| Package | Description |
|---|---|
| org.kramerlab.bmad.algorithms |
Contains many different implementations of essential parts of a
boolean matrix decomposition algorithm, that is: candidate generators,
basis selectors, and row combinators.
|
| org.kramerlab.bmad.general |
Contains few workarounds that allow to work with tuples and functions.
|
| org.kramerlab.bmad.matrix |
Contains implementations of elementary data structures, required to
perform boolean matrix decompositions: matrices.
|
| Modifier and Type | Method and Description |
|---|---|
Tuple<Integer,Integer> |
Cover.apply(int c)
returns tuple
(#TRUE, #UNKNOWN) |
Tuple<Integer,Integer> |
Cover.coveredOnesAndZerosOnInclusion(BooleanMatrix coveredRow,
BooleanMatrix basisRow)
Predicts the number of covered ones and zeros, given the row that is
being covered, and the basis row that could be included into this cover.
|
Tuple<BooleanMatrix,BooleanMatrix> |
BooleanMatrixDecomposition.decompose(BooleanMatrix a,
int dimension)
Decomposes matrix into two factor matrices.
|
Tuple<weka.core.Instances,weka.core.Instances> |
BooleanMatrixDecomposition.decompose(weka.core.Instances a,
int dimension)
Thin wrapper around the other decompose method, that translates input and
output into Weka Instances.
|
Tuple<BooleanMatrix,BooleanMatrix> |
BasisSelector.selectBasis(BooleanMatrix candidates,
BooleanMatrix a,
int dimension,
double onesWeight)
Calculates a boolean basis matrix B (second factor), and (optionally) a
coarse approximation of the combination matrix C (first factor).
|
Tuple<BooleanMatrix,BooleanMatrix> |
GreedySelector.selectBasis(BooleanMatrix candidates,
BooleanMatrix a,
int dimension,
double onesWeight) |
Tuple<BooleanMatrix,BooleanMatrix> |
FastLoc.selectBasis(BooleanMatrix candidates,
BooleanMatrix a,
int dimension,
double onesWeight) |
Tuple<Integer,Integer> |
Cover.uncoveredOnesAndZerosOnExclusion(BooleanMatrix coveredRow,
BooleanMatrix basisRow)
analogous to coveredOnesAndZerosOnInclusion
|
| Modifier and Type | Method and Description |
|---|---|
static <A,B> Tuple<A,B> |
Package.tuple(A a,
B b) |
| Modifier and Type | Method and Description |
|---|---|
Tuple<Double,Double> |
BooleanMatrix.relativeOneZeroZeroOneReconstructionError(BooleanMatrix reconstruction,
double onesWeight)
Calculates the relative
1->0 and 0->1
reconstruction error, that is, the total error divided by total
maximum possible error. |
Tuple<Integer,Integer> |
BooleanMatrix.size()
Returns width and height of the matrix.
|
Tuple<Integer,Integer> |
RowMajor.size()
Height and width of the matrix.
|
| Modifier and Type | Method and Description |
|---|---|
<Y> RowMajor<Y> |
RowMajor.mapWithIndices(Function<Tuple<X,Tuple<Integer,Integer>>,Y> f)
Pointwise application of a function that can depend on the value and
the index.
|
<Y> RowMajor<Y> |
RowMajor.mapWithIndices(Function<Tuple<X,Tuple<Integer,Integer>>,Y> f)
Pointwise application of a function that can depend on the value and
the index.
|
| Constructor and Description |
|---|
RowMajor(int h,
int w,
Function<Tuple<Integer,Integer>,X> fill)
Creates new matrix of the specified size, with entries given by the function
entry. |
Copyright © 2015. All rights reserved.