ArmNN
 24.08
UnidirectionalSequenceLstmLayer Class Reference

This layer represents a LSTM operation. More...

#include <UnidirectionalSequenceLstmLayer.hpp>

Inheritance diagram for UnidirectionalSequenceLstmLayer:
[legend]
Collaboration diagram for UnidirectionalSequenceLstmLayer:
[legend]

Public Member Functions

virtual std::unique_ptr< IWorkloadCreateWorkload (const IWorkloadFactory &factory) const override
 Makes a workload for the UnidirectionalSequence LSTM type. More...
 
UnidirectionalSequenceLstmLayerClone (Graph &graph) const override
 Creates a dynamically-allocated copy of this layer. More...
 
void ValidateTensorShapesFromInputs () override
 Check if the input tensor shape(s) will lead to a valid configuration of UnidirectionalSequenceLstmLayer. More...
 
std::vector< TensorShapeInferOutputShapes (const std::vector< TensorShape > &inputShapes) const override
 By default returns inputShapes if the number of inputs are equal to number of outputs, otherwise infers the output shapes from given input shapes and layer properties. More...
 
void ExecuteStrategy (IStrategy &strategy) const override
 Apply a visitor to this layer. More...
 
- Public Member Functions inherited from LayerWithParameters< LstmDescriptor >
const LstmDescriptorGetParameters () const override
 If the layer has a descriptor return it. More...
 
void SerializeLayerParameters (ParameterStringifyFunction &fn) const override
 Helper to serialize the layer parameters to string (currently used in DotSerializer and company). More...
 
- Public Member Functions inherited from Layer
 Layer (unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, const char *name)
 
 Layer (unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, DataLayout layout, const char *name)
 
const std::string & GetNameStr () const
 
const OutputHandlerGetOutputHandler (unsigned int i=0) const
 
OutputHandlerGetOutputHandler (unsigned int i=0)
 
ShapeInferenceMethod GetShapeInferenceMethod () const
 
bool GetAllowExpandedDims () const
 
const std::vector< InputSlot > & GetInputSlots () const
 
const std::vector< OutputSlot > & GetOutputSlots () const
 
std::vector< InputSlot >::iterator BeginInputSlots ()
 
std::vector< InputSlot >::iterator EndInputSlots ()
 
std::vector< OutputSlot >::iterator BeginOutputSlots ()
 
std::vector< OutputSlot >::iterator EndOutputSlots ()
 
bool IsOutputUnconnected ()
 
void ResetPriority () const
 
LayerPriority GetPriority () const
 
LayerType GetType () const override
 Returns the armnn::LayerType of this layer. More...
 
DataType GetDataType () const
 
const BackendIdGetBackendId () const
 
void SetBackendId (const BackendId &id) override
 Set the backend of the IConnectableLayer. More...
 
virtual void CreateTensorHandles (const TensorHandleFactoryRegistry &registry, const IWorkloadFactory &factory, const bool IsMemoryManaged=true)
 
void VerifyLayerConnections (unsigned int expectedConnections, const CheckLocation &location) const
 
virtual void ReleaseConstantData ()
 
template<typename Op >
void OperateOnConstantTensors (Op op)
 
const char * GetName () const override
 Returns the name of the layer. More...
 
unsigned int GetNumInputSlots () const override
 Returns the number of connectable input slots. More...
 
unsigned int GetNumOutputSlots () const override
 Returns the number of connectable output slots. More...
 
const InputSlotGetInputSlot (unsigned int index) const override
 Get a const input slot handle by slot index. More...
 
InputSlotGetInputSlot (unsigned int index) override
 Get the input slot handle by slot index. More...
 
const OutputSlotGetOutputSlot (unsigned int index=0) const override
 Get the const output slot handle by slot index. More...
 
OutputSlotGetOutputSlot (unsigned int index=0) override
 Get the output slot handle by slot index. More...
 
void SetGuid (LayerGuid guid)
 
LayerGuid GetGuid () const final
 Returns the unique id of the layer. More...
 
void AddRelatedLayerName (const std::string layerName)
 
const std::list< std::string > & GetRelatedLayerNames ()
 
virtual void Reparent (Graph &dest, std::list< Layer * >::const_iterator iterator)=0
 
void BackendSelectionHint (Optional< BackendId > backend) final
 Provide a hint for the optimizer as to which backend to prefer for this layer. More...
 
Optional< BackendIdGetBackendHint () const
 
void SetShapeInferenceMethod (ShapeInferenceMethod shapeInferenceMethod)
 
void SetAllowExpandedDims (bool allowExpandedDims)
 
template<typename T >
std::shared_ptr< T > GetAdditionalInformation () const
 
void SetAdditionalInfoForObject (const AdditionalInfoObjectPtr &additionalInfo)
 
virtual const BaseDescriptorGetParameters () const override
 If the layer has a descriptor return it. More...
 

Public Attributes

LstmBasicParameters m_BasicParameters
 
LstmOptCifgParameters m_CifgParameters
 
LstmOptProjectionParameters m_ProjectionParameters
 
LstmOptPeepholeParameters m_PeepholeParameters
 
LstmOptLayerNormParameters m_LayerNormParameters
 

Protected Member Functions

 UnidirectionalSequenceLstmLayer (const LstmDescriptor &param, const char *name)
 Constructor to create a UnidirectionalSequenceLstmLayer. More...
 
 ~UnidirectionalSequenceLstmLayer ()=default
 Default destructor. More...
 
Layer::ImmutableConstantTensors GetConstantTensorsByRef () const override
 Retrieve the handles to the constant values stored by the layer. More...
 
- Protected Member Functions inherited from LayerWithParameters< LstmDescriptor >
 LayerWithParameters (unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, const LstmDescriptor &param, const char *name)
 
 ~LayerWithParameters ()=default
 
WorkloadInfo PrepInfoAndDesc (QueueDescriptor &descriptor) const
 Helper function to reduce duplication in *Layer::CreateWorkload. More...
 
void ExecuteStrategy (IStrategy &strategy) const override
 Apply a visitor to this layer. More...
 
