ArmNN
 24.02
QLstmLayer Class Reference

This layer represents a QLstm operation. More...

#include <QLstmLayer.hpp>

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

Public Member Functions

virtual std::unique_ptr< IWorkloadCreateWorkload (const IWorkloadFactory &factory) const override
 Makes a workload for the QLstm type. More...
 
QLstmLayerClone (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 QLstmLayer. 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< QLstmDescriptor >
const QLstmDescriptorGetParameters () 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

QLstmBasicParameters m_BasicParameters
 
QLstmOptCifgParameters m_CifgParameters
 
QLstmOptProjectionParameters m_ProjectionParameters
 
QLstmOptPeepholeParameters m_PeepholeParameters
 
QLstmOptLayerNormParameters m_LayerNormParameters
 

Protected Member Functions

 QLstmLayer (const QLstmDescriptor &param, const char *name)
 Constructor to create a QLstmLayer. More...
 
 ~QLstmLayer ()=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< QLstmDescriptor >
 LayerWithParameters (unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, const QLstmDescriptor &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< QLstmDescriptor >
using DescriptorType = QLstmDescriptor
 
- 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< QLstmDescriptor >
QLstmDescriptor 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 QLstm operation.

Definition at line 79 of file QLstmLayer.hpp.

Constructor & Destructor Documentation

◆ QLstmLayer()

QLstmLayer ( const QLstmDescriptor param,
const char *  name 
)
protected

Constructor to create a QLstmLayer.

Parameters
[in]nameOptional name for the layer.

Definition at line 17 of file QLstmLayer.cpp.

18  : LayerWithParameters(3, 3, LayerType::QLstm, param, name)
19 {
20 }

References armnn::QLstm.

◆ ~QLstmLayer()

~QLstmLayer ( )
protecteddefault

Default destructor.

Member Function Documentation

◆ Clone()

QLstmLayer * 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 81 of file QLstmLayer.cpp.

82 {
83  auto layer = CloneBase<QLstmLayer>(graph, m_Param, GetName());
84 
85  layer->m_BasicParameters.m_InputToForgetWeights = m_BasicParameters.m_InputToForgetWeights ?
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  layer->m_PeepholeParameters.m_CellToInputWeights = m_PeepholeParameters.m_CellToInputWeights ?
127  }
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  if (!m_Param.m_CifgEnabled) {
138  layer->m_LayerNormParameters.m_InputLayerNormWeights = m_LayerNormParameters.m_InputLayerNormWeights ?
140  }
141 
142  layer->m_LayerNormParameters.m_ForgetLayerNormWeights = m_LayerNormParameters.m_ForgetLayerNormWeights ?
144  layer->m_LayerNormParameters.m_CellLayerNormWeights = m_LayerNormParameters.m_CellLayerNormWeights ?
146  layer->m_LayerNormParameters.m_OutputLayerNormWeights = m_LayerNormParameters.m_OutputLayerNormWeights ?
148  }
149 
150  return std::move(layer);
151 }

References Layer::GetName(), QLstmLayer::m_BasicParameters, QLstmBasicParameters::m_CellBias, QLstmOptLayerNormParameters::m_CellLayerNormWeights, QLstmOptPeepholeParameters::m_CellToForgetWeights, QLstmOptPeepholeParameters::m_CellToInputWeights, QLstmOptPeepholeParameters::m_CellToOutputWeights, QLstmDescriptor::m_CifgEnabled, QLstmLayer::m_CifgParameters, QLstmBasicParameters::m_ForgetGateBias, QLstmOptLayerNormParameters::m_ForgetLayerNormWeights, QLstmOptCifgParameters::m_InputGateBias, QLstmOptLayerNormParameters::m_InputLayerNormWeights, QLstmBasicParameters::m_InputToCellWeights, QLstmBasicParameters::m_InputToForgetWeights, QLstmOptCifgParameters::m_InputToInputWeights, QLstmBasicParameters::m_InputToOutputWeights, QLstmDescriptor::m_LayerNormEnabled, QLstmLayer::m_LayerNormParameters, QLstmBasicParameters::m_OutputGateBias, QLstmOptLayerNormParameters::m_OutputLayerNormWeights, LayerWithParameters< QLstmDescriptor >::m_Param, QLstmDescriptor::m_PeepholeEnabled, QLstmLayer::m_PeepholeParameters, QLstmOptProjectionParameters::m_ProjectionBias, QLstmDescriptor::m_ProjectionEnabled, QLstmLayer::m_ProjectionParameters, QLstmOptProjectionParameters::m_ProjectionWeights, QLstmBasicParameters::m_RecurrentToCellWeights, QLstmBasicParameters::m_RecurrentToForgetWeights, QLstmOptCifgParameters::m_RecurrentToInputWeights, and QLstmBasicParameters::m_RecurrentToOutputWeights.

