diff options
| author | Giorgio Arena <giorgio.arena@arm.com> | 2018-04-11 19:07:17 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:49:37 +0000 |
| commit | 9373c8b2650f34b2804d3685588bad8e408ebe63 (patch) | |
| tree | 4cc2330827578fcc87e6445071a105d48e1dd4c6 /src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp | |
| parent | fc1d1e2200f3056572b158b8208bac456f48339f (diff) | |
| download | ComputeLibrary-9373c8b2650f34b2804d3685588bad8e408ebe63.tar.gz | |
COMPMID-1037 Add support for F(4x4, 5x5) in CLWinogradFilterTransformKernel
Change-Id: I6dd639bf5df9bc0c133996f75bdee767f70a6cfb
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/127469
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp')
| -rw-r--r-- | src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp b/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp index 5b8921b8e4..d3a33c01f9 100644 --- a/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp +++ b/src/core/CL/kernels/CLWinogradFilterTransformKernel.cpp @@ -55,10 +55,11 @@ Status validate_arguments(const ITensorInfo *input, const ITensorInfo *output, c const size_t idx_w = get_data_layout_dimension_index(input->data_layout(), DataLayoutDimension::WIDTH); const size_t idx_h = get_data_layout_dimension_index(input->data_layout(), DataLayoutDimension::HEIGHT); - ARM_COMPUTE_RETURN_ERROR_ON(kernel_size != Size2D(3U, 3U)); + ARM_COMPUTE_RETURN_ERROR_ON(kernel_size != Size2D(3U, 3U) && kernel_size != Size2D(5U, 5U)); + ARM_COMPUTE_RETURN_ERROR_ON(kernel_size == Size2D(3U, 3U) && output_tile_size != Size2D(2U, 2U) && output_tile_size != Size2D(4U, 4U)); + ARM_COMPUTE_RETURN_ERROR_ON(kernel_size == Size2D(5U, 5U) && output_tile_size != Size2D(4U, 4U)); ARM_COMPUTE_RETURN_ERROR_ON(input->dimension(idx_w) != kernel_size.width || input->dimension(idx_h) != kernel_size.height); ARM_COMPUTE_RETURN_ERROR_ON(input->num_dimensions() > 4); - ARM_COMPUTE_RETURN_ERROR_ON(output_tile_size != Size2D(2U, 2U) && output_tile_size != Size2D(4U, 4U)); // Checks performed when output is configured if(output->total_size() != 0) @@ -76,8 +77,8 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input, ITen { ARM_COMPUTE_ERROR_ON_NULLPTR(input, output); - constexpr unsigned int num_elems_processed_per_iteration_x = 3; - constexpr unsigned int num_elems_processed_per_iteration_y = 3; + const unsigned int num_elems_processed_per_iteration_x = input->dimension(get_data_layout_dimension_index(input->data_layout(), DataLayoutDimension::WIDTH)); + const unsigned int num_elems_processed_per_iteration_y = input->dimension(get_data_layout_dimension_index(input->data_layout(), DataLayoutDimension::HEIGHT)); Window win = calculate_max_window(*input, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); bool window_changed = false; |