diff options
| author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2018-04-11 15:59:10 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:49:37 +0000 |
| commit | e52a3000d2c13bc1b66ca66b3d12b6b836982394 (patch) | |
| tree | 70e8ef5ba216762604f84228805aac9bd65747b6 /tests/validation/CL/DilatedConvolutionLayer.cpp | |
| parent | dd03870b63784abe499761da2b26b209b33f2db2 (diff) | |
| download | ComputeLibrary-e52a3000d2c13bc1b66ca66b3d12b6b836982394.tar.gz | |
COMPMID-1026 - Add support for 4x4 output tile in CLWinogradConvolutionLayer
The performance achieved can be found at the following confluence page:
https://confluence.arm.com/display/MLENG/GEMM-based+convolution+vs+Winograd-based+convolution+on+OpenCL
Change-Id: I4b690cfdd4eb4ff0cd17b14fdd49ccaa1d1dc85c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/127729
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Diffstat (limited to 'tests/validation/CL/DilatedConvolutionLayer.cpp')
| -rw-r--r-- | tests/validation/CL/DilatedConvolutionLayer.cpp | 37 |
1 files changed, 15 insertions, 22 deletions
diff --git a/tests/validation/CL/DilatedConvolutionLayer.cpp b/tests/validation/CL/DilatedConvolutionLayer.cpp index 532d04b431..e6a765bbe1 100644 --- a/tests/validation/CL/DilatedConvolutionLayer.cpp +++ b/tests/validation/CL/DilatedConvolutionLayer.cpp @@ -63,25 +63,19 @@ const auto CNNDataTypes = framework::dataset::make("DataType", TEST_SUITE(CL) TEST_SUITE(DilatedConvolutionLayer) -DATA_TEST_CASE(ValidateConvolutionMethod, framework::DatasetMode::ALL, zip(zip(zip(zip(zip(zip(zip( - framework::dataset::make("InputInfo", { TensorInfo(TensorShape(17U, 31U, 2U), 1, DataType::F32, 0), - TensorInfo(TensorShape(17U, 31U, 2U), 1, DataType::F32, 0), - TensorInfo(TensorShape(23U, 27U, 5U, 4U), 1, DataType::F32, 0), - TensorInfo(TensorShape(3U, 3U, 2U, 1U), 1, DataType::F32, 0), - TensorInfo(TensorShape(33U, 27U, 7U, 4U), 1, DataType::F32, 0) - }), - framework::dataset::make("WeightsInfo", { TensorInfo(TensorShape(5U, 5U, 2U, 19U), 1, DataType::F32, 0), - TensorInfo(TensorShape(5U, 5U, 2U, 19U), 1, DataType::F32, 0), - TensorInfo(TensorShape(3U, 3U, 5U, 21U), 1, DataType::F32, 0), - TensorInfo(TensorShape(3U, 3U, 5U, 21U), 1, DataType::F32, 0), - TensorInfo(TensorShape(5U, 5U, 7U, 16U), 1, DataType::F16, 0) - })), - framework::dataset::make("BiasesInfo", { TensorInfo(TensorShape(19U), 1, DataType::F32, 0), - TensorInfo(TensorShape(19U), 1, DataType::F32, 0), - TensorInfo(TensorShape(21U), 1, DataType::F32, 0), - TensorInfo(TensorShape(21U), 1, DataType::F32, 0), - TensorInfo(TensorShape(16U), 1, DataType::F32, 0) - })), +DATA_TEST_CASE(ValidateConvolutionMethod, framework::DatasetMode::ALL, zip(zip(zip(zip(zip(zip( + framework::dataset::make("InputInfo", { TensorInfo(TensorShape(17U, 31U, 2U), 1, DataType::F32, 0), + TensorInfo(TensorShape(17U, 31U, 2U), 1, DataType::F32, 0), + TensorInfo(TensorShape(23U, 27U, 5U, 4U), 1, DataType::F32, 0), + TensorInfo(TensorShape(3U, 3U, 2U, 1U), 1, DataType::F32, 0), + TensorInfo(TensorShape(33U, 27U, 7U, 4U), 1, DataType::F32, 0) + }), + framework::dataset::make("WeightsInfo", { TensorInfo(TensorShape(5U, 5U, 2U, 19U), 1, DataType::F32, 0), + TensorInfo(TensorShape(5U, 5U, 2U, 19U), 1, DataType::F32, 0), + TensorInfo(TensorShape(3U, 3U, 5U, 21U), 1, DataType::F32, 0), + TensorInfo(TensorShape(3U, 3U, 5U, 21U), 1, DataType::F32, 0), + TensorInfo(TensorShape(5U, 5U, 7U, 16U), 1, DataType::F16, 0) + })), framework::dataset::make("OutputInfo", { TensorInfo(TensorShape(15U, 15U, 19U), 1, DataType::F32, 0), TensorInfo(TensorShape(15U, 15U, 19U), 1, DataType::F32, 0), TensorInfo(TensorShape(21U, 25U, 21U, 4U), 1, DataType::F32, 0), @@ -107,12 +101,11 @@ DATA_TEST_CASE(ValidateConvolutionMethod, framework::DatasetMode::ALL, zip(zip(z Size2D(3U, 3U) })), - framework::dataset::make("Expected", { ConvolutionMethod::GEMM, ConvolutionMethod::GEMM, ConvolutionMethod::GEMM, ConvolutionMethod::GEMM, ConvolutionMethod::GEMM })), - input_info, weights_info, biases_info, output_info, conv_info, gpu_target, dilation, expected) + framework::dataset::make("Expected", { ConvolutionMethod::GEMM, ConvolutionMethod::GEMM, ConvolutionMethod::WINOGRAD, ConvolutionMethod::GEMM, ConvolutionMethod::GEMM })), + input_info, weights_info, output_info, conv_info, gpu_target, dilation, expected) { ConvolutionMethod is_valid = CLConvolutionLayer::get_convolution_method(&input_info.clone()->set_is_resizable(false), &weights_info.clone()->set_is_resizable(false), - &biases_info.clone()->set_is_resizable(false), &output_info.clone()->set_is_resizable(false), conv_info, WeightsInfo(), ActivationLayerInfo(), gpu_target, dilation); ARM_COMPUTE_EXPECT(is_valid == expected, framework::LogLevel::ERRORS); } |