◆ CreateWorkload()

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

Makes a workload for the QLstm 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 QLstmLayer.cpp.

23 {
24  QLstmQueueDescriptor 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 
60  descriptor.m_CellToForgetWeights = m_PeepholeParameters.m_CellToForgetWeights.get();
61  descriptor.m_CellToOutputWeights = m_PeepholeParameters.m_CellToOutputWeights.get();
62  }
63 
64  // Layer normalisation parameters
66  {
68  {
69  descriptor.m_InputLayerNormWeights = m_LayerNormParameters.m_InputLayerNormWeights.get();
70  }
71  descriptor.m_ForgetLayerNormWeights = m_LayerNormParameters.m_ForgetLayerNormWeights.get();
72  descriptor.m_CellLayerNormWeights = m_LayerNormParameters.m_CellLayerNormWeights.get();
73  descriptor.m_OutputLayerNormWeights = m_LayerNormParameters.m_OutputLayerNormWeights.get();
74  }
75 
76  SetAdditionalInfo(descriptor);
77 
78  return factory.CreateWorkload(LayerType::QLstm, descriptor, PrepInfoAndDesc(descriptor));
79 }

References IWorkloadFactory::CreateWorkload(), QLstmLayer::m_BasicParameters, QLstmBasicParameters::m_CellBias, QLstmQueueDescriptor::m_CellBias, QLstmOptLayerNormParameters::m_CellLayerNormWeights, QLstmQueueDescriptor::m_CellLayerNormWeights, QLstmOptPeepholeParameters::m_CellToForgetWeights, QLstmQueueDescriptor::m_CellToForgetWeights, QLstmOptPeepholeParameters::m_CellToInputWeights, QLstmQueueDescriptor::m_CellToInputWeights, QLstmOptPeepholeParameters::m_CellToOutputWeights, QLstmQueueDescriptor::m_CellToOutputWeights, QLstmDescriptor::m_CifgEnabled, QLstmLayer::m_CifgParameters, QLstmBasicParameters::m_ForgetGateBias, QLstmQueueDescriptor::m_ForgetGateBias, QLstmOptLayerNormParameters::m_ForgetLayerNormWeights, QLstmQueueDescriptor::m_ForgetLayerNormWeights, QLstmOptCifgParameters::m_InputGateBias, QLstmQueueDescriptor::m_InputGateBias, QLstmOptLayerNormParameters::m_InputLayerNormWeights, QLstmQueueDescriptor::m_InputLayerNormWeights, QLstmBasicParameters::m_InputToCellWeights, QLstmQueueDescriptor::m_InputToCellWeights, QLstmBasicParameters::m_InputToForgetWeights, QLstmQueueDescriptor::m_InputToForgetWeights, QLstmOptCifgParameters::m_InputToInputWeights, QLstmQueueDescriptor::m_InputToInputWeights, QLstmBasicParameters::m_InputToOutputWeights, QLstmQueueDescriptor::m_InputToOutputWeights, QLstmDescriptor::m_LayerNormEnabled, QLstmLayer::m_LayerNormParameters, QLstmBasicParameters::m_OutputGateBias, QLstmQueueDescriptor::m_OutputGateBias, QLstmOptLayerNormParameters::m_OutputLayerNormWeights, QLstmQueueDescriptor::m_OutputLayerNormWeights, LayerWithParameters< QLstmDescriptor >::m_Param, QLstmDescriptor::m_PeepholeEnabled, QLstmLayer::m_PeepholeParameters, QLstmOptProjectionParameters::m_ProjectionBias, QLstmQueueDescriptor::m_ProjectionBias, QLstmDescriptor::m_ProjectionEnabled, QLstmLayer::m_ProjectionParameters, QLstmOptProjectionParameters::m_ProjectionWeights, QLstmQueueDescriptor::m_ProjectionWeights, QLstmBasicParameters::m_RecurrentToCellWeights, QLstmQueueDescriptor::m_RecurrentToCellWeights, QLstmBasicParameters::m_RecurrentToForgetWeights, QLstmQueueDescriptor::m_RecurrentToForgetWeights, QLstmOptCifgParameters::m_RecurrentToInputWeights, QLstmQueueDescriptor::m_RecurrentToInputWeights, QLstmBasicParameters::m_RecurrentToOutputWeights, QLstmQueueDescriptor::m_RecurrentToOutputWeights, LayerWithParameters< QLstmDescriptor >::PrepInfoAndDesc(), armnn::QLstm, and Layer::SetAdditionalInfo().

