Package edu.wpi.first.math

Class Matrix<R extends Num,​C extends Num>

java.lang.Object
edu.wpi.first.math.Matrix<R,​C>
Type Parameters:
R - The number of rows in this matrix.
C - The number of columns in this matrix.
Direct Known Subclasses:
Vector
public class Matrix<R extends Num,​C extends Num>
extends Object
A shape-safe wrapper over Efficient Java Matrix Library (EJML) matrices.

This class is intended to be used alongside the state space library.

  • Field Summary

    Fields 
    Modifier and Type Field Description
    protected org.ejml.simple.SimpleMatrix m_storage  

  • Constructor Summary

    Constructors 
    Constructor Description
    Matrix​(Matrix<R,​C> other)
    Constructs a new matrix with the storage of the supplied matrix.
    Matrix​(Nat<R> rows, Nat<C> columns)
    Constructs an empty zero matrix of the given dimensions.
    Matrix​(org.ejml.simple.SimpleMatrix storage)
    Constructs a new Matrix with the given storage.

  • Method Summary

    Modifier and Type Method Description
    <R2 extends Num,​ C2 extends Num>
    void    		 assignBlock​(int startingRow, int startingCol, Matrix<R2,​C2> other)
    Assign a matrix of a given size and start position.
    <R2 extends Num,​ C2 extends Num>
    Matrix<R2,​C2>    		 block​(int height, int width, int startingRow, int startingCol)
    Extracts a matrix of a given size and start position with new underlying storage.
    <R2 extends Num,​ C2 extends Num>
    Matrix<R2,​C2>    		 block​(Nat<R2> height, Nat<C2> width, int startingRow, int startingCol)
    Extracts a matrix of a given size and start position with new underlying storage.
    static <R1 extends Num,​ C1 extends Num>
    Matrix<R1,​C1>    		 changeBoundsUnchecked​(Matrix<?,​?> mat)
    Reassigns dimensions of a Matrix to allow for operations with other matrices that have wildcard dimensions.
    Matrix<R,​C> copy()
    Returns a copy of this matrix.
    double det()
    Returns the determinant of this matrix.
    Matrix<R,​C> diag()
    Returns the diagonal elements inside a vector or square matrix.
    Matrix<R,​C> div​(double value)
    Divides all elements of this matrix by the given value.
    Matrix<R,​C> div​(int value)
    Divides all elements of this matrix by the given value.
    Matrix<R,​C> elementPower​(double b)
    Returns a matrix which is the result of an element by element power of "this" and b.
    Matrix<R,​C> elementPower​(int b)
    Returns a matrix which is the result of an element by element power of "this" and b.
    double elementSum()
    Computes the sum of all the elements in the matrix.
    Matrix<R,​C> elementTimes​(Matrix<R,​C> other)
    Returns a matrix which is the result of an element by element multiplication of "this" and other.
    boolean equals​(Object other)
    Checks if an object is equal to this Matrix.
    Matrix<R,​C> exp()
    Computes the matrix exponential using Eigen's solver.
    Matrix<R,​N1> extractColumnVector​(int column)
    Extracts a given column into a column vector with new underlying storage.
    <R2 extends Num,​ C2 extends Num>
    void    		 extractFrom​(int startingRow, int startingCol, Matrix<R2,​C2> other)
    Extracts a submatrix from the supplied matrix and inserts it in a submatrix in "this".
    Matrix<N1,​C> extractRowVector​(int row)
    Extracts a given row into a row vector with new underlying storage.
    static <D extends Num>
    Matrix<D,​D>    		 eye​(D dim)
    Creates the identity matrix of the given dimension.
    static <D extends Num>
    Matrix<D,​D>    		 eye​(Nat<D> dim)
    Creates the identity matrix of the given dimension.
    void fill​(double value)
    Sets all the elements in "this" matrix equal to the specified value.
    double get​(int row, int col)
    Get an element of this matrix.
