ONE - On-device Neural Engine
Loading...
Searching...
No Matches
onert::backend::builtin::kernel::CallLayer Class Reference

#include <CallLayer.h>

Collaboration diagram for onert::backend::builtin::kernel::CallLayer:

Public Member Functions

 CallLayer (const std::vector< backend::IPortableTensor * > input_tensors, const std::vector< backend::IPortableTensor * > output_tensors, const ir::SubgraphIndex &callee_subg_index, exec::IExecutors *executors, const ir::ModelIndex &model_index, const std::shared_ptr< ExternalContext > &external_context)
 
void run () override
 
- Public Member Functions inherited from onert::exec::IFunction
virtual ~IFunction ()=default
 
virtual void prepare ()
 

Detailed Description

Definition at line 28 of file CallLayer.h.

Constructor & Destructor Documentation

◆ CallLayer()

onert::backend::builtin::kernel::CallLayer::CallLayer ( const std::vector< backend::IPortableTensor * >  input_tensors,
const std::vector< backend::IPortableTensor * >  output_tensors,
const ir::SubgraphIndex callee_subg_index,
exec::IExecutors executors,
const ir::ModelIndex model_index,
const std::shared_ptr< ExternalContext > &  external_context 
)

Definition at line 22 of file CallLayer.cc.

27 : _input_tensors{input_tensors}, _output_tensors{output_tensors},
28 _callee_subg_index{callee_subg_index}, _executors{executors}, _model_index{model_index},
29 _external_context{external_context}
30{
31 // DO NOTHING
32}

Member Function Documentation

◆ run()

void onert::backend::builtin::kernel::CallLayer::run ( )
overridevirtual

Implements onert::exec::IFunction.

Definition at line 34 of file CallLayer.cc.

35{
36 exec::IExecutor *subg_exec = _executors->at(_model_index, _callee_subg_index);
37 subg_exec->execute(_input_tensors, _output_tensors,
38 _executors->entryExecutor()->currentOptions());
39}
virtual IExecutor * entryExecutor() const
Definition IExecutors.h:58
virtual IExecutor * at(const ir::ModelIndex &model_index, const ir::SubgraphIndex &subg_index) const =0
Return executor of index.
virtual void execute(const std::vector< backend::IPortableTensor * > &inputs, const std::vector< backend::IPortableTensor * > &outputs, const ExecutionOptions &options)=0
Execute with given input/output tensors.
virtual const ExecutionOptions & currentOptions() const =0
Return current execution configuration.

References onert::exec::IExecutors::at(), onert::exec::IExecutor::currentOptions(), onert::exec::IExecutors::entryExecutor(), and onert::exec::IExecutor::execute().


The documentation for this class was generated from the following files: