tvm
Public Member Functions | Static Public Member Functions | List of all members
tvm::meta_schedule::FeatureExtractor Class Reference

Managed reference to FeatureExtractorNode. More...

#include <feature_extractor.h>

Inheritance diagram for tvm::meta_schedule::FeatureExtractor:
Collaboration diagram for tvm::meta_schedule::FeatureExtractor:

Public Member Functions

 TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS (FeatureExtractor, ObjectRef, FeatureExtractorNode)
 

Static Public Member Functions

static FeatureExtractor PerStoreFeature (int buffers_per_store=5, int arith_intensity_curve_num_samples=10, int cache_line_bytes=64, bool extract_workload=false)
 Create a feature extractor that extracts features from each BufferStore. More...
 
static FeatureExtractor PyFeatureExtractor (PyFeatureExtractorNode::FExtractFrom f_extract_from, PyFeatureExtractorNode::FAsString f_as_string)
 Create a feature extractor with customized methods on the python-side. More...
 

Detailed Description

Managed reference to FeatureExtractorNode.

See also
FeatureExtractorNode

Member Function Documentation

◆ PerStoreFeature()

static FeatureExtractor tvm::meta_schedule::FeatureExtractor::PerStoreFeature ( int  buffers_per_store = 5,
int  arith_intensity_curve_num_samples = 10,
int  cache_line_bytes = 64,
bool  extract_workload = false 
)
static

Create a feature extractor that extracts features from each BufferStore.

Parameters
buffers_per_storeThe number of buffers in each BufferStore; Pad or truncate if necessary.
arith_intensity_curve_num_samplesThe number of samples used in the arithmetic intensity curve.
cache_line_bytesThe number of bytes in a cache line.
extract_workloadWhether to extract features in the workload in tuning context or not.
Returns
The feature extractor created.

◆ PyFeatureExtractor()

static FeatureExtractor tvm::meta_schedule::FeatureExtractor::PyFeatureExtractor ( PyFeatureExtractorNode::FExtractFrom  f_extract_from,
PyFeatureExtractorNode::FAsString  f_as_string 
)
static

Create a feature extractor with customized methods on the python-side.

Parameters
f_extract_fromThe packed function of ExtractFrom.
f_as_stringThe packed function of AsString.
Returns
The feature extractor created.

◆ TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS()

tvm::meta_schedule::FeatureExtractor::TVM_DEFINE_MUTABLE_OBJECT_REF_METHODS ( FeatureExtractor  ,
ObjectRef  ,
FeatureExtractorNode   
)

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