◆ ExecuteStrategy()

void ExecuteStrategy ( IStrategy strategy) const
overridevirtual

Apply a visitor to this layer.

Reimplemented from Layer.

Definition at line 307 of file QLstmLayer.cpp.

308 {
309  std::vector<ConstTensor> constTensors;
310  ManagedConstTensorHandle managedInputToForgetWeights(m_BasicParameters.m_InputToForgetWeights);
311  ManagedConstTensorHandle managedInputToCellWeights(m_BasicParameters.m_InputToCellWeights);
312  ManagedConstTensorHandle managedInputToOutputWeights(m_BasicParameters.m_InputToOutputWeights);
313  ManagedConstTensorHandle managedRecurrentToForgetWeights(m_BasicParameters.m_RecurrentToForgetWeights);
314  ManagedConstTensorHandle managedRecurrentToCellWeights(m_BasicParameters.m_RecurrentToCellWeights);
315  ManagedConstTensorHandle managedRecurrentToOutputWeights(m_BasicParameters.m_RecurrentToOutputWeights);
316  ManagedConstTensorHandle managedForgetGateBias(m_BasicParameters.m_ForgetGateBias);
317  ManagedConstTensorHandle managedCellBias(m_BasicParameters.m_CellBias);
318  ManagedConstTensorHandle managedOutputGateBias(m_BasicParameters.m_OutputGateBias);
319 
320  // Cifg parameters
321  ManagedConstTensorHandle managedInputToInputWeights(m_CifgParameters.m_InputToInputWeights);
322  ManagedConstTensorHandle managedRecurrentToInputWeights(m_CifgParameters.m_RecurrentToInputWeights);
323  ManagedConstTensorHandle managedInputGateBias(m_CifgParameters.m_InputGateBias);
324 
325  // Projection parameters
326  ManagedConstTensorHandle managedProjectionWeights(m_ProjectionParameters.m_ProjectionWeights);
327  ManagedConstTensorHandle managedProjectionBias(m_ProjectionParameters.m_ProjectionBias);
328 
329  // Peephole parameters
330  ManagedConstTensorHandle managedCellToInputWeights(m_PeepholeParameters.m_CellToInputWeights);
331  ManagedConstTensorHandle managedCellToForgetWeights(m_PeepholeParameters.m_CellToForgetWeights);
332  ManagedConstTensorHandle managedCellToOutputWeights(m_PeepholeParameters.m_CellToOutputWeights);
333 
334  // Layer normalisation parameters
335  ManagedConstTensorHandle managedInputLayerNormWeights(m_LayerNormParameters.m_InputLayerNormWeights);
336  ManagedConstTensorHandle managedForgetLayerNormWeights(m_LayerNormParameters.m_ForgetLayerNormWeights);
337  ManagedConstTensorHandle managedCellLayerNormWeights(m_LayerNormParameters.m_CellLayerNormWeights);
338  ManagedConstTensorHandle managedOutputLayerNormWeights(m_LayerNormParameters.m_OutputLayerNormWeights);
339 
340  // First add mandatory/basic parameters
342  {
343  constTensors.emplace_back(ConstTensor(managedInputToForgetWeights.GetTensorInfo(),
344  managedInputToForgetWeights.Map()));
345  }
347  {
348  constTensors.emplace_back(ConstTensor(managedInputToCellWeights.GetTensorInfo(),
349  managedInputToCellWeights.Map()));
350  }
352  {
353  constTensors.emplace_back(ConstTensor(managedInputToOutputWeights.GetTensorInfo(),
354  managedInputToOutputWeights.Map()));
355  }
357  {
358  constTensors.emplace_back(ConstTensor(
359  managedRecurrentToForgetWeights.GetTensorInfo(),
360  managedRecurrentToForgetWeights.Map()));
361  }
363  {
364  constTensors.emplace_back(ConstTensor(
365  managedRecurrentToCellWeights.GetTensorInfo(),
366  managedRecurrentToCellWeights.Map()));
367  }
369  {
370  constTensors.emplace_back(ConstTensor(
371  managedRecurrentToOutputWeights.GetTensorInfo(),
372  managedRecurrentToOutputWeights.Map()));
373  }
374  if (m_BasicParameters.m_ForgetGateBias != nullptr)
375  {
376  constTensors.emplace_back(ConstTensor(managedForgetGateBias.GetTensorInfo(),
377  managedForgetGateBias.Map()));
378  }
379  if (m_BasicParameters.m_CellBias != nullptr)
380  {
381  constTensors.emplace_back(ConstTensor(managedCellBias.GetTensorInfo(),
382  managedCellBias.Map()));
383  }
384  if (m_BasicParameters.m_OutputGateBias != nullptr)
385  {
386  constTensors.emplace_back(ConstTensor(managedOutputGateBias.GetTensorInfo(),
387  managedOutputGateBias.Map()));
388  }
389 
390  // Add cifig parameters
392  {
393  constTensors.emplace_back(ConstTensor(managedInputToInputWeights.GetTensorInfo(),
394  managedInputToInputWeights.Map()));
395  }
397  {
398  constTensors.emplace_back(ConstTensor(
399  managedRecurrentToInputWeights.GetTensorInfo(),
400  managedRecurrentToInputWeights.Map()));
401  }
402  if (m_CifgParameters.m_InputGateBias != nullptr)
403  {
404  constTensors.emplace_back(ConstTensor(managedInputGateBias.GetTensorInfo(),
405  managedInputGateBias.Map()));
406  }
407 
408  // Add peephole parameters
410  {
411  constTensors.emplace_back(ConstTensor(managedCellToInputWeights.GetTensorInfo(),
412  managedCellToInputWeights.Map()));
413  }
415  {
416  constTensors.emplace_back(ConstTensor(managedCellToForgetWeights.GetTensorInfo(),
417  managedCellToForgetWeights.Map()));
418  }
420  {
421  constTensors.emplace_back(ConstTensor(managedCellToOutputWeights.GetTensorInfo(),
422  managedCellToOutputWeights.Map()));
423  }
424 
425  // Add projection parameters
427  {
428  constTensors.emplace_back(ConstTensor(managedProjectionWeights.GetTensorInfo(),
429  managedProjectionWeights.Map()));
430  }
432  {
433  constTensors.emplace_back(ConstTensor(managedProjectionBias.GetTensorInfo(),
434  managedProjectionBias.Map()));
435  }
436 
437  // Add norm parameters
439  {
440  constTensors.emplace_back(ConstTensor(managedInputLayerNormWeights.GetTensorInfo(),
441  managedInputLayerNormWeights.Map()));
442  }
444  {
445  constTensors.emplace_back(ConstTensor(managedForgetLayerNormWeights.GetTensorInfo(),
446  managedForgetLayerNormWeights.Map()));
447  }
449  {
450  constTensors.emplace_back(ConstTensor(managedCellLayerNormWeights.GetTensorInfo(),
451  managedCellLayerNormWeights.Map()));
452  }
454  {
455  constTensors.emplace_back(ConstTensor(managedOutputLayerNormWeights.GetTensorInfo(),
456  managedOutputLayerNormWeights.Map()));
457  }
458  strategy.ExecuteStrategy(this, GetParameters(), constTensors, GetName());
459 }