    double[] getData()
    Returns the row major data of this matrix as a double array.
    int getNumCols()
    Gets the number of columns in this matrix.
    int getNumRows()
    Gets the number of rows in this matrix.
    org.ejml.simple.SimpleMatrix getStorage()
    Gets the underlying SimpleMatrix that this Matrix wraps.
    int hashCode()  
    Matrix<R,​C> inv()
    Returns the inverse matrix of "this" matrix.
    boolean isEqual​(Matrix<?,​?> other, double tolerance)
    Checks if another Matrix is equal to "this" within a specified tolerance.
    boolean isIdentical​(Matrix<?,​?> other, double tolerance)
    Checks if another Matrix is identical to "this" one within a specified tolerance.
    Matrix<R,​C> lltDecompose​(boolean lowerTriangular)
    Decompose "this" matrix using Cholesky Decomposition.
    static <R extends Num,​ C extends Num>
    MatBuilder<R,​C>    		 mat​(Nat<R> rows, Nat<C> cols)
    Entrypoint to the MatBuilder class for creation of custom matrices with the given dimensions and contents.
    double max()
    Returns the largest element of this matrix.
    double maxAbs()
    Returns the absolute value of the element in this matrix with the largest absolute value.
    double mean()
    Calculates the mean of the elements in this matrix.
    double minInternal()
    Returns the smallest element of this matrix.
    Matrix<R,​C> minus​(double value)
    Subtracts the given value from all the elements of this matrix.
    Matrix<R,​C> minus​(Matrix<R,​C> value)
    Subtracts the given matrix from this matrix.
    double normF()
    Computes the Frobenius normal of the matrix.
    double normIndP1()
    Computes the induced p = 1 matrix norm.
    Matrix<R,​C> plus​(double value)
    Adds the given value to all the elements of this matrix.
    Matrix<R,​C> plus​(Matrix<R,​C> value)
    Adds the given matrix to this matrix.
    Matrix<R,​C> pow​(double exponent)
    Computes the matrix power using Eigen's solver.
    void rankUpdate​(Matrix<R,​N1> v, double sigma, boolean lowerTriangular)
    Performs an inplace Cholesky rank update (or downdate).
    void set​(int row, int col, double value)
    Sets the value at the given indices.
    void setColumn​(int column, Matrix<R,​N1> val)
    Sets a column to a given column vector.
    void setRow​(int row, Matrix<N1,​C> val)
    Sets a row to a given row vector.
    <C2 extends Num>
    Matrix<C,​C2>    		 solve​(Matrix<R,​C2> b)
    Returns the solution x to the equation Ax = b, where A is "this" matrix.
    <R2 extends Num,​ C2 extends Num>
    Matrix<C,​C2>    		 solveFullPivHouseholderQr​(Matrix<R2,​C2> other)
    Solves the least-squares problem Ax=B using a QR decomposition with full pivoting, where this matrix is A.
    Matrix<R,​C> times​(double value)
    Multiplies all the elements of this matrix by the given scalar.
    <C2 extends Num>
    Matrix<R,​C2>    		 times​(Matrix<C,​C2> other)
    Multiplies this matrix with another that has C rows.
    String toString()  
    double trace()
    Computes the trace of the matrix.
    Matrix<C,​R> transpose()
    Calculates the transpose, Mᵀ of this matrix.

    Methods inherited from class java.lang.Object

    clone, finalize, getClass, notify, notifyAll, wait, wait, wait

  • Field Details

  • Constructor Details

    • Matrix

      public Matrix​(Nat<R> rows, Nat<C> columns)
      Constructs an empty zero matrix of the given dimensions.
      Parameters:
      rows - The number of rows of the matrix.
      columns - The number of columns of the matrix.

    • Matrix

      public Matrix​(org.ejml.simple.SimpleMatrix storage)
      Constructs a new Matrix with the given storage. Caller should make sure that the provided generic bounds match the shape of the provided Matrix.

