Poplar and PopLibs
popsparse::dynamic::Partitioner< T > Class Template Reference

Class to translate and encode sparsity information for a fully connected layer. More...

#include <SparsePartitioner.hpp>

Public Member Functions

SparsityDataImpl< T > createSparsityDataImpl (const CSCMatrix< T > &matrix_) const
 Create implementation sparsity representation for a compressed sparse columns (CSC) matrix.
 
SparsityDataImpl< T > createSparsityDataImpl (const CSRMatrix< T > &matrix_) const
 Creates implementation sparsity representation for a compressed sparse rows (CSR) matrix.
 
SparsityDataImpl< T > createSparsityDataImpl (const COOMatrix< T > &matrix_) const
 Creates implementation sparsity representation for a coordinate (COO) format matrix.
 
COOMatrix< T > sparsityDataImplToCOOMatrix (const SparsityDataImpl< T > &sparsityDataImpl) const
 Create a coordinate (COO) representation matrix from implementation sparsity representation. More...
 
CSRMatrix< T > sparsityDataImplToCSRMatrix (const SparsityDataImpl< T > &sparsityDataImpl) const
 Create compressed sparse rows (CSR) representation from implementation sparsity representation.
 
CSCMatrix< T > sparsityDataImplToCSCMatrix (const SparsityDataImpl< T > &sparsityDataImpl) const
 Create compressed sparse columns (CSC) representation from implementation sparsity representation.
 
std::array< std::vector< std::size_t >, 3 > getPlanPartitions (void) const
 Fetch the partitions in X, Y and Z to reveal the plan.
 
std::size_t getmetaInfoBucketElements (void) const
 Fetch the number of elements in a meta info bucket.
 

Detailed Description

template<typename T>
class popsparse::dynamic::Partitioner< T >

Class to translate and encode sparsity information for a fully connected layer.

See createFullyConnectedWeights() for details of the options.

Member Function Documentation

◆ sparsityDataImplToCOOMatrix()

template<typename T >
COOMatrix< T > popsparse::dynamic::Partitioner< T >::sparsityDataImplToCOOMatrix ( const SparsityDataImpl< T > &  sparsityDataImpl) const

Create a coordinate (COO) representation matrix from implementation sparsity representation.

The COO entries are ordered by row first, and then columns.


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