|
Parallel Colt 0.7.2 | |||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Object cern.colt.PersistentObject cern.colt.matrix.tfloat.FloatFactory2D
public class FloatFactory2D
Factory for convenient construction of 2-d matrices holding float cells. Also provides convenient methods to compose (concatenate) and decompose (split) matrices from/to constituent blocks.
Construction | Use idioms like FloatFactory2D.dense.make(4,4) to construct dense matrices, FloatFactory2D.sparse.make(4,4) to construct sparse matrices. |
Construction with initial values | Use other make methods to construct matrices with given initial values. |
Appending rows and columns | Use methods appendColumns , appendRows and repeat to append rows
and columns. |
General block matrices | Use methods compose and
decompose to work with
general block matrices. |
Diagonal matrices | Use methods diagonal(vector) ,
diagonal(matrix) and identity to work with diagonal matrices. |
Diagonal block matrices | Use method
composeDiagonal to work with diagonal block matrices. |
Random | Use methods random and
sample to construct random matrices. |
If the factory is used frequently it might be useful to streamline the notation. For example by aliasing:
FloatFactory2D F = FloatFactory2D.dense; F.make(4,4); F.descending(10,20); F.random(4,4); ... |
Field Summary | |
---|---|
static FloatFactory2D |
dense
A factory producing dense matrices. |
static FloatFactory2D |
rowCompressed
A factory producing sparse row compressed matrices. |
static FloatFactory2D |
sparse
A factory producing sparse hash matrices. |
Fields inherited from class cern.colt.PersistentObject |
---|
serialVersionUID |
Method Summary | |
---|---|
FloatMatrix2D |
appendColumn(FloatMatrix2D A,
FloatMatrix1D b)
|
FloatMatrix2D |
appendColumns(FloatMatrix2D A,
FloatMatrix2D B)
C = A||B; Constructs a new matrix which is the column-wise concatenation of two other matrices. |
FloatMatrix2D |
appendRow(FloatMatrix2D A,
FloatMatrix1D b)
|
FloatMatrix2D |
appendRows(FloatMatrix2D A,
FloatMatrix2D B)
C = A||B; Constructs a new matrix which is the row-wise concatenation of two other matrices. |
FloatMatrix2D |
ascending(int rows,
int columns)
Constructs a matrix with cells having ascending values. |
FloatMatrix2D |
compose(FloatMatrix2D[][] parts)
Constructs a block matrix made from the given parts. |
FloatMatrix2D |
composeBidiagonal(FloatMatrix2D A,
FloatMatrix2D B)
|
FloatMatrix2D |
composeDiagonal(FloatMatrix2D A,
FloatMatrix2D B)
Constructs a diagonal block matrix from the given parts (the direct sum of two matrices). |
FloatMatrix2D |
composeDiagonal(FloatMatrix2D A,
FloatMatrix2D B,
FloatMatrix2D C)
Constructs a diagonal block matrix from the given parts. |
void |
decompose(FloatMatrix2D[][] parts,
FloatMatrix2D matrix)
Splits a block matrix into its constituent blocks; Copies blocks of a matrix into the given parts. |
void |
demo1()
Demonstrates usage of this class. |
void |
demo2()
Demonstrates usage of this class. |
FloatMatrix2D |
descending(int rows,
int columns)
Constructs a matrix with cells having descending values. |
FloatMatrix2D |
diagonal(float[] vector)
Constructs a new diagonal matrix whose diagonal elements are the elements of vector. |
FloatMatrix2D |
diagonal(FloatMatrix1D vector)
Constructs a new diagonal matrix whose diagonal elements are the elements of vector. |
FloatMatrix1D |
diagonal(FloatMatrix2D A)
Constructs a new vector consisting of the diagonal elements of A . |
FloatMatrix2D |
identity(int rowsAndColumns)
Constructs an identity matrix (having ones on the diagonal and zeros elsewhere). |
FloatMatrix2D |
make(float[][] values)
Constructs a matrix with the given cell values. |
FloatMatrix2D |