Layer::ImmutableConstantTensors GetConnectedConstantAsInputTensors () const
 
- Protected Member Functions inherited from Layer
virtual ~Layer ()=default
 
template<typename QueueDescriptor >
void CollectQueueDescriptorInputs (QueueDescriptor &descriptor, WorkloadInfo &info) const
 
template<typename QueueDescriptor >
void CollectQueueDescriptorOutputs (QueueDescriptor &descriptor, WorkloadInfo &info) const
 
void ValidateAndCopyShape (const TensorShape &outputShape, const TensorShape &inferredShape, const ShapeInferenceMethod shapeInferenceMethod, const std::string &layerName, const unsigned int outputSlotIndex=0)
 
void VerifyShapeInferenceType (const TensorShape &outputShape, ShapeInferenceMethod shapeInferenceMethod)
 
template<typename QueueDescriptor >
WorkloadInfo PrepInfoAndDesc (QueueDescriptor &descriptor) const
 Helper function to reduce duplication in *Layer::CreateWorkload. More...
 
template<typename LayerType , typename ... Params>
LayerTypeCloneBase (Graph &graph, Params &&... params) const
 
virtual ConstantTensors GetConstantTensorsByRef () override final
 
void SetAdditionalInfo (QueueDescriptor &descriptor) const
 
- Protected Member Functions inherited from IConnectableLayer
 ~IConnectableLayer ()
 Objects are not deletable via the handle. More...
 

Additional Inherited Members

- Public Types inherited from LayerWithParameters< LstmDescriptor >
using DescriptorType = LstmDescriptor
 
- Public Types inherited from IConnectableLayer
using ConstantTensors = std::vector< std::reference_wrapper< std::shared_ptr< ConstTensorHandle > >>
 
using ImmutableConstantTensors = std::vector< std::reference_wrapper< const std::shared_ptr< ConstTensorHandle > >>
 
- Protected Attributes inherited from LayerWithParameters< LstmDescriptor >
LstmDescriptor m_Param
 The parameters for the layer (not including tensor-valued weights etc.). More...
 
- Protected Attributes inherited from Layer
AdditionalInfoObjectPtr m_AdditionalInfoObject
 
std::vector< OutputHandlerm_OutputHandlers
 
ShapeInferenceMethod m_ShapeInferenceMethod
 

Detailed Description

This layer represents a LSTM operation.

Definition at line 16 of file UnidirectionalSequenceLstmLayer.hpp.

Constructor & Destructor Documentation

◆ UnidirectionalSequenceLstmLayer()

UnidirectionalSequenceLstmLayer ( const LstmDescriptor param,
const char *  name 
)
protected

Constructor to create a UnidirectionalSequenceLstmLayer.

Parameters
[in]paramLstmDescriptor to configure the lstm operation.
[in]nameOptional name for the layer.

Definition at line 17 of file UnidirectionalSequenceLstmLayer.cpp.

19 {
20 }

References armnn::UnidirectionalSequenceLstm.

◆ ~UnidirectionalSequenceLstmLayer()

~UnidirectionalSequenceLstmLayer ( )
protecteddefault

Default destructor.

Member Function Documentation

◆ Clone()

UnidirectionalSequenceLstmLayer * Clone ( Graph graph) const
overridevirtual

Creates a dynamically-allocated copy of this layer.

Parameters
[in]graphThe graph into which this layer is being cloned.

Implements Layer.

Definition at line 80 of file UnidirectionalSequenceLstmLayer.cpp.

81 {
82  auto layer = CloneBase<UnidirectionalSequenceLstmLayer>(graph, m_Param, GetName());
83 
84  layer->m_BasicParameters.m_InputToForgetWeights = m_BasicParameters.m_InputToForgetWeights ?
86  : nullptr;
87  layer->m_BasicParameters.m_InputToCellWeights = m_BasicParameters.m_InputToCellWeights ?
89  layer->m_BasicParameters.m_InputToOutputWeights = m_BasicParameters.m_InputToOutputWeights ?
91  layer->m_BasicParameters.m_RecurrentToForgetWeights = m_BasicParameters.m_RecurrentToForgetWeights ?
93  layer->m_BasicParameters.m_RecurrentToCellWeights = m_BasicParameters.m_RecurrentToCellWeights ?
95  layer->m_BasicParameters.m_RecurrentToOutputWeights = m_BasicParameters.m_RecurrentToOutputWeights ?
97  layer->m_BasicParameters.m_ForgetGateBias = m_BasicParameters.m_ForgetGateBias ?
99  layer->m_BasicParameters.m_CellBias = m_BasicParameters.m_CellBias ?
100  m_BasicParameters.m_CellBias : nullptr;
101  layer->m_BasicParameters.m_OutputGateBias = m_BasicParameters.m_OutputGateBias ?
103 
104  if (!m_Param.m_CifgEnabled)
105  {
106  layer->m_CifgParameters.m_InputToInputWeights = m_CifgParameters.m_InputToInputWeights ?
108  layer->m_CifgParameters.m_RecurrentToInputWeights = m_CifgParameters.m_RecurrentToInputWeights ?
110  layer->m_CifgParameters.m_InputGateBias = m_CifgParameters.m_InputGateBias ?
112  }
113 
115  {
116  layer->m_ProjectionParameters.m_ProjectionWeights = m_ProjectionParameters.m_ProjectionWeights ?
118  layer->m_ProjectionParameters.m_ProjectionBias = m_ProjectionParameters.m_ProjectionBias ?
120  }
121 
123  {
124  if (!m_Param.m_CifgEnabled)
125  {
126  layer->m_PeepholeParameters.m_CellToInputWeights = m_PeepholeParameters.m_CellToInputWeights ?
128  }
129  layer->m_PeepholeParameters.m_CellToForgetWeights = m_PeepholeParameters.m_CellToForgetWeights ?
131  layer->m_PeepholeParameters.m_CellToOutputWeights = m_PeepholeParameters.m_CellToOutputWeights ?
133  }
134 
136  {
137  layer->m_LayerNormParameters.m_InputLayerNormWeights = m_LayerNormParameters.m_InputLayerNormWeights ?
139  layer->m_LayerNormParameters.m_ForgetLayerNormWeights = m_LayerNormParameters.m_ForgetLayerNormWeights ?
141  layer->m_LayerNormParameters.m_CellLayerNormWeights = m_LayerNormParameters.m_CellLayerNormWeights ?
143  layer->m_LayerNormParameters.m_OutputLayerNormWeights = m_LayerNormParameters.m_OutputLayerNormWeights ?
145  }
146 
147  return std::move(layer);
148 }

