diff options
Diffstat (limited to 'arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h')
| -rw-r--r-- | arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h | 37 |
1 files changed, 31 insertions, 6 deletions
diff --git a/arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h b/arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h index 9169b75d19..2f44d19b4f 100644 --- a/arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h +++ b/arm_compute/core/NEON/kernels/NEWinogradLayerKernel.h @@ -34,6 +34,7 @@ namespace arm_compute { class ITensor; +/** Interface for the NEON kernel to perform Winograd input transform. */ template <typename T> class INEWinogradLayerTransformInputKernel : public INEKernel { @@ -46,6 +47,8 @@ public: * @param[in] n_rows Number of rows in each feature map. * @param[in] n_cols Number of columns in each feature map. * @param[in] same_padding Use "SAME" padding, otherwise use "VALID". + * + * @return Storage size (in units of TIn) required. */ virtual unsigned int get_input_storage_size(int n_batches, int n_channels, int n_rows, int n_cols, bool same_padding) const = 0; @@ -72,11 +75,13 @@ public: */ virtual void configure(const T *const input, const int n_batches, const int n_rows, const int n_cols, const int n_channels, const PaddingType padding, T *const output, const int matrix_stride) = 0; + /** Destructor */ virtual ~INEWinogradLayerTransformInputKernel() { } }; +/** NEON kernel to perform Winograd input transform. */ template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols> class NEWinogradLayerTransformInputKernel : public INEWinogradLayerTransformInputKernel<T> { @@ -89,6 +94,8 @@ public: * @param[in] n_rows Number of rows in each feature map. * @param[in] n_cols Number of columns in each feature map. * @param[in] same_padding Use "SAME" padding, otherwise use "VALID". + * + * @return Storage size (in units of TIn) required. */ unsigned int get_input_storage_size( int n_batches, @@ -107,6 +114,7 @@ public: */ int get_matrix_stride(const KernelShape &kernel_shape, const Tensor4DShape &input_shape, const PaddingType padding_type) const override; + /** Default constructor */ NEWinogradLayerTransformInputKernel(); const char *name() const override @@ -139,7 +147,9 @@ public: void run(const Window &window, const ThreadInfo &info) override; bool is_parallelisable() const override; + /** Winograd base kernel */ using WinogradBase = winograd::WinogradGEMM<OutputTileRows, OutputTileCols, KernelCols, KernelCols>; + /** Winograd convolution kernel */ using WinogradConv = typename WinogradBase::template Convolution<T, T>; private: @@ -147,6 +157,7 @@ private: std::unique_ptr<InputTransform> _transform; }; +/** Interface for the NEON kernel to perform Winograd output transform. */ template <typename T> class INEWinogradLayerTransformOutputKernel : public INEKernel { @@ -159,6 +170,8 @@ public: * @param[in] n_cols Number of columns in each feature map of the input tensor. * @param[in] n_output_channels Number of feature maps in the output tensor. * @param[in] same_padding Use "SAME" padding, otherwise use "VALID". + * + * @return Storage size (in units of TOut) required. */ virtual unsigned int get_output_storage_size(int n_batches, int n_rows, int n_cols, int n_output_channels, bool same_padding) const = 0; @@ -208,6 +221,7 @@ public: } }; +/** NEON kernel to perform Winograd output transform. */ template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols> class NEWinogradLayerTransformOutputKernel : public INEWinogradLayerTransformOutputKernel<T> { @@ -227,7 +241,7 @@ public: NEWinogradLayerTransformOutputKernel(NEWinogradLayerTransformOutputKernel &&) = default; /** Allow instances of this class to be moved */ NEWinogradLayerTransformOutputKernel &operator=(NEWinogradLayerTransformOutputKernel &&) = default; - + /** Default destructor */ ~NEWinogradLayerTransformOutputKernel() = default; // Inherited methods overridden: @@ -239,6 +253,8 @@ public: * @param[in] n_cols Number of columns in each feature map of the input tensor. * @param[in] n_output_channels Number of feature maps in the output tensor. * @param[in] same_padding Use "SAME" padding, otherwise use "VALID". + * + * @return Storage size (in units of TOut) required. */ unsigned int get_output_storage_size(int n_batches, int n_rows, int n_cols, int n_output_channels, bool same_padding) const override; @@ -301,6 +317,7 @@ private: int _n_channels; }; +/** Interface for the NEON kernel to perform Winograd weights transform. */ template <typename T> class INEWinogradLayerTransformWeightsKernel : public INEKernel { @@ -310,6 +327,8 @@ public: * * @param[in] n_output_channels Number of output feature maps. * @param[in] n_input_channels Number of input feature maps. + * + * @return Storage size (in units of T) required. */ virtual unsigned int get_weight_storage_size(int n_output_channels, int n_input_channels) const = 0; /** Gets the stride between matrices in the kernel worspace @@ -335,10 +354,12 @@ public: } }; +/** NEON kernel to perform Winograd weights transform. */ template <typename T, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols> class NEWinogradLayerTransformWeightsKernel final : public INEWinogradLayerTransformWeightsKernel<T> { public: + /** Default constructor. */ NEWinogradLayerTransformWeightsKernel(); const char *name() const override { @@ -359,6 +380,7 @@ private: std::unique_ptr<WeightsTransform> _transform; }; +/** Interface for the NEON kernel to perform Winograd. */ template <typename TIn, typename TOut> class INEWinogradLayerBatchedGEMMKernel : public INEKernel { @@ -406,16 +428,17 @@ public: virtual int get_number_blocks() const = 0; }; +/** NEON kernel to perform Winograd. */ template <typename TIn, typename TOut, int OutputTileRows, int OutputTileCols, int KernelRows, int KernelCols> class NEWinogradLayerBatchedGEMMKernel : public INEWinogradLayerBatchedGEMMKernel<TIn, TOut> { public: + /** Winograd base kernel */ using WinogradBase = winograd::WinogradGEMM<OutputTileRows, OutputTileCols, KernelRows, KernelCols>; + /** Winograd convolution kernel */ using WinogradConv = typename WinogradBase::template Convolution<TIn, TOut>; - using MultiGEMM = winograd::BatchedBlockedGemm<WinogradConv::M_BLOCK, WinogradConv::N_BLOCK, TIn, TOut>; - - static const int _output_tile_rows = OutputTileRows; - static const int _output_tile_cols = OutputTileCols; + /** Winograd batched blocked GEMM operator */ + using MultiGEMM = winograd::BatchedBlockedGemm<WinogradConv::M_BLOCK, WinogradConv::N_BLOCK, TIn, TOut>; const char *name() const override { @@ -432,7 +455,7 @@ public: NEWinogradLayerBatchedGEMMKernel(NEWinogradLayerBatchedGEMMKernel &&) = default; /** Allow instances of this class to be moved */ NEWinogradLayerBatchedGEMMKernel &operator=(NEWinogradLayerBatchedGEMMKernel &&) = default; - + /** Default destructor. */ ~NEWinogradLayerBatchedGEMMKernel() = default; // Inherited methods overridden: @@ -474,6 +497,8 @@ public: void run(const Window &window, const ThreadInfo &info) override; private: + static const int _output_tile_rows = OutputTileRows; + static const int _output_tile_cols = OutputTileCols; std::unique_ptr<MultiGEMM> _gemms; }; |