (c) 2018-2021, 2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -139,7 +139,7 @@ TEST_SUITE(FP16) FIXTURE_DATA_TEST_CASE(RunSmall, NERNNLayerFixture<half>, framework::DatasetMode::ALL, combine(datasets::SmallRNNLayerDataset(), framework::dataset::make("DataType", DataType::F16))) { // Validate output - validate(Accessor(_target), _reference, tolerance_f16, 0.f, abs_tolerance_f16); + validate(Accessor(_target), _reference, tolerance_f16, 0.02f, abs_tolerance_f16); } TEST_SUITE_END() // FP16 #endif /* __ARM_FEATURE_FP16_VECTOR_ARITHMETIC */ |