References Layer::GetName(), UnidirectionalSequenceLstmLayer::m_BasicParameters, LstmBasicParameters::m_CellBias, LstmOptLayerNormParameters::m_CellLayerNormWeights, LstmOptPeepholeParameters::m_CellToForgetWeights, LstmOptPeepholeParameters::m_CellToInputWeights, LstmOptPeepholeParameters::m_CellToOutputWeights, LstmDescriptor::m_CifgEnabled, UnidirectionalSequenceLstmLayer::m_CifgParameters, LstmBasicParameters::m_ForgetGateBias, LstmOptLayerNormParameters::m_ForgetLayerNormWeights, LstmOptCifgParameters::m_InputGateBias, LstmOptLayerNormParameters::m_InputLayerNormWeights, LstmBasicParameters::m_InputToCellWeights, LstmBasicParameters::m_InputToForgetWeights, LstmOptCifgParameters::m_InputToInputWeights, LstmBasicParameters::m_InputToOutputWeights, LstmDescriptor::m_LayerNormEnabled, UnidirectionalSequenceLstmLayer::m_LayerNormParameters, LstmBasicParameters::m_OutputGateBias, LstmOptLayerNormParameters::m_OutputLayerNormWeights, LayerWithParameters< LstmDescriptor >::m_Param, LstmDescriptor::m_PeepholeEnabled, UnidirectionalSequenceLstmLayer::m_PeepholeParameters, LstmOptProjectionParameters::m_ProjectionBias, LstmDescriptor::m_ProjectionEnabled, UnidirectionalSequenceLstmLayer::m_ProjectionParameters, LstmOptProjectionParameters::m_ProjectionWeights, LstmBasicParameters::m_RecurrentToCellWeights, LstmBasicParameters::m_RecurrentToForgetWeights, LstmOptCifgParameters::m_RecurrentToInputWeights, and LstmBasicParameters::m_RecurrentToOutputWeights.

◆ CreateWorkload()

std::unique_ptr< IWorkload > CreateWorkload ( const IWorkloadFactory factory) const
overridevirtual

Makes a workload for the UnidirectionalSequence LSTM type.

Parameters
[in]graphThe graph where this layer can be found.
[in]factoryThe workload factory which will create the workload.
Returns
A pointer to the created workload, or nullptr if not created.

Implements Layer.

Definition at line 22 of file UnidirectionalSequenceLstmLayer.cpp.

23 {
24  UnidirectionalSequenceLstmQueueDescriptor descriptor;
25 
26  // Basic parameters
27  descriptor.m_InputToForgetWeights = m_BasicParameters.m_InputToForgetWeights.get();
28  descriptor.m_InputToCellWeights = m_BasicParameters.m_InputToCellWeights.get();
29  descriptor.m_InputToOutputWeights = m_BasicParameters.m_InputToOutputWeights.get();
30  descriptor.m_RecurrentToForgetWeights = m_BasicParameters.m_RecurrentToForgetWeights.get();
31  descriptor.m_RecurrentToCellWeights = m_BasicParameters.m_RecurrentToCellWeights.get();
32  descriptor.m_RecurrentToOutputWeights = m_BasicParameters.m_RecurrentToOutputWeights.get();
33  descriptor.m_ForgetGateBias = m_BasicParameters.m_ForgetGateBias.get();
34  descriptor.m_CellBias = m_BasicParameters.m_CellBias.get();
35  descriptor.m_OutputGateBias = m_BasicParameters.m_OutputGateBias.get();
36 
37  // Cifg parameters
39  {
40  descriptor.m_InputToInputWeights = m_CifgParameters.m_InputToInputWeights.get();
41  descriptor.m_RecurrentToInputWeights = m_CifgParameters.m_RecurrentToInputWeights.get();
42  descriptor.m_InputGateBias = m_CifgParameters.m_InputGateBias.get();
43  }
44 
45  // Projection parameters
47  {
48  descriptor.m_ProjectionWeights = m_ProjectionParameters.m_ProjectionWeights.get();
49  descriptor.m_ProjectionBias = m_ProjectionParameters.m_ProjectionBias.get();
50  }
51 
52  // Peephole parameters
54  {
56  {
57  descriptor.m_CellToInputWeights = m_PeepholeParameters.m_CellToInputWeights.get();
58  }
59  descriptor.m_CellToForgetWeights = m_PeepholeParameters.m_CellToForgetWeights.get();
60  descriptor.m_CellToOutputWeights = m_PeepholeParameters.m_CellToOutputWeights.get();
61  }
62 
63  // Layer normalisation parameters
65  {
67  {
68  descriptor.m_InputLayerNormWeights = m_LayerNormParameters.m_InputLayerNormWeights.get();
69  }
70  descriptor.m_ForgetLayerNormWeights = m_LayerNormParameters.m_ForgetLayerNormWeights.get();
71  descriptor.m_CellLayerNormWeights = m_LayerNormParameters.m_CellLayerNormWeights.get();
72  descriptor.m_OutputLayerNormWeights = m_LayerNormParameters.m_OutputLayerNormWeights.get();
73  }
74 
75  SetAdditionalInfo(descriptor);
76 
77  return factory.CreateWorkload(LayerType::UnidirectionalSequenceLstm, descriptor, PrepInfoAndDesc(descriptor));
78 }