References IStrategy::ExecuteStrategy(), Layer::GetName(), LayerWithParameters< QLstmDescriptor >::GetParameters(), ManagedConstTensorHandle::GetTensorInfo(), QLstmLayer::m_BasicParameters, QLstmBasicParameters::m_CellBias, QLstmOptLayerNormParameters::m_CellLayerNormWeights, QLstmOptPeepholeParameters::m_CellToForgetWeights, QLstmOptPeepholeParameters::m_CellToInputWeights, QLstmOptPeepholeParameters::m_CellToOutputWeights, QLstmLayer::m_CifgParameters, QLstmBasicParameters::m_ForgetGateBias, QLstmOptLayerNormParameters::m_ForgetLayerNormWeights, QLstmOptCifgParameters::m_InputGateBias, QLstmOptLayerNormParameters::m_InputLayerNormWeights, QLstmBasicParameters::m_InputToCellWeights, QLstmBasicParameters::m_InputToForgetWeights, QLstmOptCifgParameters::m_InputToInputWeights, QLstmBasicParameters::m_InputToOutputWeights, QLstmLayer::m_LayerNormParameters, QLstmBasicParameters::m_OutputGateBias, QLstmOptLayerNormParameters::m_OutputLayerNormWeights, QLstmLayer::m_PeepholeParameters, QLstmOptProjectionParameters::m_ProjectionBias, QLstmLayer::m_ProjectionParameters, QLstmOptProjectionParameters::m_ProjectionWeights, QLstmBasicParameters::m_RecurrentToCellWeights, QLstmBasicParameters::m_RecurrentToForgetWeights, QLstmOptCifgParameters::m_RecurrentToInputWeights, QLstmBasicParameters::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 272 of file QLstmLayer.cpp.

