class SparseMatrixD extends MatriD with Error with Serializable
The SparseMatrixD class stores and operates on Matrices of Doubles. Rather
than storing the matrix as a 2 dimensional array, it is stored as an array
of sorted-linked-maps, which record all the non-zero values for each particular
row, along with their j-index as (j, v) pairs.
- Alphabetic
- By Inheritance
- SparseMatrixD
- Serializable
- MatriD
- Error
- AnyRef
- Any
- Hide All
- Show All
- Public
- Protected
Instance Constructors
- new SparseMatrixD(b: MatrixD)
Construct a sparse matrix and assign values from dense matrix
MatrixD'b'.Construct a sparse matrix and assign values from dense matrix
MatrixD'b'.- b
the matrix of values to assign
- new SparseMatrixD(b: SparseMatrixD)
Construct a sparse matrix and assign values from matrix 'b'.
Construct a sparse matrix and assign values from matrix 'b'.
- b
the matrix of values to assign
- new SparseMatrixD(dim: (Int, Int), u: Double*)
Construct a matrix from repeated values.
Construct a matrix from repeated values.
- dim
the (row, column) dimensions
- u
the repeated values
- new SparseMatrixD(dim1: Int, dim2: Int, x: Double)
Construct a 'dim1' by 'dim2' sparse matrix and assign each element the value 'x'.
Construct a 'dim1' by 'dim2' sparse matrix and assign each element the value 'x'.
- dim1
the row dimension
- dim2
the column dimension
- x
the scalar value to assign
- new SparseMatrixD(dim1: Int)
Construct a 'dim1' by 'dim1' square sparse matrix.
Construct a 'dim1' by 'dim1' square sparse matrix.
- dim1
the row and column dimension
- new SparseMatrixD(dim1: Int, dim2: Int, u: Array[TreeMap[Int, Double]])
Construct a 'dim1' by 'dim2' sparse matrix from an array of sorted-linked-maps.
Construct a 'dim1' by 'dim2' sparse matrix from an array of sorted-linked-maps.
- dim1
the row dimension
- dim2
the column dimension
- u
the array of sorted-linked-maps
- new SparseMatrixD(d1: Int, d2: Int)
- d1
the first/row dimension
- d2
the second/column dimension
Value Members
- final def !=(arg0: Any): Boolean
- Definition Classes
- AnyRef → Any
- final def ##: Int
- Definition Classes
- AnyRef → Any
- def *(x: Double): SparseMatrixD
Multiply 'this' sparse matrix by scalar 'x'.
Multiply 'this' sparse matrix by scalar 'x'.
- x
the scalar to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def *(u: VectoD): VectorD
Multiply 'this' sparse matrix by vector 'u' (vector elements beyond 'dim2' ignored).
Multiply 'this' sparse matrix by vector 'u' (vector elements beyond 'dim2' ignored).
- u
the vector to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def *(b: MatriD): SparseMatrixD
Multiply 'this' sparse matrix by dense matrix 'b'.
Multiply 'this' sparse matrix by dense matrix 'b'.
- b
the matrix to multiply by (requires 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def **(u: VectoD): SparseMatrixD
Multiply 'this' sparse matrix by vector 'u' to produce another matrix 'a_ij * u_j'.
Multiply 'this' sparse matrix by vector 'u' to produce another matrix 'a_ij * u_j'.
- u
the vector to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def **(b: MatriD): MatriD
Multiply 'this' matrix by matrix 'b' elementwise (Hadamard product).
Multiply 'this' matrix by matrix 'b' elementwise (Hadamard product).
- b
the matrix to multiply by
- Definition Classes
- MatriD
- See also
en.wikipedia.org/wiki/Hadamard_product_(matrices) FIX - remove ??? and implement in all implementing classes
- def **:(u: VectoD): SparseMatrixD
Multiply vector 'u' by 'this' matrix to produce another matrix 'u_i * a_ij'.
Multiply vector 'u' by 'this' matrix to produce another matrix 'u_i * a_ij'. E.g., multiply a diagonal matrix represented as a vector by a matrix. This operator is right associative.
- u
the vector to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def **=(u: VectoD): SparseMatrixD
Multiply in-place 'this' sparse matrix by vector 'u' to produce another matrix 'a_ij * u_j'.
Multiply in-place 'this' sparse matrix by vector 'u' to produce another matrix 'a_ij * u_j'.
- u
the vector to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def *:(u: VectoD): VectoD
Multiply (row) vector 'u' by 'this' matrix.
Multiply (row) vector 'u' by 'this' matrix. Note '*:' is right associative. vector = vector *: matrix
- u
the vector to multiply by
- Definition Classes
- MatriD
- def *=(x: Double): SparseMatrixD
Multiply in-place 'this' sparse matrix by scalar 'x'.
Multiply in-place 'this' sparse matrix by scalar 'x'.