References IWorkloadFactory::CreateWorkload(), UnidirectionalSequenceLstmLayer::m_BasicParameters, LstmBasicParameters::m_CellBias, UnidirectionalSequenceLstmQueueDescriptor::m_CellBias, LstmOptLayerNormParameters::m_CellLayerNormWeights, UnidirectionalSequenceLstmQueueDescriptor::m_CellLayerNormWeights, LstmOptPeepholeParameters::m_CellToForgetWeights, UnidirectionalSequenceLstmQueueDescriptor::m_CellToForgetWeights, LstmOptPeepholeParameters::m_CellToInputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_CellToInputWeights, LstmOptPeepholeParameters::m_CellToOutputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_CellToOutputWeights, LstmDescriptor::m_CifgEnabled, UnidirectionalSequenceLstmLayer::m_CifgParameters, LstmBasicParameters::m_ForgetGateBias, UnidirectionalSequenceLstmQueueDescriptor::m_ForgetGateBias, LstmOptLayerNormParameters::m_ForgetLayerNormWeights, UnidirectionalSequenceLstmQueueDescriptor::m_ForgetLayerNormWeights, LstmOptCifgParameters::m_InputGateBias, UnidirectionalSequenceLstmQueueDescriptor::m_InputGateBias, LstmOptLayerNormParameters::m_InputLayerNormWeights, UnidirectionalSequenceLstmQueueDescriptor::m_InputLayerNormWeights, LstmBasicParameters::m_InputToCellWeights, UnidirectionalSequenceLstmQueueDescriptor::m_InputToCellWeights, LstmBasicParameters::m_InputToForgetWeights, UnidirectionalSequenceLstmQueueDescriptor::m_InputToForgetWeights, LstmOptCifgParameters::m_InputToInputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_InputToInputWeights, LstmBasicParameters::m_InputToOutputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_InputToOutputWeights, LstmDescriptor::m_LayerNormEnabled, UnidirectionalSequenceLstmLayer::m_LayerNormParameters, LstmBasicParameters::m_OutputGateBias, UnidirectionalSequenceLstmQueueDescriptor::m_OutputGateBias, LstmOptLayerNormParameters::m_OutputLayerNormWeights, UnidirectionalSequenceLstmQueueDescriptor::m_OutputLayerNormWeights, LayerWithParameters< LstmDescriptor >::m_Param, LstmDescriptor::m_PeepholeEnabled, UnidirectionalSequenceLstmLayer::m_PeepholeParameters, LstmOptProjectionParameters::m_ProjectionBias, UnidirectionalSequenceLstmQueueDescriptor::m_ProjectionBias, LstmDescriptor::m_ProjectionEnabled, UnidirectionalSequenceLstmLayer::m_ProjectionParameters, LstmOptProjectionParameters::m_ProjectionWeights, UnidirectionalSequenceLstmQueueDescriptor::m_ProjectionWeights, LstmBasicParameters::m_RecurrentToCellWeights, UnidirectionalSequenceLstmQueueDescriptor::m_RecurrentToCellWeights, LstmBasicParameters::m_RecurrentToForgetWeights, UnidirectionalSequenceLstmQueueDescriptor::m_RecurrentToForgetWeights, LstmOptCifgParameters::m_RecurrentToInputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_RecurrentToInputWeights, LstmBasicParameters::m_RecurrentToOutputWeights, UnidirectionalSequenceLstmQueueDescriptor::m_RecurrentToOutputWeights, LayerWithParameters< LstmDescriptor >::PrepInfoAndDesc(), Layer::SetAdditionalInfo(), and armnn::UnidirectionalSequenceLstm.

◆ ExecuteStrategy()

void ExecuteStrategy ( IStrategy strategy) const
overridevirtual

Apply a visitor to this layer.

Reimplemented from Layer.

Definition at line 397 of file UnidirectionalSequenceLstmLayer.cpp.

