diff options
author | Gunes Bayir <gunes.bayir@arm.com> | 2023-01-19 15:56:00 +0000 |
---|---|---|
committer | Gunes Bayir <gunes.bayir@arm.com> | 2023-01-24 09:40:01 +0000 |
commit | cc2877368d5e15d9ea89d31c84ec651fc0fffd13 (patch) | |
tree | c57a3a406125b3a31e2d4aff6126ce99f4ade395 /tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h | |
parent | a6a153817302793732e28b07c3b4046df3f91a60 (diff) | |
download | ComputeLibrary-cc2877368d5e15d9ea89d31c84ec651fc0fffd13.tar.gz |
Change dynamic fusion API to return destination tensor info
The new dynamic fusion API is introduced in the following patch:
https://review.mlplatform.org/c/ml/ComputeLibrary/+/8906
For each operator (except Conv2D, which is migrated in the above patch), we
- remove destination tensor from is_supported, validate and create calls
- make create_op return ITensorInfo* to the intermediate destination object
Affected operators:
- DepthwiseConv2D
- Cast
- Elementwise Ops
- Clamp
- Reshape
- Resize
Resolves: COMPMID-5777
Change-Id: Ib60ec8a5f081752808455d7a7d790f2ed0627059
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8991
Reviewed-by: Ramy Elgammal <ramy.elgammal@arm.com>
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Dynamic-Fusion: Ramy Elgammal <ramy.elgammal@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/fixtures/dynamic_fusion/operators/ClampFixture.h')
-rw-r--r-- | tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h | 20 |
1 files changed, 7 insertions, 13 deletions
diff --git a/tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h b/tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h index fe87d9a022..a1fd22582f 100644 --- a/tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h +++ b/tests/validation/fixtures/dynamic_fusion/operators/ClampFixture.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -94,8 +94,8 @@ protected: template <typename U> void fill(U &&tensor) { - float min_bound = 0; - float max_bound = 0; + float min_bound = 0; + float max_bound = 0; std::tie(min_bound, max_bound) = get_activation_layer_test_bounds<T>(ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU, _data_type); library->fill_static_values(tensor, get_boundary_values(static_cast<T>(min_bound), static_cast<T>(max_bound))); } @@ -111,21 +111,15 @@ protected: TensorInfo src_info = sketch.create_tensor_info(TensorInfo(shape, 1, _data_type)); TensorInfo dst_info = sketch.create_tensor_info(TensorInfo(shape, 1, _data_type)); - auto ans_0_info = sketch.create_tensor_info(); - TensorInfo ans_1_info; - - FunctionType::create_op(sketch, &src_info, &ans_0_info, attributes); - + ITensorInfo *ans_0_info = FunctionType::create_op(sketch, &src_info, attributes); if(_fuse) { - ans_1_info = sketch.create_tensor_info(); - - FunctionType::create_op(sketch, &ans_0_info, &ans_1_info, attributes); - GpuOutput::create_op(sketch, &ans_1_info, &dst_info); + ITensorInfo *ans_1_info = FunctionType::create_op(sketch, ans_0_info, attributes); + GpuOutput::create_op(sketch, ans_1_info, &dst_info); } else { - GpuOutput::create_op(sketch, &ans_0_info, &dst_info); + GpuOutput::create_op(sketch, ans_0_info, &dst_info); } // Configure runtime |