tvm::meta_schedule::FeatureExtractorNode Class Reference

Extractor for features from measure candidates for use in cost model. More...

#include <feature_extractor.h>

Inheritance diagram for tvm::meta_schedule::FeatureExtractorNode:

Collaboration diagram for tvm::meta_schedule::FeatureExtractorNode:

Public Member Functions
virtual	~FeatureExtractorNode ()=default
	Virtual destructor. More...
virtual ffi::Array< tvm::runtime::Tensor >	ExtractFrom (const TuneContext &context, const ffi::Array< MeasureCandidate > &candidates)=0
	Extract features from the given measure candidate. More...
	TVM_FFI_DECLARE_OBJECT_INFO ("meta_schedule.FeatureExtractor", FeatureExtractorNode, Object)

Static Public Member Functions
static void	RegisterReflection ()

Detailed Description

Extractor for features from measure candidates for use in cost model.

Constructor & Destructor Documentation

~FeatureExtractorNode()

virtual tvm::meta_schedule::FeatureExtractorNode::~FeatureExtractorNode ( )

virtualdefault

Virtual destructor.

Member Function Documentation

ExtractFrom()

virtual ffi::Array<tvm::runtime::Tensor> tvm::meta_schedule::FeatureExtractorNode::ExtractFrom ( const TuneContext &  context, const ffi::Array< MeasureCandidate > &  candidates  )

pure virtual

Extract features from the given measure candidate.

Parameters

context	The tuning context for feature extraction.
candidates	The measure candidates to extract features from.

Returns

The feature tensor extracted.

Implemented in tvm::meta_schedule::PyFeatureExtractorNode.

RegisterReflection()

static void tvm::meta_schedule::FeatureExtractorNode::RegisterReflection ( )

inlinestatic

TVM_FFI_DECLARE_OBJECT_INFO()

tvm::meta_schedule::FeatureExtractorNode::TVM_FFI_DECLARE_OBJECT_INFO	(	"meta_schedule.FeatureExtractor"	,
		FeatureExtractorNode	,
		Object
	)

---

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

• include/tvm/meta_schedule/feature_extractor.h