398 {
399  std::vector<ConstTensor> constTensors;
400 
401  LstmDescriptor descriptor = GetParameters();
402 
403  ManagedConstTensorHandle managedInputToForgetWeights(m_BasicParameters.m_InputToForgetWeights);
404  ManagedConstTensorHandle managedInputToCellWeights(m_BasicParameters.m_InputToCellWeights);
405  ManagedConstTensorHandle managedInputToOutputWeights(m_BasicParameters.m_InputToOutputWeights);
406  ManagedConstTensorHandle managedRecurrentToForgetWeights(m_BasicParameters.m_RecurrentToForgetWeights);
407  ManagedConstTensorHandle managedRecurrentToCellWeights(m_BasicParameters.m_RecurrentToCellWeights);
408  ManagedConstTensorHandle managedRecurrentToOutputWeights(m_BasicParameters.m_RecurrentToOutputWeights);
409  ManagedConstTensorHandle managedForgetGateBias(m_BasicParameters.m_ForgetGateBias);
410  ManagedConstTensorHandle managedCellBias(m_BasicParameters.m_CellBias);
411  ManagedConstTensorHandle managedOutputGateBias(m_BasicParameters.m_OutputGateBias);
412 
413  // Cifg parameters
414  ManagedConstTensorHandle managedInputToInputWeights(m_CifgParameters.m_InputToInputWeights);
415  ManagedConstTensorHandle managedRecurrentToInputWeights(m_CifgParameters.m_RecurrentToInputWeights);
416  ManagedConstTensorHandle managedInputGateBias(m_CifgParameters.m_InputGateBias);
417 
418  // Projection parameters
419  ManagedConstTensorHandle managedProjectionWeights(m_ProjectionParameters.m_ProjectionWeights);
420  ManagedConstTensorHandle managedProjectionBias(m_ProjectionParameters.m_ProjectionBias);
421 
422  // Peephole parameters
423  ManagedConstTensorHandle managedCellToInputWeights(m_PeepholeParameters.m_CellToInputWeights);
424  ManagedConstTensorHandle managedCellToForgetWeights(m_PeepholeParameters.m_CellToForgetWeights);
425  ManagedConstTensorHandle managedCellToOutputWeights(m_PeepholeParameters.m_CellToOutputWeights);
426 
427  // Layer normalisation parameters
428  ManagedConstTensorHandle managedInputLayerNormWeights(m_LayerNormParameters.m_InputLayerNormWeights);
429  ManagedConstTensorHandle managedForgetLayerNormWeights(m_LayerNormParameters.m_ForgetLayerNormWeights);
430  ManagedConstTensorHandle managedCellLayerNormWeights(m_LayerNormParameters.m_CellLayerNormWeights);
431  ManagedConstTensorHandle managedOutputLayerNormWeights(m_LayerNormParameters.m_OutputLayerNormWeights);
432 
433  // First add mandatory/basic parameters
435  {
436  constTensors.emplace_back(ConstTensor(managedInputToForgetWeights.GetTensorInfo(),
437  managedInputToForgetWeights.Map()));
438  }
440  {
441  constTensors.emplace_back(ConstTensor(managedInputToCellWeights.GetTensorInfo(),
442  managedInputToCellWeights.Map()));
443  }
445  {
446  constTensors.emplace_back(ConstTensor(managedInputToOutputWeights.GetTensorInfo(),
447  managedInputToOutputWeights.Map()));
448  }
450  {
451  constTensors.emplace_back(ConstTensor(
452  managedRecurrentToForgetWeights.GetTensorInfo(),
453  managedRecurrentToForgetWeights.Map()));
454  }
456  {
457  constTensors.emplace_back(ConstTensor(
458  managedRecurrentToCellWeights.GetTensorInfo(),
459  managedRecurrentToCellWeights.Map()));
460  }
462  {
463  constTensors.emplace_back(ConstTensor(
464  managedRecurrentToOutputWeights.GetTensorInfo(),
465  managedRecurrentToOutputWeights.Map()));
466  }
467  if (m_BasicParameters.m_ForgetGateBias != nullptr)
468  {
469  constTensors.emplace_back(ConstTensor(managedForgetGateBias.GetTensorInfo(),
470  managedForgetGateBias.Map()));
471  }
472  if (m_BasicParameters.m_CellBias != nullptr)
473  {
474  constTensors.emplace_back(ConstTensor(managedCellBias.GetTensorInfo(),
475  managedCellBias.Map()));
476  }
477  if (m_BasicParameters.m_OutputGateBias != nullptr)
478  {
479  constTensors.emplace_back(ConstTensor(managedOutputGateBias.GetTensorInfo(),
480  managedOutputGateBias.Map()));
481  }
482 
483  // Add cifg parameters
484  if (!descriptor.m_CifgEnabled)
485  {
487  {
488  constTensors.emplace_back(ConstTensor(managedInputToInputWeights.GetTensorInfo(),
489  managedInputToInputWeights.Map()));
490  }
492  {
493  constTensors.emplace_back(ConstTensor(
494  managedRecurrentToInputWeights.GetTensorInfo(),
495  managedRecurrentToInputWeights.Map()));
496  }
497  if (m_CifgParameters.m_InputGateBias != nullptr)
498  {
499  constTensors.emplace_back(ConstTensor(managedInputGateBias.GetTensorInfo(),
500  managedInputGateBias.Map()));
501  }
502  }
503 
504  // Add peephole parameters
505  if (descriptor.m_PeepholeEnabled)
506  {
507  if (!descriptor.m_CifgEnabled)
508  {
510  {
511  constTensors.emplace_back(ConstTensor(managedCellToInputWeights.GetTensorInfo(),
512  managedCellToInputWeights.Map()));
513  }
514  }
516  {
517  constTensors.emplace_back(ConstTensor(managedCellToForgetWeights.GetTensorInfo(),
518  managedCellToForgetWeights.Map()));
519  }
521  {
522  constTensors.emplace_back(ConstTensor(managedCellToOutputWeights.GetTensorInfo(),
523  managedCellToOutputWeights.Map()));
524  }
525  }
526 
527  // Add projection parameters
528  if (descriptor.m_ProjectionEnabled)
529  {
531  {
532  constTensors.emplace_back(ConstTensor(managedProjectionWeights.GetTensorInfo(),
533  managedProjectionWeights.Map()));
534  }
536  {
537  constTensors.emplace_back(ConstTensor(managedProjectionBias.GetTensorInfo(),
538  managedProjectionBias.Map()));
539  }
540  }
541 
542  // Add norm parameters
543  if (descriptor.m_LayerNormEnabled)
544  {
545  if (!descriptor.m_CifgEnabled)
546  {
548  {
549  constTensors.emplace_back(ConstTensor(managedInputLayerNormWeights.GetTensorInfo(),
550  managedInputLayerNormWeights.Map()));
551  }
552  }
554  {
555  constTensors.emplace_back(ConstTensor(managedForgetLayerNormWeights.GetTensorInfo(),
556  managedForgetLayerNormWeights.Map()));
557  }
559  {
560  constTensors.emplace_back(ConstTensor(managedCellLayerNormWeights.GetTensorInfo(),
561  managedCellLayerNormWeights.Map()));
562  }
564  {
565  constTensors.emplace_back(ConstTensor(managedOutputLayerNormWeights.GetTensorInfo(),
566  managedOutputLayerNormWeights.Map()));
567  }
568  }
569 
570  strategy.ExecuteStrategy(this, GetParameters(), constTensors, GetName());
571 }

References IStrategy::ExecuteStrategy(), Layer::GetName(), LayerWithParameters< LstmDescriptor >::GetParameters(), ManagedConstTensorHandle::GetTensorInfo(), UnidirectionalSequenceLstmLayer::m_BasicParameters, LstmBasicParameters::m_CellBias, LstmOptLayerNormParameters::m_CellLayerNormWeights, LstmOptPeepholeParameters::m_CellToForgetWeights, LstmOptPeepholeParameters::m_CellToInputWeights, LstmOptPeepholeParameters::m_CellToOutputWeights, LstmDescriptor::m_CifgEnabled, UnidirectionalSequenceLstmLayer::m_CifgParameters, LstmBasicParameters::m_ForgetGateBias, LstmOptLayerNormParameters::m_ForgetLayerNormWeights, LstmOptCifgParameters::m_InputGateBias, LstmOptLayerNormParameters::m_InputLayerNormWeights, LstmBasicParameters::m_InputToCellWeights, LstmBasicParameters::m_InputToForgetWeights, LstmOptCifgParameters::m_InputToInputWeights, LstmBasicParameters::m_InputToOutputWeights, LstmDescriptor::m_LayerNormEnabled, UnidirectionalSequenceLstmLayer::m_LayerNormParameters, LstmBasicParameters::m_OutputGateBias, LstmOptLayerNormParameters::m_OutputLayerNormWeights, LstmDescriptor::m_PeepholeEnabled, UnidirectionalSequenceLstmLayer::m_PeepholeParameters, LstmOptProjectionParameters::m_ProjectionBias, LstmDescriptor::m_ProjectionEnabled, UnidirectionalSequenceLstmLayer::m_ProjectionParameters, LstmOptProjectionParameters::m_ProjectionWeights, LstmBasicParameters::m_RecurrentToCellWeights, LstmBasicParameters::m_RecurrentToForgetWeights, LstmOptCifgParameters::m_RecurrentToInputWeights, LstmBasicParameters::m_RecurrentToOutputWeights, and ManagedConstTensorHandle::Map().

◆ GetConstantTensorsByRef()

Layer::ImmutableConstantTensors GetConstantTensorsByRef ( ) const
overrideprotectedvirtual

Retrieve the handles to the constant values stored by the layer.

Returns
A vector of the constant tensors stored by this layer.

Reimplemented from Layer.

Definition at line 363 of file UnidirectionalSequenceLstmLayer.cpp.

References UnidirectionalSequenceLstmLayer::m_BasicParameters, LstmBasicParameters::m_CellBias, LstmOptLayerNormParameters::m_CellLayerNormWeights, LstmOptPeepholeParameters::m_CellToForgetWeights, LstmOptPeepholeParameters::m_CellToInputWeights, LstmOptPeepholeParameters::m_CellToOutputWeights, UnidirectionalSequenceLstmLayer::m_CifgParameters, LstmBasicParameters::m_ForgetGateBias, LstmOptLayerNormParameters::m_ForgetLayerNormWeights, LstmOptCifgParameters::m_InputGateBias, LstmOptLayerNormParameters::m_InputLayerNormWeights, LstmBasicParameters::m_InputToCellWeights, LstmBasicParameters::m_InputToForgetWeights, LstmOptCifgParameters::m_InputToInputWeights, LstmBasicParameters::m_InputToOutputWeights, UnidirectionalSequenceLstmLayer::m_LayerNormParameters, LstmBasicParameters::m_OutputGateBias, LstmOptLayerNormParameters::m_OutputLayerNormWeights, UnidirectionalSequenceLstmLayer::m_PeepholeParameters, LstmOptProjectionParameters::m_ProjectionBias, UnidirectionalSequenceLstmLayer::m_ProjectionParameters, LstmOptProjectionParameters::m_ProjectionWeights, LstmBasicParameters::m_RecurrentToCellWeights, LstmBasicParameters::m_RecurrentToForgetWeights, LstmOptCifgParameters::m_RecurrentToInputWeights, and LstmBasicParameters::m_RecurrentToOutputWeights.

◆ InferOutputShapes()

std::vector< TensorShape > InferOutputShapes ( const std::vector< TensorShape > &  inputShapes) const
overridevirtual

By default returns inputShapes if the number of inputs are equal to number of outputs, otherwise infers the output shapes from given input shapes and layer properties.

Parameters
[in]inputShapesThe input shapes layer has.
Returns
A vector to the inferred output shape.

Reimplemented from Layer.

Definition at line 150 of file UnidirectionalSequenceLstmLayer.cpp.

152 {
153  ARMNN_THROW_INVALIDARG_MSG_IF_FALSE(inputShapes.size() == 3,
154  "inputShapes' size is \"" + std::to_string(inputShapes.size()) +
155  "\" - should be \"3\".");
156 
157  // Get input values for validation
158  unsigned int outputSize = inputShapes[1][1];
159 
160  std::vector<TensorShape> outShapes;
161  if (m_Param.m_TimeMajor)
162  {
163  outShapes.push_back(TensorShape({inputShapes[0][0], inputShapes[0][1], outputSize}));
164  }
165  else
166  {
167  outShapes.push_back(TensorShape({inputShapes[0][0], inputShapes[0][1], outputSize}));
168  }
169  return outShapes;
170 }

References ARMNN_THROW_INVALIDARG_MSG_IF_FALSE, LayerWithParameters< LstmDescriptor >::m_Param, and LstmDescriptor::m_TimeMajor.

Referenced by UnidirectionalSequenceLstmLayer::ValidateTensorShapesFromInputs().

◆ ValidateTensorShapesFromInputs()

void ValidateTensorShapesFromInputs ( )
overridevirtual

Check if the input tensor shape(s) will lead to a valid configuration of UnidirectionalSequenceLstmLayer.

Parameters
[in]shapeInferenceMethodIndicates if output shape shall be overwritten or just validated.

Implements Layer.

Definition at line 172 of file UnidirectionalSequenceLstmLayer.cpp.

173 {
175 
176  const TensorShape& outputShape = GetOutputSlot(2).GetTensorInfo().GetShape();
177 
179 
180  auto inferredShapes = InferOutputShapes( {
184  });
185 
186  if (inferredShapes.size() != 1)
187  {
188  throw armnn::LayerValidationException("inferredShapes has "
189  + std::to_string(inferredShapes.size()) +
190  " elements - should only have 1.");
191  }
192 
193  // Check if the weights are nullptr
195  {
196  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
197  "m_BasicParameters.m_InputToForgetWeights should not be null.");
198  }
199 
201  {
202  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
203  "m_BasicParameters.m_InputToCellWeights should not be null.");
204  }
205 
207  {
208  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
209  "m_BasicParameters.m_InputToOutputWeights should not be null.");
210  }
211 
213  {
214  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
215  "m_BasicParameters.m_RecurrentToForgetWeights should not be null.");
216  }
217 
219  {
220  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
221  "m_BasicParameters.m_RecurrentToCellWeights should not be null.");
222  }
223 
225  {
226  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
227  "m_BasicParameters.m_RecurrentToOutputWeights should not be null.");
228  }
229 
231  {
232  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
233  "m_BasicParameters.m_ForgetGateBias should not be null.");
234  }
235 
237  {
238  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
239  "m_BasicParameters.m_CellBias should not be null.");
240  }
241 
243  {
244  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
245  "m_BasicParameters.m_OutputGateBias should not be null.");
246  }
247 
248  if (!m_Param.m_CifgEnabled)
249  {
251  {
252  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
253  "m_CifgParameters.m_InputToInputWeights should not be null.");
254  }
255 
257  {
258  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
259  "m_CifgParameters.m_RecurrentToInputWeights should not be null.");
260  }
261 
263  {
264  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
265  "m_CifgParameters.m_InputGateBias should not be null.");
266  }
267  }
268  else
269  {
271  {
272  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
273  "m_CifgParameters.m_InputToInputWeights should not have a value "
274  "when CIFG is enabled.");
275  }
276 
278  {
279  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
280  "m_CifgParameters.m_RecurrentToInputWeights should not have a value "
281  "when CIFG is enabled.");
282  }
283 
285  {
286  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
287  "m_CifgParameters.m_InputGateBias should not have a value "
288  "when CIFG is enabled.");
289  }
290  }
291 
293  {
295  {
296  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
297  "m_ProjectionParameters.m_ProjectionWeights should not be null.");
298  }
299  }
300 
302  {
303  if (!m_Param.m_CifgEnabled)
304  {
306  {
307  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
308  "m_PeepholeParameters.m_CellToInputWeights should not be null "
309  "when Peephole is enabled and CIFG is disabled.");
310  }
311  }
312 
314  {
315  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
316  "m_PeepholeParameters.m_CellToForgetWeights should not be null.");
317  }
318 
320  {
321  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
322  "m_PeepholeParameters.m_CellToOutputWeights should not be null.");
323  }
324  }
325 
327  {
329  {
331  {
332  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
333  "m_LayerNormParameters.m_inputLayerNormWeights "
334  "should not be null.");
335  }
336  }
337 
339  {
340  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
341  "m_LayerNormParameters.m_forgetLayerNormWeights "
342  "should not be null.");
343  }
344 
346  {
347  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
348  "m_LayerNormParameters.m_cellLayerNormWeights "
349  "should not be null.");
350  }
351 
353  {
354  throw armnn::LayerValidationException("UnidirectionalSequenceLstmLayer: "
355  "m_LayerNormParameters.m_outputLayerNormWeights "
356  "should not be null.");
357  }
358  }
359 
360  ValidateAndCopyShape(outputShape, inferredShapes[0], m_ShapeInferenceMethod, "UnidirectionalSequenceLstmLayer");
361 }