      NOTE:It is not recommend to use this constructor unless the SimpleMatrix API is absolutely necessary due to the desired function not being accessible through the Matrix wrapper.

      Parameters:
      storage - The SimpleMatrix to back this value.

    • Matrix

      public Matrix​(Matrix<R,​C> other)
      Constructs a new matrix with the storage of the supplied matrix.
      Parameters:
      other - The Matrix to copy the storage of.

  • Method Details

    • getStorage

      public org.ejml.simple.SimpleMatrix getStorage()
      Gets the underlying SimpleMatrix that this Matrix wraps.

      NOTE:The use of this method is heavily discouraged as this removes any guarantee of type safety. This should only be called if the SimpleMatrix API is absolutely necessary due to the desired function not being accessible through the Matrix wrapper.

      Returns:
      The underlying SimpleMatrix storage.

    • getNumCols

      public final int getNumCols()
      Gets the number of columns in this matrix.
      Returns:
      The number of columns, according to the internal storage.

    • getNumRows

      public final int getNumRows()
      Gets the number of rows in this matrix.
      Returns:
      The number of rows, according to the internal storage.

    • get

      public final double get​(int row, int col)
      Get an element of this matrix.
      Parameters:
      row - The row of the element.
      col - The column of the element.
      Returns:
      The element in this matrix at row,col.

    • set

      public final void set​(int row, int col, double value)
      Sets the value at the given indices.
      Parameters:
      row - The row of the element.
      col - The column of the element.
      value - The value to insert at the given location.

    • setRow

      public final void setRow​(int row, Matrix<N1,​C> val)
      Sets a row to a given row vector.
      Parameters:
      row - The row to set.
      val - The row vector to set the given row to.

    • setColumn

      public final void setColumn​(int column, Matrix<R,​N1> val)
      Sets a column to a given column vector.
      Parameters:
      column - The column to set.
      val - The column vector to set the given row to.

    • fill

      public void fill​(double value)
      Sets all the elements in "this" matrix equal to the specified value.
      Parameters:
      value - The value each element is set to.

    • diag

      public final Matrix<R,​C> diag()
      Returns the diagonal elements inside a vector or square matrix.

      If "this" Matrix is a vector then a square matrix is returned. If a "this" Matrix is a matrix then a vector of diagonal elements is returned.

      Returns:
      The diagonal elements inside a vector or a square matrix.

    • max

      public final double max()
      Returns the largest element of this matrix.
      Returns:
      The largest element of this matrix.

    • maxAbs

      public final double maxAbs()
      Returns the absolute value of the element in this matrix with the largest absolute value.
      Returns:
      The absolute value of the element with the largest absolute value.

    • minInternal

      public final double minInternal()
      Returns the smallest element of this matrix.
      Returns:
      The smallest element of this matrix.

    • mean

      public final double mean()
      Calculates the mean of the elements in this matrix.
      Returns:
      The mean value of this matrix.

    • times

      public final <C2 extends Num> Matrix<R,​C2> times​(Matrix<C,​C2> other)
      Multiplies this matrix with another that has C rows.

      As matrix multiplication is only defined if the number of columns in the first matrix matches the number of rows in the second, this operation will fail to compile under any other circumstances.

      Type Parameters:
      C2 - The number of columns in the second matrix.
      Parameters:
      other - The other matrix to multiply by.
      Returns:
      The result of the matrix multiplication between "this" and the given matrix.

    • times

      public Matrix<R,​C> times​(double value)
      Multiplies all the elements of this matrix by the given scalar.
      Parameters:
      value - The scalar value to multiply by.
      Returns:
      A new matrix with all the elements multiplied by the given value.

    • elementTimes

      public final Matrix<R,​C> elementTimes​(Matrix<R,​C> other)
      Returns a matrix which is the result of an element by element multiplication of "this" and other.

      ci,j = ai,j*otheri,j

      Parameters:
      other - The other Matrix to perform element multiplication on.
      Returns:
      The element by element multiplication of "this" and other.

