Package smile.data

Class SparseDataset<T>

java.lang.Object

smile.data.SimpleDataset<SparseArray,T>

smile.data.SparseDataset<T>

Type Parameters:

T - the target type.

All Implemented Interfaces:

Iterable<SampleInstance<SparseArray,T>>, Dataset<SparseArray,T>

public class SparseDataset<T>
extends SimpleDataset<SparseArray,T>

List of Lists sparse matrix format. LIL stores one list per row, where each entry stores a column index and value. Typically, these entries are kept sorted by column index for faster lookup. This format is good for incremental matrix construction.

LIL is typically used to construct the matrix. Once the matrix is constructed, it is typically converted to a format, such as Harwell-Boeing column-compressed sparse matrix format, which is more efficient for matrix operations.

Constructor Summary

Constructors

Constructor	Description
SparseDataset(Collection<SampleInstance<SparseArray,T>> data)	Constructor.
SparseDataset(Collection<SampleInstance<SparseArray,T>> data, int ncol)	Constructor.

Method Summary

Modifier and Type	Method	Description
static SparseDataset<Void>	from(Path path)	Parses spare dataset in coordinate triple tuple list format.
static SparseDataset<Void>	from(Path path, int arrayIndexOrigin)	Reads spare dataset in coordinate triple tuple list format.
double	get(int i, int j)	Returns the value at entry (i, j).
int	ncol()	Returns the number of columns.
int	nrow()	Returns the number of rows.
int	nz()	Returns the number of nonzero entries.
int	nz(int j)	Returns the number of nonzero entries in column j.
static SparseDataset<Void>	of(SparseArray[] data)	Returns a default implementation of SparseDataset without targets.
static SparseDataset<Void>	of(SparseArray[] data, int ncol)	Returns a default implementation of SparseDataset without targets.
SparseMatrix	toMatrix()	Convert into Harwell-Boeing column-compressed sparse matrix format.
void	unitize()	Unitize each row so that L2 norm of x = 1.
void	unitize1()	Unitize each row so that L1 norm of x is 1.

Methods inherited from class smile.data.SimpleDataset

get, iterator, size, stream, toList

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface smile.data.Dataset

apply, batch, isEmpty, toString

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Constructor Details

SparseDataset

public SparseDataset(Collection<SampleInstance<SparseArray,T>> data)

Constructor.

Parameters:

data - The sample instances.

SparseDataset

public SparseDataset(Collection<SampleInstance<SparseArray,T>> data,
 int ncol)

Constructor.

Parameters:

data - The sample instances.

ncol - The number of columns.

Method Details

nz

public int nz()

Returns the number of nonzero entries.

Returns:

the number of nonzero entries.

nz

public int nz(int j)

Returns the number of nonzero entries in column j.

Parameters:

j - the column index.

Returns:

the number of nonzero entries in column j.

nrow

public int nrow()

Returns the number of rows.

Returns:

the number of rows.

ncol

public int ncol()

Returns the number of columns.

Returns:

the number of columns.

get

public double get(int i,
 int j)

Returns the value at entry (i, j).

Parameters:

i - the row index.

j - the column index.

Returns:

the cell value.

unitize

public void unitize()

Unitize each row so that L2 norm of x = 1.

unitize1

public void unitize1()

Unitize each row so that L1 norm of x is 1.

toMatrix

public SparseMatrix toMatrix()

Convert into Harwell-Boeing column-compressed sparse matrix format.

Returns:

the sparse matrix.

of

public static SparseDataset<Void> of(SparseArray[] data)

Returns a default implementation of SparseDataset without targets.

Parameters:

data - sparse arrays.

Returns:

the sparse dataset.

of

public static SparseDataset<Void> of(SparseArray[] data,
 int ncol)

Returns a default implementation of SparseDataset without targets.

Parameters:

data - sparse arrays.

ncol - the number of columns.

Returns:

the sparse dataset.

from

public static SparseDataset<Void> from(Path path)
                                throws IOException,
ParseException

Parses spare dataset in coordinate triple tuple list format. Coordinate file stores a list of (row, column, value) tuples.

Parameters:

path - the input file path.

Returns:

the sparse dataset.

Throws:

IOException - when fails to read file.

ParseException - when fails to parse data.

from

public static SparseDataset<Void> from(Path path,
 int arrayIndexOrigin)
                                throws IOException,
ParseException

Reads spare dataset in coordinate triple tuple list format. Coordinate file stores a list of (row, column, value) tuples:

 instanceID attributeID value
 instanceID attributeID value
 instanceID attributeID value
 instanceID attributeID value
 ...
 instanceID attributeID value
 instanceID attributeID value
 instanceID attributeID value
 

Ideally, the entries are sorted (by row index, then column index) to improve random access times. This format is good for incremental matrix construction.

In addition, there may a header line

 D W N   // The number of rows, columns and nonzero entries.
 

or 3 header lines

 D    // The number of rows
 W    // The number of columns
 N    // The total number of nonzero entries in the dataset.
 

Parameters:

path - the input file path.

arrayIndexOrigin - the starting index of array. By default, it is 0 as in C/C++ and Java. But it could be 1 to parse data produced by other programming language such as Fortran.

Returns:

the sparse dataset.

Throws:

IOException - if stream to file cannot be read or closed.

ParseException - if an index is not an integer or the value is not a double.