References CHECK_LOCATION, Layer::GetInputSlot(), Layer::GetOutputSlot(), TensorInfo::GetShape(), InputSlot::GetTensorInfo(), OutputSlot::GetTensorInfo(), UnidirectionalSequenceLstmLayer::InferOutputShapes(), UnidirectionalSequenceLstmLayer::m_BasicParameters, LstmBasicParameters::m_CellBias, LstmOptLayerNormParameters::m_CellLayerNormWeights, LstmOptPeepholeParameters::m_CellToForgetWeights, LstmOptPeepholeParameters::m_CellToInputWeights, LstmOptPeepholeParameters::m_CellToOutputWeights, LstmDescriptor::m_CifgEnabled, UnidirectionalSequenceLstmLayer::m_CifgParameters, LstmBasicParameters::m_ForgetGateBias, LstmOptLayerNormParameters::m_ForgetLayerNormWeights, LstmOptCifgParameters::m_InputGateBias, LstmOptLayerNormParameters::m_InputLayerNormWeights, LstmBasicParameters::m_InputToCellWeights, LstmBasicParameters::m_InputToForgetWeights, LstmOptCifgParameters::m_InputToInputWeights, LstmBasicParameters::m_InputToOutputWeights, LstmDescriptor::m_LayerNormEnabled, UnidirectionalSequenceLstmLayer::m_LayerNormParameters, LstmBasicParameters::m_OutputGateBias, LstmOptLayerNormParameters::m_OutputLayerNormWeights, LayerWithParameters< LstmDescriptor >::m_Param, LstmDescriptor::m_PeepholeEnabled, UnidirectionalSequenceLstmLayer::m_PeepholeParameters, LstmDescriptor::m_ProjectionEnabled, UnidirectionalSequenceLstmLayer::m_ProjectionParameters, LstmOptProjectionParameters::m_ProjectionWeights, LstmBasicParameters::m_RecurrentToCellWeights, LstmBasicParameters::m_RecurrentToForgetWeights, LstmOptCifgParameters::m_RecurrentToInputWeights, LstmBasicParameters::m_RecurrentToOutputWeights, Layer::m_ShapeInferenceMethod, Layer::ValidateAndCopyShape(), Layer::VerifyLayerConnections(), and Layer::VerifyShapeInferenceType().