    • minus

      public final Matrix<R,​C> minus​(double value)
      Subtracts the given value from all the elements of this matrix.
      Parameters:
      value - The value to subtract.
      Returns:
      The resultant matrix.

    • minus

      public final Matrix<R,​C> minus​(Matrix<R,​C> value)
      Subtracts the given matrix from this matrix.
      Parameters:
      value - The matrix to subtract.
      Returns:
      The resultant matrix.

    • plus

      public final Matrix<R,​C> plus​(double value)
      Adds the given value to all the elements of this matrix.
      Parameters:
      value - The value to add.
      Returns:
      The resultant matrix.

    • plus

      public final Matrix<R,​C> plus​(Matrix<R,​C> value)
      Adds the given matrix to this matrix.
      Parameters:
      value - The matrix to add.
      Returns:
      The resultant matrix.

    • div

      public Matrix<R,​C> div​(int value)
      Divides all elements of this matrix by the given value.
      Parameters:
      value - The value to divide by.
      Returns:
      The resultant matrix.

    • div

      public Matrix<R,​C> div​(double value)
      Divides all elements of this matrix by the given value.
      Parameters:
      value - The value to divide by.
      Returns:
      The resultant matrix.

    • transpose

      public final Matrix<C,​R> transpose()
      Calculates the transpose, Mᵀ of this matrix.
      Returns:
      The transpose matrix.

    • copy

      public final Matrix<R,​C> copy()
      Returns a copy of this matrix.
      Returns:
      A copy of this matrix.

    • inv

      public final Matrix<R,​C> inv()
      Returns the inverse matrix of "this" matrix.
      Returns:
      The inverse of "this" matrix.
      Throws:
      org.ejml.data.SingularMatrixException - If "this" matrix is non-invertable.

    • solve

      public final <C2 extends Num> Matrix<C,​C2> solve​(Matrix<R,​C2> b)
      Returns the solution x to the equation Ax = b, where A is "this" matrix.

      The matrix equation could also be written as x = A-1b. Where the pseudo inverse is used if A is not square.

      Note that this method does not support solving using a QR decomposition with full-pivoting, as only column-pivoting is supported. For full-pivoting, use solveFullPivHouseholderQr(edu.wpi.first.math.Matrix<R2, C2>).

      Type Parameters:
      C2 - Columns in b.
      Parameters:
      b - The right-hand side of the equation to solve.
      Returns:
      The solution to the linear system.

    • solveFullPivHouseholderQr

      public final <R2 extends Num,​ C2 extends Num> Matrix<C,​C2> solveFullPivHouseholderQr​(Matrix<R2,​C2> other)
      Solves the least-squares problem Ax=B using a QR decomposition with full pivoting, where this matrix is A.
      Type Parameters:
      R2 - Number of rows in B.
      C2 - Number of columns in B.
      Parameters:
      other - The B matrix.
      Returns:
      The solution matrix.

    • exp

      public final Matrix<R,​C> exp()
      Computes the matrix exponential using Eigen's solver. This method only works for square matrices, and will otherwise throw an MatrixDimensionException.
      Returns:
      The exponential of A.

    • pow

      public final Matrix<R,​C> pow​(double exponent)
      Computes the matrix power using Eigen's solver. This method only works for square matrices, and will otherwise throw an MatrixDimensionException.
      Parameters:
      exponent - The exponent.
      Returns:
      The exponential of A.

    • det

      public final double det()
      Returns the determinant of this matrix.
      Returns:
      The determinant of this matrix.

    • normF

      public final double normF()
      Computes the Frobenius normal of the matrix.

      normF = Sqrt{ ∑i=1:m ∑j=1:n { aij2} }

      Returns:
      The matrix's Frobenius normal.

    • normIndP1

      public final double normIndP1()
      Computes the induced p = 1 matrix norm.

      ||A||1= max(j=1 to n; sum(i=1 to m; |aij|))

      Returns:
      The norm.