References QLstmLayer::m_BasicParameters, QLstmBasicParameters::m_CellBias, QLstmOptLayerNormParameters::m_CellLayerNormWeights, QLstmOptPeepholeParameters::m_CellToForgetWeights, QLstmOptPeepholeParameters::m_CellToInputWeights, QLstmOptPeepholeParameters::m_CellToOutputWeights, QLstmLayer::m_CifgParameters, QLstmBasicParameters::m_ForgetGateBias, QLstmOptLayerNormParameters::m_ForgetLayerNormWeights, QLstmOptCifgParameters::m_InputGateBias, QLstmOptLayerNormParameters::m_InputLayerNormWeights, QLstmBasicParameters::m_InputToCellWeights, QLstmBasicParameters::m_InputToForgetWeights, QLstmOptCifgParameters::m_InputToInputWeights, QLstmBasicParameters::m_InputToOutputWeights, QLstmLayer::m_LayerNormParameters, QLstmBasicParameters::m_OutputGateBias, QLstmOptLayerNormParameters::m_OutputLayerNormWeights, QLstmLayer::m_PeepholeParameters, QLstmOptProjectionParameters::m_ProjectionBias, QLstmLayer::m_ProjectionParameters, QLstmOptProjectionParameters::m_ProjectionWeights, QLstmBasicParameters::m_RecurrentToCellWeights, QLstmBasicParameters::m_RecurrentToForgetWeights, QLstmOptCifgParameters::m_RecurrentToInputWeights, and QLstmBasicParameters::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 153 of file QLstmLayer.cpp.

154 {
155  ARMNN_ASSERT(inputShapes.size() == 3);
156 
157  // Get input values for validation
158  unsigned int batchSize = inputShapes[0][0];
159  unsigned int outputSize = inputShapes[1][1];
160  unsigned int numUnits = inputShapes[2][1];
161 
162  std::vector<TensorShape> outShapes;
163  outShapes.push_back(TensorShape({ batchSize, outputSize })); // outputStateOut
164  outShapes.push_back(TensorShape({ batchSize, numUnits })); // cellStateOut
165  outShapes.push_back(TensorShape({ batchSize, outputSize })); // output
166 
167  return outShapes;
168 }

References ARMNN_ASSERT.

Referenced by QLstmLayer::ValidateTensorShapesFromInputs().

◆ ValidateTensorShapesFromInputs()

void ValidateTensorShapesFromInputs ( )
overridevirtual

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

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

Implements Layer.

Definition at line 170 of file QLstmLayer.cpp.