Member Data Documentation

◆ m_BasicParameters

◆ m_CifgParameters

◆ m_LayerNormParameters

◆ m_PeepholeParameters

◆ m_ProjectionParameters


The documentation for this class was generated from the following files:
armnn::LstmOptProjectionParameters::m_ProjectionWeights
std::shared_ptr< ConstTensorHandle > m_ProjectionWeights
A unique pointer to represent 2D weights tensor with dimensions [output_size, num_units].
Definition: LstmParameters.hpp:39
armnn::LstmOptLayerNormParameters::m_OutputLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_OutputLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:23
armnn::LstmDescriptor::m_TimeMajor
bool m_TimeMajor
Enable/disable time major.
Definition: Descriptors.hpp:1154
armnn::OutputSlot::GetTensorInfo
const TensorInfo & GetTensorInfo() const override
Definition: Layer.cpp:100
armnn::LstmBasicParameters::m_InputToCellWeights
std::shared_ptr< ConstTensorHandle > m_InputToCellWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units].
Definition: LstmParameters.hpp:59
armnn::LstmOptLayerNormParameters::m_InputLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_InputLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:17
armnn::LstmOptCifgParameters::m_InputGateBias
std::shared_ptr< ConstTensorHandle > m_InputGateBias
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:33
armnn::LstmOptPeepholeParameters::m_CellToInputWeights
std::shared_ptr< ConstTensorHandle > m_CellToInputWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:47
CHECK_LOCATION
#define CHECK_LOCATION()
Definition: Exceptions.hpp:203
armnn::Layer::ValidateAndCopyShape
void ValidateAndCopyShape(const TensorShape &outputShape, const TensorShape &inferredShape, const ShapeInferenceMethod shapeInferenceMethod, const std::string &layerName, const unsigned int outputSlotIndex=0)
Definition: Layer.cpp:457
armnn::UnidirectionalSequenceLstmLayer::m_BasicParameters
LstmBasicParameters m_BasicParameters
Definition: UnidirectionalSequenceLstmLayer.hpp:20
armnn::Layer::GetOutputSlot
const OutputSlot & GetOutputSlot(unsigned int index=0) const override
Get the const output slot handle by slot index.
Definition: Layer.hpp:339
armnn::Layer::GetInputSlot
const InputSlot & GetInputSlot(unsigned int index) const override
Get a const input slot handle by slot index.
Definition: Layer.hpp:337
armnn::LayerWithParameters< LstmDescriptor >::GetParameters
const LstmDescriptor & GetParameters() const override
Definition: LayerWithParameters.hpp:19
armnn::LstmOptProjectionParameters::m_ProjectionBias
std::shared_ptr< ConstTensorHandle > m_ProjectionBias
A unique pointer to represent 1D weights tensor with dimensions [output_size].
Definition: LstmParameters.hpp:41
armnn::LstmBasicParameters::m_InputToOutputWeights
std::shared_ptr< ConstTensorHandle > m_InputToOutputWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units].
Definition: LstmParameters.hpp:61
armnn::Layer::GetName
const char * GetName() const override
Returns the name of the layer.
Definition: Layer.hpp:332
armnn::InputSlot::GetTensorInfo
const TensorInfo & GetTensorInfo() const override
Gets the TensorInfo for this InputSlot.
Definition: Layer.cpp:614
armnn::LstmBasicParameters::m_ForgetGateBias
std::shared_ptr< ConstTensorHandle > m_ForgetGateBias
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:69
armnn::LstmOptLayerNormParameters::m_CellLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_CellLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:21
armnn::LstmDescriptor::m_PeepholeEnabled
bool m_PeepholeEnabled
Enable/disable peephole.
Definition: Descriptors.hpp:1148
armnn::LayerWithParameters< LstmDescriptor >::m_Param
LstmDescriptor m_Param
The parameters for the layer (not including tensor-valued weights etc.).
Definition: LayerWithParameters.hpp:52
armnn::LstmOptCifgParameters::m_RecurrentToInputWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToInputWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units].
Definition: LstmParameters.hpp:31
armnn::LayerWithParameters< LstmDescriptor >::PrepInfoAndDesc
WorkloadInfo PrepInfoAndDesc(QueueDescriptor &descriptor) const
Helper function to reduce duplication in *Layer::CreateWorkload.
Definition: LayerWithParameters.hpp:44
armnn::LayerValidationException
Definition: Exceptions.hpp:105
armnn::Layer::VerifyShapeInferenceType
void VerifyShapeInferenceType(const TensorShape &outputShape, ShapeInferenceMethod shapeInferenceMethod)
Definition: Layer.cpp:526
armnn::Layer::SetAdditionalInfo
void SetAdditionalInfo(QueueDescriptor &descriptor) const
Definition: Layer.cpp:303
armnn::UnidirectionalSequenceLstmLayer::m_PeepholeParameters
LstmOptPeepholeParameters m_PeepholeParameters
Definition: UnidirectionalSequenceLstmLayer.hpp:23
armnn::LstmBasicParameters::m_OutputGateBias
std::shared_ptr< ConstTensorHandle > m_OutputGateBias
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:73
armnn::LstmBasicParameters::m_CellBias
std::shared_ptr< ConstTensorHandle > m_CellBias
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:71
armnn::LstmOptCifgParameters::m_InputToInputWeights
std::shared_ptr< ConstTensorHandle > m_InputToInputWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units].
Definition: LstmParameters.hpp:29
armnn::LstmBasicParameters::m_InputToForgetWeights
std::shared_ptr< ConstTensorHandle > m_InputToForgetWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units].
Definition: LstmParameters.hpp:57
armnn::LstmDescriptor::m_CifgEnabled
bool m_CifgEnabled
Enable/disable cifg (coupled input & forget gate).
Definition: Descriptors.hpp:1146
armnn::LstmBasicParameters::m_RecurrentToCellWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToCellWeights
A unique pointer to represent 2D weights tensor with dimensions [output_size, num_units].
Definition: LstmParameters.hpp:65
armnn::LstmBasicParameters::m_RecurrentToForgetWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToForgetWeights
A unique pointer to represent 2D weights tensor with dimensions [output_size, num_units].
Definition: LstmParameters.hpp:63
armnn::TensorInfo::GetShape
const TensorShape & GetShape() const
Definition: Tensor.hpp:193
armnn::UnidirectionalSequenceLstmLayer::InferOutputShapes
std::vector< TensorShape > InferOutputShapes(const std::vector< TensorShape > &inputShapes) const override
By default returns inputShapes if the number of inputs are equal to number of outputs,...
Definition: UnidirectionalSequenceLstmLayer.cpp:150
armnn::LstmBasicParameters::m_RecurrentToOutputWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToOutputWeights
A unique pointer to represent 2D weights tensor with dimensions [output_size, num_units].
Definition: LstmParameters.hpp:67
armnn::LstmDescriptor::m_LayerNormEnabled
bool m_LayerNormEnabled
Enable/disable layer normalization.
Definition: Descriptors.hpp:1152
armnn::LayerType::UnidirectionalSequenceLstm
@ UnidirectionalSequenceLstm
armnn::LstmOptPeepholeParameters::m_CellToOutputWeights
std::shared_ptr< ConstTensorHandle > m_CellToOutputWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:51
armnn::UnidirectionalSequenceLstmLayer::m_LayerNormParameters
LstmOptLayerNormParameters m_LayerNormParameters
Definition: UnidirectionalSequenceLstmLayer.hpp:24
armnn::LstmOptLayerNormParameters::m_ForgetLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_ForgetLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:19
armnn::UnidirectionalSequenceLstmLayer::m_CifgParameters
LstmOptCifgParameters m_CifgParameters
Definition: UnidirectionalSequenceLstmLayer.hpp:21
armnn::LstmDescriptor::m_ProjectionEnabled
bool m_ProjectionEnabled
Enable/disable the projection layer.
Definition: Descriptors.hpp:1150
armnn::Layer::VerifyLayerConnections
void VerifyLayerConnections(unsigned int expectedConnections, const CheckLocation &location) const
Definition: Layer.cpp:410
armnn::LayerWithParameters< LstmDescriptor >::LayerWithParameters
LayerWithParameters(unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, const LstmDescriptor &param, const char *name)
Definition: LayerWithParameters.hpp:30
armnn::Layer::m_ShapeInferenceMethod
ShapeInferenceMethod m_ShapeInferenceMethod
Definition: Layer.hpp:441
armnn::LstmOptPeepholeParameters::m_CellToForgetWeights
std::shared_ptr< ConstTensorHandle > m_CellToForgetWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units].
Definition: LstmParameters.hpp:49
armnn::UnidirectionalSequenceLstmLayer::m_ProjectionParameters
LstmOptProjectionParameters m_ProjectionParameters
Definition: UnidirectionalSequenceLstmLayer.hpp:22
ARMNN_THROW_INVALIDARG_MSG_IF_FALSE
#define ARMNN_THROW_INVALIDARG_MSG_IF_FALSE(_cond, _str)
Definition: Exceptions.hpp:210