diff options
| author | Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> | 2023-01-03 10:16:16 +0000 |
|---|---|---|
| committer | Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> | 2023-01-20 13:11:59 +0000 |
| commit | a18d85c6d2c0025938c2dc10e553eb82c01922f2 (patch) | |
| tree | e62ce3acdbe065bedba7355cdaba6cf32d7ce20e /tests/validation/reference | |
| parent | 11f7d7ed4aa6293c8ad115374b7bad9bbf5a8ae7 (diff) | |
| download | ComputeLibrary-a18d85c6d2c0025938c2dc10e553eb82c01922f2.tar.gz | |
Dynamic Fusion Pooling Layer 2d
- Adds Dynamic fusion PoolingLayer2D as Unfusable Operator
- Indices are not supported
- Adds tests for F32/F16 Datatypes
Resolves : [COMPMID-5520]
Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com>
Change-Id: I0d112545eb9209c836bf9ea153069f8627531e0a
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8893
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'tests/validation/reference')
| -rw-r--r-- | tests/validation/reference/PoolingLayer.cpp | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/tests/validation/reference/PoolingLayer.cpp b/tests/validation/reference/PoolingLayer.cpp index 9e671e3173..378d91d829 100644 --- a/tests/validation/reference/PoolingLayer.cpp +++ b/tests/validation/reference/PoolingLayer.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2021 Arm Limited. + * Copyright (c) 2017-2021, 2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -40,7 +40,6 @@ using namespace arm_compute::misc::shape_calculator; template <typename T, typename ACC_T, typename std::enable_if<is_floating_point<T>::value, int>::type> SimpleTensor<T> pooling_layer_internal(const SimpleTensor<T> &src, const PoolingLayerInfo &info, SimpleTensor<uint32_t> *indices, DataLayout data_layout) { - ARM_COMPUTE_ERROR_ON(info.is_global_pooling && (src.shape().x() != src.shape().y())); // Create reference SimpleTensor<T> dst{ compute_pool_shape(TensorInfo(src.shape(), 1, src.data_type()), info), src.data_type(), 1 }; auto pooled_shape = compute_pool_shape(TensorInfo(src.shape(), 1, src.data_type()), info); |