From 47a899017e67556ffffef78571c9be61dd7bc3f0 Mon Sep 17 00:00:00 2001 From: Michele Di Giorgio Date: Mon, 9 Mar 2020 19:32:33 +0000 Subject: COMPMID-3237: Implement NEQLSTMLayer COMPMID-3082: Extend NEQLSTMLayer with enhancements Change-Id: I88175b7bf69494a4eae510b74176fe8a0d6cd770 Signed-off-by: Michele Di Giorgio Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2969 Tested-by: Arm Jenkins Reviewed-by: Sang-Hoon Park Reviewed-by: Sheri Zhang Comments-Addressed: Arm Jenkins --- arm_compute/runtime/NEON/functions/NEQLSTMLayer.h | 332 ++++++++++++++++++++++ 1 file changed, 332 insertions(+) create mode 100644 arm_compute/runtime/NEON/functions/NEQLSTMLayer.h (limited to 'arm_compute/runtime/NEON/functions/NEQLSTMLayer.h') diff --git a/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h b/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h new file mode 100644 index 0000000000..a37909b775 --- /dev/null +++ b/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h @@ -0,0 +1,332 @@ +/* + * Copyright (c) 2020 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef ARM_COMPUTE_NEQLSTMLAYER_H +#define ARM_COMPUTE_NEQLSTMLAYER_H + +#include "arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h" +#include "arm_compute/core/NEON/kernels/NEArithmeticSubtractionKernel.h" +#include "arm_compute/core/NEON/kernels/NEGEMMLowpReductionKernel.h" +#include "arm_compute/core/NEON/kernels/NEPixelWiseMultiplicationKernel.h" +#include "arm_compute/core/Types.h" +#include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" +#include "arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h" +#include "arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h" +#include "arm_compute/runtime/NEON/functions/NETranspose.h" + +#include "arm_compute/runtime/common/LSTMParams.h" + +namespace arm_compute +{ +// Forward declarations +class ITensor; + +/** Basic function to run @ref NEQLSTMLayer + * + * This function calls the following NEON functions/kernels: + * + * -# @ref NEActivationLayer Activation functions (tanh and logistic) + * -# @ref NEArithmeticAdditionKernel Elementwise addition + * -# @ref NEArithmeticSubtractionKernel Elementwise subtraction + * -# @ref NEGEMMLowpMatrixMultiplyCore Quantized matrix multiplication core. Accumulators are 32-bit integers + * -# @ref NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint Convert 32-bit integers into QSYMM16 + * -# @ref NEGEMMLowpMatrixAReductionKernel For precomputing effective biases to use + * -# @ref NEPixelWiseMultiplicationKernel Elementwise multiplication + * -# @ref NETranspose Transpose function for reshaping the weights + * */ +class NEQLSTMLayer : public IFunction +{ +public: + /** Default constructor */ + NEQLSTMLayer(std::shared_ptr memory_manager = nullptr); + /** Prevent instances of this class from being copied (As this class contains pointers) */ + NEQLSTMLayer(const NEQLSTMLayer &) = delete; + /** Default move constructor */ + NEQLSTMLayer(NEQLSTMLayer &&) = default; + /** Prevent instances of this class from being copied (As this class contains pointers) */ + NEQLSTMLayer &operator=(const NEQLSTMLayer &) = delete; + /** Default move assignment operator */ + NEQLSTMLayer &operator=(NEQLSTMLayer &&) = default; + /** Initialize function's tensors. + * + * @param[in] input Source tensor. Input is a 2D tensor with dimensions [input_size, batch_size]. Data types supported: QASYMM8_SIGNED. + * @param[in] input_to_forget_weights 2D weights tensor with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] input_to_cell_weights 2D weights tensor with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] input_to_output_weights 2D weights tensor with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_forget_weights 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_cell_weights 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_output_weights 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] forget_gate_bias 1D weights tensor with dimensions [num_units]. Data type supported: S32. + * @param[in] cell_bias 1D weights tensor with dimensions [num_units]. Data type supported: S32. + * @param[in] output_gate_bias 1D weights tensor with dimensions [num_units]. Data type supported: S32. + * @param[in] cell_state_in 2D tensor with dimensions [output_size, batch_size]. Data type supported: QSYMM16. + * @param[in] output_state_in 2D tensor with dimensions [num_units, batch_size]. Data type supported: Same as @p input. + * @param[out] cell_state_out Destination tensor. Output is a 2D tensor with dimensions [output_size, batch_size]. Data type supported: QSYMM16. + * @param[out] output_state_out Destination tensor. Output is a 2D tensor with dimensions [num_units, batch_size].Data types supported: Same as @p input. + * @param[in] lstm_params Weights tensors used in peephole, CIFG and layer normalization optimizations: + * input_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at input gate. + * forget_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at forget gate. + * cell_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at cell gate. + * output_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at output gate. + * hidden_state_zero The zero point of the hidden state. + * hidden_state_scale The scale of the hidden state. + * input_to_input_weights (Optional) 2D weights tensor with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * recurrent_to_input_weights (Optional) 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * cell_to_input_weights (Optional) 1D weights tensor with dimensions [num_units]. Can be nullptr. Data type supported: QSYMM16. + * cell_to_forget_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_to_output_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * input_gate_bias (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: S32. + * projection_weights (Optional) 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * projection_bias (Optional) 1D weights tensor with dimensions [output_size]. S32. + * input_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * forget_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * output_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_threshold (Optional) The clipping threshold for the cell state, such that values are bound within [-cell_clip, cell_clip]. + * If set to 0.0 then clipping is disabled. + * projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within + * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. + */ + void configure(const ITensor *input, + const ITensor *input_to_forget_weights, const ITensor *input_to_cell_weights, const ITensor *input_to_output_weights, + const ITensor *recurrent_to_forget_weights, const ITensor *recurrent_to_cell_weights, const ITensor *recurrent_to_output_weights, + const ITensor *forget_gate_bias, const ITensor *cell_bias, const ITensor *output_gate_bias, + const ITensor *cell_state_in, const ITensor *output_state_in, + ITensor *cell_state_out, ITensor *output_state_out, + const LSTMParams &lstm_params); + + /** Static function to check if given info will lead to a valid configuration of @ref NEQLSTMLayer + * + * @param[in] input Source tensor info. Input is a 2D tensor info with dimensions [input_size, batch_size]. Data types supported: QASYMM8_SIGNED. + * @param[in] input_to_forget_weights 2D weights tensor info with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] input_to_cell_weights 2D weights tensor info with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] input_to_output_weights 2D weights tensor info with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_forget_weights 2D weights tensor info with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_cell_weights 2D weights tensor info with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] recurrent_to_output_weights 2D weights tensor info with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * @param[in] forget_gate_bias 1D weights tensor info with dimensions [num_units]. Data type supported: S32. + * @param[in] cell_bias 1D weights tensor info with dimensions [num_units]. Data type supported: S32. + * @param[in] output_gate_bias 1D weights tensor info with dimensions [num_units]. Data type supported: S32. + * @param[in] cell_state_in 2D tensor info with dimensions [num_units, batch_size]. Data type supported: QSYMM16. + * @param[in] output_state_in 2D tensor info with dimensions [output_size, batch_size]. Data type supported: Same as @p input. + * @param[out] cell_state_out Destination tensor info. Output is a 2D tensor info with dimensions [num_units, batch_size]. Data type supported: QSYMM16. + * @param[out] output_state_out Destination tensor info. Output is a 2D tensor info with dimensions [output_size, batch_size].Data types supported: Same as @p input. + * @param[in] lstm_params Weights tensors info used in peephole, CIFG and layer normalization optimizations: + * input_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at input gate. + * forget_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at forget gate. + * cell_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at cell gate. + * output_intermediate_scale Scale of the intermediate result of matmul, i.e. input to layer normalization, at output gate. + * hidden_state_zero The zero point of the hidden state. + * hidden_state_scale The scale of the hidden state. + * input_to_input_weights (Optional) 2D weights tensor with dimensions [input_size, num_units]. Data type supported: QSYMM8. + * recurrent_to_input_weights (Optional) 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * cell_to_input_weights (Optional) 1D weights tensor with dimensions [num_units]. Can be nullptr. Data type supported: QSYMM16. + * cell_to_forget_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_to_output_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * input_gate_bias (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: S32. + * projection_weights (Optional) 2D weights tensor with dimensions [output_size, num_units]. Data type supported: QSYMM8. + * projection_bias (Optional) 1D weights tensor with dimensions [output_size]. S32. + * input_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * forget_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * output_layer_norm_weights (Optional) 1D weights tensor with dimensions [num_units]. Data type supported: QSYMM16. + * cell_threshold (Optional) The clipping threshold for the cell state, such that values are bound within [-cell_clip, cell_clip]. + * If set to 0.0 then clipping is disabled. + * projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within + * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. + * @return a status + */ + static Status validate(const ITensorInfo *input, + const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out, + const LSTMParams &lstm_params); + + // Inherited methods overridden: + void run() override; + void prepare() override; + +private: + /** Internal method to configure matrix multiplication plus output stage of each gate. + * + * @param[in] mm Matrix multiplication function to use. + * @param[in] outstage Output stage function to use. + * @param[in] gemmlowp_info GEMMLowp metadata to be used by the output stage. + * @param[in] mm_input Input tensor to matrix multiplication function. + * @param[in] mm_weights Weights tensor to matrix multiplication function. + * @param[in] bias Bias tensor to matrix multiplication function. + * @param[in] outstage_res Tensor to be used for storing the result of the output stage. + * @param[in] gemmlowp_scale Real multiplier to be used computing multiplier and shift for requantization. + * @param[in] mm_res_info Tensor info to be used to initialize matrix multiplication result tensor. + * @param[in] mm_res_info Tensor info to be used to initialize output stage result tensor. + * + */ + void configure_mm(NEGEMMLowpMatrixMultiplyCore &mm, NEGEMMLowpOutputStage &outstage, GEMMLowpOutputStageInfo &gemmlowp_info, + const ITensor *mm_input, const ITensor *mm_weights, const ITensor *bias, Tensor *mm_res, + Tensor *outstage_res, float gemmlowp_scale, + const TensorInfo &mm_res_info, const TensorInfo &outstage_tensor_info); + + MemoryGroup _memory_group{}; + + // Functions used + NETranspose _transpose_input_to_forget_weights{}; + NETranspose _transpose_input_to_cell_weights{}; + NETranspose _transpose_input_to_output_weights{}; + NETranspose _transpose_input_to_input_weights{}; + NETranspose _transpose_recurrent_to_forget_weights{}; + NETranspose _transpose_recurrent_to_cell_weights{}; + NETranspose _transpose_recurrent_to_output_weights{}; + NETranspose _transpose_recurrent_to_input_weights{}; + NETranspose _transpose_projection_weights{}; + NEGEMMLowpMatrixAReductionKernel _input_to_input_reduction{}; + NEGEMMLowpMatrixAReductionKernel _recurrent_to_input_reduction{}; + NEGEMMLowpMatrixAReductionKernel _input_to_forget_reduction{}; + NEGEMMLowpMatrixAReductionKernel _recurrent_to_forget_reduction{}; + NEGEMMLowpMatrixAReductionKernel _input_to_cell_reduction{}; + NEGEMMLowpMatrixAReductionKernel _recurrent_to_cell_reduction{}; + NEGEMMLowpMatrixAReductionKernel _input_to_output_reduction{}; + NEGEMMLowpMatrixAReductionKernel _recurrent_to_output_reduction{}; + NEGEMMLowpMatrixAReductionKernel _projection_reduction{}; + NEArithmeticAdditionKernel _projection_bias_add{}; + NEGEMMLowpMatrixMultiplyCore _mm_input_to_forget{}; + NEGEMMLowpMatrixMultiplyCore _mm_recurrent_to_forget{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_cell_to_forget{}; + NEGEMMLowpOutputStage _input_to_forget_outstage{}; + NEGEMMLowpOutputStage _recurrent_to_forget_outstage{}; + NEGEMMLowpOutputStage _cell_to_forget_outstage{}; + NEArithmeticAdditionKernel _accumulate_input_recurrent_forget{}; + NEArithmeticAdditionKernel _accumulate_cell_forget{}; + NEActivationLayer _forget_gate_sigmoid{}; + NEGEMMLowpMatrixMultiplyCore _mm_input_to_cell{}; + NEGEMMLowpOutputStage _input_to_cell_outstage{}; + NEGEMMLowpMatrixMultiplyCore _mm_recurrent_to_cell{}; + NEGEMMLowpOutputStage _recurrent_to_cell_outstage{}; + NEArithmeticAdditionKernel _accumulate_input_recurrent_modulation{}; + NEActivationLayer _cell_gate_tanh{}; + NEArithmeticSubtractionKernel _input_gate_sub{}; + NEGEMMLowpMatrixMultiplyCore _mm_input_to_input{}; + NEGEMMLowpOutputStage _input_to_input_outstage{}; + NEGEMMLowpMatrixMultiplyCore _mm_recurrent_to_input{}; + NEGEMMLowpOutputStage _recurrent_to_input_outstage{}; + NEArithmeticAdditionKernel _accumulate_input_recurrent_input{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_cell_to_input{}; + NEGEMMLowpOutputStage _cell_to_input_outstage{}; + NEArithmeticAdditionKernel _accumulate_cell_input{}; + NEActivationLayer _input_gate_tanh{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_forget_cell{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_input_cell{}; + NEArithmeticAdditionKernel _add_forget_cell{}; + NEActivationLayer _cell_clip{}; + NEGEMMLowpMatrixMultiplyCore _mm_input_to_output{}; + NEGEMMLowpOutputStage _input_to_output_outstage{}; + NEGEMMLowpMatrixMultiplyCore _mm_recurrent_to_output{}; + NEGEMMLowpOutputStage _recurrent_to_output_outstage{}; + NEArithmeticAdditionKernel _accumulate_input_recurrent_output{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_cell_to_output{}; + NEArithmeticAdditionKernel _accumulate_cell_to_output{}; + NEActivationLayer _output_gate_sigmoid{}; + NEActivationLayer _hidden_tanh{}; + NEPixelWiseMultiplicationKernel _pixelwise_mul_hidden{}; + NEGEMMLowpOutputStage _hidden_outstage{}; + NEGEMMLowpMatrixMultiplyCore _mm_projection{}; + NEGEMMLowpOutputStage _projection_outstage{}; + NEArithmeticAdditionKernel _accumulate_projection{}; + NEActivationLayer _projection_clip{}; + + // Tensor pointers + const ITensor *_input_to_input_weights + { + nullptr + }; + const ITensor *_recurrent_to_input_weights{ nullptr }; + const ITensor *_projection_bias{ nullptr }; + const ITensor *_input_to_forget_weights{ nullptr }; + const ITensor *_input_to_cell_weights{ nullptr }; + const ITensor *_input_to_output_weights{ nullptr }; + const ITensor *_recurrent_to_forget_weights{ nullptr }; + const ITensor *_recurrent_to_cell_weights{ nullptr }; + const ITensor *_recurrent_to_output_weights{ nullptr }; + const ITensor *_projection_weights{ nullptr }; + + // Temporary tensors + Tensor _input_to_forget_weights_transposed{ nullptr }; + Tensor _input_to_cell_weights_transposed{ nullptr }; + Tensor _input_to_output_weights_transposed{ nullptr }; + Tensor _input_to_input_weights_transposed{ nullptr }; + Tensor _recurrent_to_forget_weights_transposed{ nullptr }; + Tensor _recurrent_to_cell_weights_transposed{ nullptr }; + Tensor _recurrent_to_output_weights_transposed{ nullptr }; + Tensor _recurrent_to_input_weights_transposed{ nullptr }; + Tensor _projection_weights_transposed{ nullptr }; + Tensor _input_to_input_eff_bias{ nullptr }; + Tensor _recurrent_to_input_eff_bias{ nullptr }; + Tensor _input_to_forget_eff_bias{ nullptr }; + Tensor _recurrent_to_forget_eff_bias{ nullptr }; + Tensor _input_to_cell_eff_bias{ nullptr }; + Tensor _recurrent_to_cell_eff_bias{ nullptr }; + Tensor _input_to_output_eff_bias{ nullptr }; + Tensor _recurrent_to_output_eff_bias{ nullptr }; + Tensor _projection_reduction_res{ nullptr }; + Tensor _projection_eff_bias{ nullptr }; + Tensor _mm_input_to_forget_res{ nullptr }; + Tensor _mm_recurrent_to_forget_res{ nullptr }; + Tensor _mul_cell_to_forget_res{ nullptr }; + Tensor _input_to_forget_outstage_res{ nullptr }; + Tensor _cell_to_forget_outstage_res{ nullptr }; + Tensor _recurrent_to_forget_outstage_res{ nullptr }; + Tensor _forget_gate{ nullptr }; + Tensor _mm_input_to_cell_res{ nullptr }; + Tensor _input_to_cell_outstage_res{ nullptr }; + Tensor _mm_recurrent_to_cell_res{ nullptr }; + Tensor _recurrent_to_cell_outstage_res{ nullptr }; + Tensor _cell_gate{ nullptr }; + Tensor _mul_input_cell_res{ nullptr }; + Tensor _mm_input_to_input_res{ nullptr }; + Tensor _input_to_input_outstage_res{ nullptr }; + Tensor _mm_recurrent_to_input_res{ nullptr }; + Tensor _mul_cell_to_input_res{ nullptr }; + Tensor _cell_to_input_outstage_res{ nullptr }; + Tensor _recurrent_to_input_outstage_res{ nullptr }; + Tensor _input_gate{ nullptr }; + Tensor _mm_input_to_output_res{ nullptr }; + Tensor _input_to_output_outstage_res{ nullptr }; + Tensor _mm_recurrent_to_output_res{ nullptr }; + Tensor _mul_cell_to_output_res{ nullptr }; + Tensor _recurrent_to_output_outstage_res{ nullptr }; + Tensor _output_gate{ nullptr }; + Tensor _hidden_mul_res{ nullptr }; + Tensor _mm_projection_res{ nullptr }; + Tensor _projection_outstage_res{ nullptr }; + Tensor _ones{ nullptr }; + + bool _is_prepared{ false }; + bool _has_cifg{ false }; + bool _has_cell_clipping{ false }; + bool _has_projection{ false }; + bool _has_projection_clipping{ false }; + bool _has_peephole{ false }; +}; +} // namespace arm_compute +#endif /* ARM_COMPUTE_NEQLSTMLAYER_H */ -- cgit v1.2.1