171 {
173 
174  const TensorShape& outputShape = GetOutputSlot(0).GetTensorInfo().GetShape();
175 
177 
178  auto inferredShapes = InferOutputShapes(
179  {
180  GetInputSlot(0).GetTensorInfo().GetShape(), // input
181  GetInputSlot(1).GetTensorInfo().GetShape(), // previousOutputIn
182  GetInputSlot(2).GetTensorInfo().GetShape() // previousCellStateIn
183  });
184 
185  ARMNN_ASSERT(inferredShapes.size() == 3);
186 
187  // Check if the weights are nullptr for basic params
189  "QLstmLayer: m_BasicParameters.m_InputToForgetWeights should not be null.");
191  "QLstmLayer: m_BasicParameters.m_InputToCellWeights should not be null.");
193  "QLstmLayer: m_BasicParameters.m_InputToOutputWeights should not be null.");
195  "QLstmLayer: m_BasicParameters.m_RecurrentToForgetWeights should not be null.");
197  "QLstmLayer: m_BasicParameters.m_RecurrentToCellWeights should not be null.");
199  "QLstmLayer: m_BasicParameters.m_RecurrentToOutputWeights should not be null.");
201  "QLstmLayer: m_BasicParameters.m_ForgetGateBias should not be null.");
203  "QLstmLayer: m_BasicParameters.m_CellBias should not be null.");
205  "QLstmLayer: m_BasicParameters.m_OutputGateBias should not be null.");
206 
207  if (!m_Param.m_CifgEnabled)
208  {
210  "QLstmLayer: m_CifgParameters.m_InputToInputWeights should not be null.");
212  "QLstmLayer: m_CifgParameters.m_RecurrentToInputWeights should not be null.");
214  "QLstmLayer: m_CifgParameters.m_InputGateBias should not be null.");
215 
216  ValidateAndCopyShape(outputShape, inferredShapes[0], m_ShapeInferenceMethod, "QLstmLayer");
217  }
218  else
219  {
221  "QLstmLayer: m_CifgParameters.m_InputToInputWeights should not have a value when CIFG is enabled.");
223  "QLstmLayer: m_CifgParameters.m_RecurrentToInputWeights should "
224  "not have a value when CIFG is enabled.");
226  "QLstmLayer: m_CifgParameters.m_InputGateBias should not have a value when CIFG is enabled.");
227 
228  ValidateAndCopyShape(outputShape, inferredShapes[0], m_ShapeInferenceMethod, "QLstmLayer");
229  }
230 
232  {
234  "QLstmLayer: m_ProjectionParameters.m_ProjectionWeights should not be null.");
235  }
236 
238  {
239  if (!m_Param.m_CifgEnabled) {
241  "QLstmLayer: m_PeepholeParameters.m_CellToInputWeights should not be null "
242  "when Peephole is enabled and CIFG is disabled.");
243  }
244 
246  "QLstmLayer: m_PeepholeParameters.m_CellToForgetWeights should not be null.");
248  "QLstmLayer: m_PeepholeParameters.m_CellToOutputWeights should not be null.");
249  }
250 
252  GetOutputSlot(1).GetTensorInfo().GetShape(), inferredShapes[1], m_ShapeInferenceMethod, "QLstmLayer", 1);
254  GetOutputSlot(2).GetTensorInfo().GetShape(), inferredShapes[2], m_ShapeInferenceMethod, "QLstmLayer", 2);
255 
257  {
259  {
261  "QLstmLayer: m_LayerNormParameters.m_InputLayerNormWeights should not be null.");
262  }
264  "QLstmLayer: m_LayerNormParameters.m_ForgetLayerNormWeights should not be null.");
266  "QLstmLayer: m_LayerNormParameters.m_CellLayerNormWeights should not be null.");
268  "QLstmLayer: m_LayerNormParameters.m_UutputLayerNormWeights should not be null.");
269  }
270 }