make(float[] values,
int rows)
Construct a matrix from a one-dimensional column-major packed array, ala Fortran. |
FloatMatrix2D |
make(int rows,
int columns)
Constructs a matrix with the given shape, each cell initialized with zero. |
FloatMatrix2D |
make(int rows,
int columns,
float initialValue)
Constructs a matrix with the given shape, each cell initialized with the given value. |
FloatMatrix2D |
random(int rows,
int columns)
Constructs a matrix with uniformly distributed values in (0,1) (exclusive). |
FloatMatrix2D |
repeat(FloatMatrix2D A,
int rowRepeat,
int columnRepeat)
C = A||A||..||A; Constructs a new matrix which is duplicated both along the row and column dimension. |
FloatMatrix2D |
reshape(FloatMatrix1D a,
int rows,
int columns)
|
FloatMatrix2D |
sample(FloatMatrix2D matrix,
float value,
float nonZeroFraction)
Modifies the given matrix to be a randomly sampled matrix. |
FloatMatrix2D |
sample(int rows,
int columns,
float value,
float nonZeroFraction)
Constructs a randomly sampled matrix with the given shape. |
Methods inherited from class cern.colt.PersistentObject |
---|
clone |
Methods inherited from class java.lang.Object |
---|
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Field Detail |
---|
public static final FloatFactory2D dense
public static final FloatFactory2D sparse
public static final FloatFactory2D rowCompressed
Method Detail |
---|
public FloatMatrix2D appendColumns(FloatMatrix2D A, FloatMatrix2D B)
0 1 2 3 4 5 appendColumns 6 7 8 9 --> 0 1 2 6 7 3 4 5 8 9
public FloatMatrix2D appendColumn(FloatMatrix2D A, FloatMatrix1D b)
public FloatMatrix2D appendRows(FloatMatrix2D A, FloatMatrix2D B)
0 1 2 3 4 5 appendRows 6 7 8 9 --> 0 1 2 3 4 5 6 7 8 9
public FloatMatrix2D appendRow(FloatMatrix2D A, FloatMatrix1D b)
public FloatMatrix2D ascending(int rows, int columns)
0 1 2 3 4 5
public FloatMatrix2D reshape(FloatMatrix1D a, int rows, int columns)
public FloatMatrix2D compose(FloatMatrix2D[][] parts)
decompose(FloatMatrix2D[][], FloatMatrix2D)
.
All matrices of a given column within parts must have the same number of columns. All matrices of a given row within parts must have the same number of rows. Otherwise an IllegalArgumentException is thrown. Note that nulls within parts[row,col] are an exception to this rule: they are ignored. Cells are copied. Example:
Code | Result |
FloatMatrix2D[][] parts1 = { { null, make(2, 2, 1), null }, { make(4, 4, 2), null, make(4, 3, 3) }, { null, make(2, 2, 4), null } }; System.out.println(compose(parts1)); |
8 x 9 matrix 0 0 0 0 1 1 0 0 0 0 0 0 0 1 1 0 0 0 2 2 2 2 0 0 3 3 3 2 2 2 2 0 0 3 3 3 2 2 2 2 0 0 3 3 3 2 2 2 2 0 0 3 3 3 0 0 0 0 4 4 0 0 0 0 0 0 0 4 4 0 0 0 |
FloatMatrix2D[][] parts3 = { { identity(3), null, }, { null, identity(3).viewColumnFlip() }, { identity(3).viewRowFlip(), null } }; System.out.println("\n" + make(parts3)); |
9 x 6 matrix 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 |
FloatMatrix2D A = ascending(2, 2); FloatMatrix2D B = descending(2, 2); FloatMatrix2D _ = null; FloatMatrix2D[][] parts4 = { { A, _, A, _ }, { _, A, _, B } }; System.out.println("\n" + make(parts4)); |
4 x 8 matrix 1 2 0 0 1 2 0 0 3 4 0 0 3 4 0 0 0 0 1 2 0 0 3 2 0 0 3 4 0 0 1 0 |
FloatMatrix2D[][] parts2 = { { null, make(2, 2, 1), null }, { make(4, 4, 2), null, make(4, 3, 3) }, { null, make(2, 3, 4), null } }; System.out.println("\n" + Factory2D.make(parts2)); |
IllegalArgumentException A[0,1].cols != A[2,1].cols (2 != 3) |
IllegalArgumentException
- subject to the conditions outlined above.public FloatMatrix2D composeDiagonal(FloatMatrix2D A, FloatMatrix2D B)
A 0 0 B(The direct sum has A.rows()+B.rows() rows and A.columns()+B.columns() columns). Cells are copied.