    • elementSum

      public final double elementSum()
      Computes the sum of all the elements in the matrix.
      Returns:
      Sum of all the elements.

    • trace

      public final double trace()
      Computes the trace of the matrix.
      Returns:
      The trace of the matrix.

    • elementPower

      public final Matrix<R,​C> elementPower​(double b)
      Returns a matrix which is the result of an element by element power of "this" and b.

      ci,j = ai,j ^ b

      Parameters:
      b - Scalar.
      Returns:
      The element by element power of "this" and b.

    • elementPower

      public final Matrix<R,​C> elementPower​(int b)
      Returns a matrix which is the result of an element by element power of "this" and b.

      ci,j = ai,j ^ b

      Parameters:
      b - Scalar.
      Returns:
      The element by element power of "this" and b.

    • extractRowVector

      public final Matrix<N1,​C> extractRowVector​(int row)
      Extracts a given row into a row vector with new underlying storage.
      Parameters:
      row - The row to extract a vector from.
      Returns:
      A row vector from the given row.

    • extractColumnVector

      public final Matrix<R,​N1> extractColumnVector​(int column)
      Extracts a given column into a column vector with new underlying storage.
      Parameters:
      column - The column to extract a vector from.
      Returns:
      A column vector from the given column.

    • block

      public final <R2 extends Num,​ C2 extends Num> Matrix<R2,​C2> block​(Nat<R2> height, Nat<C2> width, int startingRow, int startingCol)
      Extracts a matrix of a given size and start position with new underlying storage.
      Type Parameters:
      R2 - Number of rows to extract.
      C2 - Number of columns to extract.
      Parameters:
      height - The number of rows of the extracted matrix.
      width - The number of columns of the extracted matrix.
      startingRow - The starting row of the extracted matrix.
      startingCol - The starting column of the extracted matrix.
      Returns:
      The extracted matrix.

    • block

      public final <R2 extends Num,​ C2 extends Num> Matrix<R2,​C2> block​(int height, int width, int startingRow, int startingCol)
      Extracts a matrix of a given size and start position with new underlying storage.
      Type Parameters:
      R2 - Number of rows to extract.
      C2 - Number of columns to extract.
      Parameters:
      height - The number of rows of the extracted matrix.
      width - The number of columns of the extracted matrix.
      startingRow - The starting row of the extracted matrix.
      startingCol - The starting column of the extracted matrix.
      Returns:
      The extracted matrix.

    • assignBlock

      public <R2 extends Num,​ C2 extends Num> void assignBlock​(int startingRow, int startingCol, Matrix<R2,​C2> other)
      Assign a matrix of a given size and start position.
      Type Parameters:
      R2 - Rows in block assignment.
      C2 - Columns in block assignment.
      Parameters:
      startingRow - The row to start at.
      startingCol - The column to start at.
      other - The matrix to assign the block to.

    • extractFrom

      public <R2 extends Num,​ C2 extends Num> void extractFrom​(int startingRow, int startingCol, Matrix<R2,​C2> other)
      Extracts a submatrix from the supplied matrix and inserts it in a submatrix in "this". The shape of "this" is used to determine the size of the matrix extracted.
      Type Parameters:
      R2 - Number of rows to extract.
      C2 - Number of columns to extract.
      Parameters:
      startingRow - The starting row in the supplied matrix to extract the submatrix.
      startingCol - The starting column in the supplied matrix to extract the submatrix.
      other - The matrix to extract the submatrix from.

    • lltDecompose

      public Matrix<R,​C> lltDecompose​(boolean lowerTriangular)
      Decompose "this" matrix using Cholesky Decomposition. If the "this" matrix is zeros, it will return the zero matrix.
      Parameters:
      lowerTriangular - Whether we want to decompose to the lower triangular Cholesky matrix.
      Returns:
      The decomposed matrix.
      Throws:
      RuntimeException - if the matrix could not be decomposed(i.e. is not positive semidefinite).