References ARMNN_ASSERT, ARMNN_ASSERT_MSG, CHECK_LOCATION, Layer::GetInputSlot(), Layer::GetOutputSlot(), TensorInfo::GetShape(), armnn::GetTensorInfo(), InputSlot::GetTensorInfo(), OutputSlot::GetTensorInfo(), QLstmLayer::InferOutputShapes(), QLstmLayer::m_BasicParameters, QLstmBasicParameters::m_CellBias, QLstmOptPeepholeParameters::m_CellToForgetWeights, QLstmOptPeepholeParameters::m_CellToInputWeights, QLstmOptPeepholeParameters::m_CellToOutputWeights, QLstmDescriptor::m_CifgEnabled, QLstmLayer::m_CifgParameters, QLstmBasicParameters::m_ForgetGateBias, QLstmOptCifgParameters::m_InputGateBias, QLstmBasicParameters::m_InputToCellWeights, QLstmBasicParameters::m_InputToForgetWeights, QLstmOptCifgParameters::m_InputToInputWeights, QLstmBasicParameters::m_InputToOutputWeights, QLstmBasicParameters::m_OutputGateBias, LayerWithParameters< QLstmDescriptor >::m_Param, QLstmDescriptor::m_PeepholeEnabled, QLstmLayer::m_PeepholeParameters, QLstmDescriptor::m_ProjectionEnabled, QLstmLayer::m_ProjectionParameters, QLstmOptProjectionParameters::m_ProjectionWeights, QLstmBasicParameters::m_RecurrentToCellWeights, QLstmBasicParameters::m_RecurrentToForgetWeights, QLstmOptCifgParameters::m_RecurrentToInputWeights, QLstmBasicParameters::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_ASSERT
#define ARMNN_ASSERT(COND)
Definition: Assert.hpp:14
armnn::QLstmLayer::m_CifgParameters
QLstmOptCifgParameters m_CifgParameters
Definition: QLstmLayer.hpp:84
armnn::QLstmOptPeepholeParameters::m_CellToForgetWeights
std::shared_ptr< ConstTensorHandle > m_CellToForgetWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:51
armnn::QLstmBasicParameters::m_ForgetGateBias
std::shared_ptr< ConstTensorHandle > m_ForgetGateBias
A unique pointer to represent 1D bias tensor with dimensions [num_units] (int32).
Definition: QLstmLayer.hpp:31
armnn::QLstmDescriptor::m_ProjectionEnabled
bool m_ProjectionEnabled
Enable/disable the projection layer.
Definition: Descriptors.hpp:1422
armnn::OutputSlot::GetTensorInfo
const TensorInfo & GetTensorInfo() const override
Definition: Layer.cpp:92
armnn::QLstmOptLayerNormParameters::m_ForgetLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_ForgetLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:71
CHECK_LOCATION
#define CHECK_LOCATION()
Definition: Exceptions.hpp:203
armnn::QLstmOptPeepholeParameters::m_CellToOutputWeights
std::shared_ptr< ConstTensorHandle > m_CellToOutputWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:53
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:435
armnn::QLstmOptLayerNormParameters::m_OutputLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_OutputLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:75
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_ASSERT_MSG
#define ARMNN_ASSERT_MSG(COND, MSG)
Definition: Assert.hpp:15
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< QLstmDescriptor >::GetParameters
const QLstmDescriptor & GetParameters() const override
Definition: LayerWithParameters.hpp:19
armnn::QLstmOptProjectionParameters::m_ProjectionWeights
std::shared_ptr< ConstTensorHandle > m_ProjectionWeights
A unique pointer to represent 2D weights tensor with dimensions [output_size, num_units] (QSymmS8).
Definition: QLstmLayer.hpp:41
armnn::Layer::GetName
const char * GetName() const override
Returns the name of the layer.
Definition: Layer.hpp:332
armnn::QLstmOptLayerNormParameters::m_CellLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_CellLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:73
armnn::InputSlot::GetTensorInfo
const TensorInfo & GetTensorInfo() const override
Gets the TensorInfo for this InputSlot.
Definition: Layer.cpp:592
armnn::LayerWithParameters< QLstmDescriptor >::m_Param
QLstmDescriptor m_Param
The parameters for the layer (not including tensor-valued weights etc.).
Definition: LayerWithParameters.hpp:52
armnn::QLstmOptPeepholeParameters::m_CellToInputWeights
std::shared_ptr< ConstTensorHandle > m_CellToInputWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:49
armnn::LayerWithParameters< QLstmDescriptor >::PrepInfoAndDesc
WorkloadInfo PrepInfoAndDesc(QueueDescriptor &descriptor) const
Helper function to reduce duplication in *Layer::CreateWorkload.
Definition: LayerWithParameters.hpp:44
armnn::QLstmOptCifgParameters::m_RecurrentToInputWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToInputWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units] (QSymmS8).
Definition: QLstmLayer.hpp:61
armnn::Layer::VerifyShapeInferenceType
void VerifyShapeInferenceType(const TensorShape &outputShape, ShapeInferenceMethod shapeInferenceMethod)