public FloatMatrix2D composeDiagonal(FloatMatrix2D A, FloatMatrix2D B, FloatMatrix2D C)
A 0 0 0 B 0 0 0 Cfrom the given parts. Cells are copied.
public FloatMatrix2D composeBidiagonal(FloatMatrix2D A, FloatMatrix2D B)
public void decompose(FloatMatrix2D[][] parts, FloatMatrix2D matrix)
compose(FloatMatrix2D[][])
.
All matrices of a given column within parts must have the same number of columns. All matrices of a given row within parts must have the same number of rows. Otherwise an IllegalArgumentException is thrown. Note that nulls within parts[row,col] are an exception to this rule: they are ignored. Cells are copied. Example:
Code | matrix | --> parts |
FloatMatrix2D matrix = ... ; FloatMatrix2D _ = null; FloatMatrix2D A,B,C,D; A = make(2,2); B = make (4,4); C = make(4,3); D = make (2,2); FloatMatrix2D[][] parts = { { _, A, _ }, { B, _, C }, { _, D, _ } }; decompose(parts,matrix); System.out.println("\nA = "+A); System.out.println("\nB = "+B); System.out.println("\nC = "+C); System.out.println("\nD = "+D); |
8 x 9 matrix 9 9 9 9 1 1 9 9 9 9 9 9 9 1 1 9 9 9 2 2 2 2 9 9 3 3 3 2 2 2 2 9 9 3 3 3 2 2 2 2 9 9 3 3 3 2 2 2 2 9 9 3 3 3 9 9 9 9 4 4 9 9 9 9 9 9 9 4 4 9 9 9 |
A = 2 x 2 matrix
B = 4 x 4 matrix
C = 4 x 3 matrix
D = 2 x 2 matrix |
IllegalArgumentException
- subject to the conditions outlined above.public void demo1()
public void demo2()
public FloatMatrix2D descending(int rows, int columns)
5 4 3 2 1 0
public FloatMatrix2D diagonal(FloatMatrix1D vector)
5 4 3 --> 5 0 0 0 4 0 0 0 3
public FloatMatrix2D diagonal(float[] vector)
5 4 3 --> 5 0 0 0 4 0 0 0 3
public FloatMatrix1D diagonal(FloatMatrix2D A)
5 0 0 9 0 4 0 9 0 0 3 9 --> 5 4 3
A
- the matrix, need not be square.
public FloatMatrix2D identity(int rowsAndColumns)
public FloatMatrix2D make(float[][] values)
The values are copied. So subsequent changes in values are not reflected in the matrix, and vice-versa.
values
- The values to be filled into the new matrix.
IllegalArgumentException
- if
for any 1 <= row < values.length: values[row].length != values[row-1].length
.public FloatMatrix2D make(float[] values, int rows)
values
- One-dimensional array of floats, packed by columns (ala
Fortran).rows
- the number of rows.
IllegalArgumentException
- values.length must be a multiple of rows
.public FloatMatrix2D make(int rows, int columns)
public FloatMatrix2D make(int rows, int columns, float initialValue)
public FloatMatrix2D random(int rows, int columns)
public FloatMatrix2D repeat(FloatMatrix2D A, int rowRepeat, int columnRepeat)
0 1 2 3 repeat(2,3) --> 0 1 0 1 0 1 2 3 2 3 2 3 0 1 0 1 0 1 2 3 2 3 2 3
public FloatMatrix2D sample(int rows, int columns, float value, float nonZeroFraction)
IllegalArgumentException
- if nonZeroFraction < 0 || nonZeroFraction > 1.FloatRandomSampler
public FloatMatrix2D sample(FloatMatrix2D matrix, float value, float nonZeroFraction)
IllegalArgumentException
- if nonZeroFraction < 0 || nonZeroFraction > 1.FloatRandomSampler
|
Parallel Colt 0.7.2 | |||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |