LCOV - code coverage report
Current view: top level - nntrainer/layers - mol_attention_layer.h (source / functions) Coverage Total Hit
Test: coverage_filtered.info Lines: 100.0 % 3 3
Test Date: 2025-12-14 20:38:17 Functions: 100.0 % 2 2

            Line data    Source code
       1              : // SPDX-License-Identifier: Apache-2.0
       2              : /**
       3              :  * Copyright (C) 2021 Parichay Kapoor <pk.kapoor@samsung.com>
       4              :  *
       5              :  * @file   mol_attention_layer.h
       6              :  * @date   11 November 2021
       7              :  * @see    https://github.com/nnstreamer/nntrainer
       8              :  * @author Parichay Kapoor <pk.kapoor@samsung.com>
       9              :  * @bug    No known bugs except for NYI items
      10              :  * @brief  This is MoL Attention Layer Class for Neural Network
      11              :  *
      12              :  */
      13              : 
      14              : #ifndef __MOL_ATTENTION_LAYER_H__
      15              : #define __MOL_ATTENTION_LAYER_H__
      16              : #ifdef __cplusplus
      17              : 
      18              : #include <attention_layer.h>
      19              : #include <layer_impl.h>
      20              : 
      21              : namespace nntrainer {
      22              : 
      23              : /**
      24              :  * @class   MoL Attention Layer
      25              :  * @brief   Mixture of Logistics Attention Layer
      26              :  */
      27              : class MoLAttentionLayer : public LayerImpl {
      28              : public:
      29              :   /**
      30              :    * @brief     Constructor of MoL Attention Layer
      31              :    */
      32              :   MoLAttentionLayer();
      33              : 
      34              :   /**
      35              :    * @brief     Destructor of MoL Attention Layer
      36              :    */
      37              :   ~MoLAttentionLayer();
      38              : 
      39              :   /**
      40              :    *  @brief  Move constructor of MoLAttentionLayer.
      41              :    *  @param[in] MoLAttentionLayer &&
      42              :    */
      43              :   MoLAttentionLayer(MoLAttentionLayer &&rhs) noexcept = default;
      44              : 
      45              :   /**
      46              :    * @brief  Move assignment operator.
      47              :    * @parma[in] rhs MoLAttentionLayer to be moved.
      48              :    */
      49              :   MoLAttentionLayer &operator=(MoLAttentionLayer &&rhs) = default;
      50              : 
      51              :   /**
      52              :    * @copydoc Layer::finalize(InitLayerContext &context)
      53              :    */
      54              :   void finalize(InitLayerContext &context) override;
      55              : 
      56              :   /**
      57              :    * @copydoc Layer::forwarding(RunLayerContext &context, bool training)
      58              :    */
      59              :   void forwarding(RunLayerContext &context, bool training) override;
      60              : 
      61              :   /**
      62              :    * @copydoc Layer::calcDerivative(RunLayerContext &context)
      63              :    */
      64              :   void calcDerivative(RunLayerContext &context) override;
      65              : 
      66              :   /**
      67              :    * @copydoc Layer::calcGradient(RunLayerContext &context)
      68              :    */
      69              :   void calcGradient(RunLayerContext &context) override;
      70              : 
      71              :   /**
      72              :    * @copydoc bool supportBackwarding() const
      73              :    */
      74            2 :   bool supportBackwarding() const override { return true; };
      75              : 
      76              :   /**
      77              :    * @copydoc Layer::exportTo(Exporter &exporter, ml::train::ExportMethods
      78              :    * method)
      79              :    */
      80              :   void exportTo(Exporter &exporter,
      81              :                 const ml::train::ExportMethods &method) const override;
      82              : 
      83              :   /**
      84              :    * @copydoc Layer::setProperty(const std::vector<std::string> &values)
      85              :    */
      86              :   void setProperty(const std::vector<std::string> &values) override;
      87              : 
      88              :   /**
      89              :    * @copydoc Layer::getType()
      90              :    */
      91           24 :   const std::string getType() const override {
      92           24 :     return MoLAttentionLayer::type;
      93              :   };
      94              : 
      95              :   /**
      96              :    * @copydoc Layer::setBatch(RunLayerContext &context, unsigned int batch)
      97              :    */
      98              :   void setBatch(RunLayerContext &context, unsigned int batch) override;
      99              : 
     100              :   static constexpr const char *type = "mol_attention";
     101              : 
     102              : private:
     103              :   std::tuple<props::Unit, props::MoL_K>
     104              :     mol_props; /**< mol attention layer properties : unit - number of output
     105              :                   neurons */
     106              : 
     107              :   bool helper_exec; /** check if the helper function has already ran */
     108              :   ActiFunc softmax; /** softmax activation operation */
     109              :   ActiFunc tanh;    /** softmax activation operation */
     110              :   ActiFunc sigmoid; /** softmax activation operation */
     111              :   std::array<unsigned int, 17>
     112              :     wt_idx; /**< indices of the weights and tensors */
     113              : 
     114              :   /**
     115              :    * @brief Helper function for calculation of the derivative
     116              :    *
     117              :    * @param context layer context
     118              :    * @param dstate to store the derivative of the state
     119              :    */
     120              :   void calcDerivativeHelper(RunLayerContext &context, Tensor &dstate);
     121              : };
     122              : 
     123              : } // namespace nntrainer
     124              : 
     125              : #endif /* __cplusplus */
     126              : #endif /* __MOL_ATTENTION_LAYER_H__ */
        

Generated by: LCOV version 2.0-1