Definition: Layer.cpp:504
armnn::GetTensorInfo
const TensorInfo & GetTensorInfo(const ITensorHandle *tensorHandle)
float32 helpers
Definition: RefWorkloadUtils.hpp:33
armnn::Layer::SetAdditionalInfo
void SetAdditionalInfo(QueueDescriptor &descriptor) const
Definition: Layer.cpp:287
armnn::QLstmOptCifgParameters::m_InputGateBias
std::shared_ptr< ConstTensorHandle > m_InputGateBias
A unique pointer to represent 1D weights tensor with dimensions [num_units] (int32).
Definition: QLstmLayer.hpp:63
armnn::QLstmOptLayerNormParameters::m_InputLayerNormWeights
std::shared_ptr< ConstTensorHandle > m_InputLayerNormWeights
A unique pointer to represent 1D weights tensor with dimensions [num_units] (QSymmS16).
Definition: QLstmLayer.hpp:69
armnn::QLstmLayer::m_LayerNormParameters
QLstmOptLayerNormParameters m_LayerNormParameters
Definition: QLstmLayer.hpp:87
armnn::QLstmBasicParameters::m_RecurrentToOutputWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToOutputWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, outputSize] (QSymmS8).
Definition: QLstmLayer.hpp:28
armnn::QLstmLayer::m_PeepholeParameters
QLstmOptPeepholeParameters m_PeepholeParameters
Definition: QLstmLayer.hpp:86
armnn::TensorInfo::GetShape
const TensorShape & GetShape() const
Definition: Tensor.hpp:193
armnn::QLstmBasicParameters::m_CellBias
std::shared_ptr< ConstTensorHandle > m_CellBias
A unique pointer to represent 1D bias tensor with dimensions [num_units] (int32).
Definition: QLstmLayer.hpp:33
armnn::QLstmBasicParameters::m_OutputGateBias
std::shared_ptr< ConstTensorHandle > m_OutputGateBias
A unique pointer to represent 1D bias tensor with dimensions [num_units] (int32).
Definition: QLstmLayer.hpp:35
armnn::QLstmBasicParameters::m_RecurrentToCellWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToCellWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, outputSize] (QSymmS8).
Definition: QLstmLayer.hpp:26
armnn::QLstmDescriptor::m_CifgEnabled
bool m_CifgEnabled
Enable/disable CIFG (coupled input & forget gate).
Definition: Descriptors.hpp:1418
armnn::QLstmOptCifgParameters::m_InputToInputWeights
std::shared_ptr< ConstTensorHandle > m_InputToInputWeights
A unique pointer to represent 2D weights tensor with dimensions [input_size, num_units] (QSymmS8).
Definition: QLstmLayer.hpp:59
armnn::QLstmLayer::m_ProjectionParameters
QLstmOptProjectionParameters m_ProjectionParameters
Definition: QLstmLayer.hpp:85
armnn::QLstmLayer::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: QLstmLayer.cpp:153
armnn::Layer::VerifyLayerConnections
void VerifyLayerConnections(unsigned int expectedConnections, const CheckLocation &location) const
Definition: Layer.cpp:391
armnn::QLstmBasicParameters::m_RecurrentToForgetWeights
std::shared_ptr< ConstTensorHandle > m_RecurrentToForgetWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, outputSize] (QSymmS8).
Definition: QLstmLayer.hpp:24
armnn::QLstmDescriptor::m_LayerNormEnabled
bool m_LayerNormEnabled
Enable/disable layer normalization.
Definition: Descriptors.hpp:1424
armnn::QLstmOptProjectionParameters::m_ProjectionBias
std::shared_ptr< ConstTensorHandle > m_ProjectionBias
A unique pointer to represent 1D weights tensor with dimensions [output_size] (int32).
Definition: QLstmLayer.hpp:43
armnn::LayerWithParameters< QLstmDescriptor >::LayerWithParameters
LayerWithParameters(unsigned int numInputSlots, unsigned int numOutputSlots, LayerType type, const QLstmDescriptor &param, const char *name)
Definition: LayerWithParameters.hpp:30
armnn::QLstmLayer::m_BasicParameters
QLstmBasicParameters m_BasicParameters
Definition: QLstmLayer.hpp:83
armnn::Layer::m_ShapeInferenceMethod
ShapeInferenceMethod m_ShapeInferenceMethod
Definition: Layer.hpp:441
armnn::QLstmBasicParameters::m_InputToOutputWeights
std::shared_ptr< ConstTensorHandle > m_InputToOutputWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, inputSize] (QSymmS8).
Definition: QLstmLayer.hpp:21
armnn::QLstmDescriptor::m_PeepholeEnabled
bool m_PeepholeEnabled
Enable/disable peephole.
Definition: Descriptors.hpp:1420
armnn::LayerType::QLstm
@ QLstm
armnn::QLstmBasicParameters::m_InputToForgetWeights
std::shared_ptr< ConstTensorHandle > m_InputToForgetWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, inputSize] (QSymmS8).
Definition: QLstmLayer.hpp:17
armnn::QLstmBasicParameters::m_InputToCellWeights
std::shared_ptr< ConstTensorHandle > m_InputToCellWeights
A unique pointer to represent 2D weights tensor with dimensions [num_units, inputSize] (QSymmS8).
Definition: QLstmLayer.hpp:19