    • getData

      public double[] getData()
      Returns the row major data of this matrix as a double array.
      Returns:
      The row major data of this matrix as a double array.

    • eye

      public static <D extends Num> Matrix<D,​D> eye​(Nat<D> dim)
      Creates the identity matrix of the given dimension.
      Type Parameters:
      D - The dimension of the desired matrix as a generic.
      Parameters:
      dim - The dimension of the desired matrix as a Nat.
      Returns:
      The DxD identity matrix.

    • eye

      public static <D extends Num> Matrix<D,​D> eye​(D dim)
      Creates the identity matrix of the given dimension.
      Type Parameters:
      D - The dimension of the desired matrix as a generic.
      Parameters:
      dim - The dimension of the desired matrix as a Num.
      Returns:
      The DxD identity matrix.

    • mat

      public static <R extends Num,​ C extends Num> MatBuilder<R,​C> mat​(Nat<R> rows, Nat<C> cols)
      Entrypoint to the MatBuilder class for creation of custom matrices with the given dimensions and contents.
      Type Parameters:
      R - The number of rows of the desired matrix as a generic.
      C - The number of columns of the desired matrix as a generic.
      Parameters:
      rows - The number of rows of the desired matrix.
      cols - The number of columns of the desired matrix.
      Returns:
      A builder to construct the matrix.

    • changeBoundsUnchecked

      public static <R1 extends Num,​ C1 extends Num> Matrix<R1,​C1> changeBoundsUnchecked​(Matrix<?,​?> mat)
      Reassigns dimensions of a Matrix to allow for operations with other matrices that have wildcard dimensions.
      Type Parameters:
      R1 - Row dimension to assign.
      C1 - Column dimension to assign.
      Parameters:
      mat - The Matrix to remove the dimensions from.
      Returns:
      The matrix with reassigned dimensions.

    • isIdentical

      public boolean isIdentical​(Matrix<?,​?> other, double tolerance)
      Checks if another Matrix is identical to "this" one within a specified tolerance.

      This will check if each element is in tolerance of the corresponding element from the other Matrix or if the elements have the same symbolic meaning. For two elements to have the same symbolic meaning they both must be either Double.NaN, Double.POSITIVE_INFINITY, or Double.NEGATIVE_INFINITY.

      NOTE:It is recommended to use isEqual(Matrix, double) over this method when checking if two matrices are equal as isEqual(Matrix, double) will return false if an element is uncountable. This method should only be used when uncountable elements need to be compared.

      Parameters:
      other - The Matrix to check against this one.
      tolerance - The tolerance to check equality with.
      Returns:
      true if this matrix is identical to the one supplied.

    • isEqual

      public boolean isEqual​(Matrix<?,​?> other, double tolerance)
      Checks if another Matrix is equal to "this" within a specified tolerance.

      This will check if each element is in tolerance of the corresponding element from the other Matrix.

      tol ≥ |aij - bij|

      Parameters:
      other - The Matrix to check against this one.
      tolerance - The tolerance to check equality with.
      Returns:
      true if this matrix is equal to the one supplied.

    • rankUpdate

      public void rankUpdate​(Matrix<R,​N1> v, double sigma, boolean lowerTriangular)
      Performs an inplace Cholesky rank update (or downdate).

      If this matrix contains L where A = LLᵀ before the update, it will contain L where LLᵀ = A + σvvᵀ after the update.

      Parameters:
      v - Vector to use for the update.
      sigma - Sigma to use for the update.
      lowerTriangular - Whether this matrix is lower triangular.

    • toString

      public String toString()
      Overrides:
      toString in class Object

    • equals

      public boolean equals​(Object other)
      Checks if an object is equal to this Matrix.

      aij == bij

      Overrides:
      equals in class Object
      Parameters:
      other - The Object to check against this Matrix.
      Returns:
      true if the object supplied is a Matrix and is equal to this matrix.

    • hashCode

      public int hashCode()
      Overrides:
      hashCode in class Object