- x
the scalar to multiply by
- Definition Classes
- SparseMatrixD → MatriD
- def *=(b: MatriD): SparseMatrixD
Multiply in-place 'this' sparse matrix by dense matrix 'b'.
Multiply in-place 'this' sparse matrix by dense matrix 'b'.
- b
the matrix to multiply by (requires square and 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def +(x: Double): MatrixD
Add 'this' sparse matrix and scalar 'x'.
Add 'this' sparse matrix and scalar 'x'. Note: every element will be likely filled, hence the return type is a dense matrix.
- x
the scalar to add
- Definition Classes
- SparseMatrixD → MatriD
- def +(u: VectoD): SparseMatrixD
Add 'this' sparse matrix and (row) vector 'u'.
Add 'this' sparse matrix and (row) vector 'u'.
- u
the vector to add
- Definition Classes
- SparseMatrixD → MatriD
- def +(b: MatriD): SparseMatrixD
Add 'this' sparse matrix and matrix 'b'.
Add 'this' sparse matrix and matrix 'b'. 'b' may be any subtype of
MatriD. Note, subtypes ofMatriDshould also implement a more efficient version, e.g.,def + (b: SparseMatrixD): SparseMatrixD.- b
the matrix to add (requires 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def +(b: SparseMatrixD): SparseMatrixD
Add 'this' sparse matrix and sparse matrix 'b'.
Add 'this' sparse matrix and sparse matrix 'b'.
- b
the matrix to add (requires 'sameCrossDimensions')
- def ++(b: MatriD): SparseMatrixD
Concatenate (row-wise) 'this' matrix and matrix 'b'.
Concatenate (row-wise) 'this' matrix and matrix 'b'.
- b
the matrix to be concatenated as the new last rows in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- def ++^(b: MatriD): SparseMatrixD
Concatenate (column-wise) 'this' matrix and matrix 'b'.
Concatenate (column-wise) 'this' matrix and matrix 'b'.
- b
the matrix to be concatenated as the new last columns in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- def +:(u: VectoD): SparseMatrixD
Concatenate (row) vector 'u' and 'this' matrix, i.e., prepend 'u' to 'this'.
Concatenate (row) vector 'u' and 'this' matrix, i.e., prepend 'u' to 'this'.
- u
the vector to be prepended as the new first row in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- def +=(x: Double): SparseMatrixD
Add in-place 'this' sparse matrix and scalar 'x'.
Add in-place 'this' sparse matrix and scalar 'x'.
- x
the scalar to add
- Definition Classes
- SparseMatrixD → MatriD
- def +=(u: VectoD): SparseMatrixD
Add in-place this matrix and (row) vector 'u'.
Add in-place this matrix and (row) vector 'u'.
- u
the vector to add
- Definition Classes
- SparseMatrixD → MatriD
- def +=(b: MatriD): SparseMatrixD
Add in-place 'this' sparse matrix and matrix 'b'.
Add in-place 'this' sparse matrix and matrix 'b'.
- b
the matrix to add (requires 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def +=(b: SparseMatrixD): SparseMatrixD
Add in-place 'this' sparse matrix and sparse matrix 'b'.
Add in-place 'this' sparse matrix and sparse matrix 'b'.
- b
the matrix to add (requires 'sameCrossDimensions')
- def +^:(u: VectoD): SparseMatrixD
Concatenate (column) vector 'u' and 'this' matrix, i.e., prepend 'u' to 'this'.
Concatenate (column) vector 'u' and 'this' matrix, i.e., prepend 'u' to 'this'.
- u
the vector to be prepended as the new first column in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- def -(x: Double): MatrixD
From 'this' sparse matrix subtract scalar 'x'.
From 'this' sparse matrix subtract scalar 'x'. Note: every element will be likely filled, hence the return type is a dense matrix.
- x
the scalar to subtract
- Definition Classes
- SparseMatrixD → MatriD
- def -(u: VectoD): SparseMatrixD
From
thissparse matrix subtract (row) vector 'u'.From
thissparse matrix subtract (row) vector 'u'.- u
the vector to subtract
- Definition Classes
- SparseMatrixD → MatriD
- def -(b: MatriD): SparseMatrixD
From 'this' sparse matrix subtract matrix 'b'.
From 'this' sparse matrix subtract matrix 'b'.
- b
the matrix to subtract (requires 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def -(b: SparseMatrixD): SparseMatrixD
From 'this' sparse matrix subtract matrix 'b'.
From 'this' sparse matrix subtract matrix 'b'.
- b
the sparse matrix to subtract (requires 'sameCrossDimensions')
- def -=(x: Double): SparseMatrixD
From 'this' sparse matrix subtract in-place scalar 'x'.
From 'this' sparse matrix subtract in-place scalar 'x'.
- x
the scalar to subtract
- Definition Classes
- SparseMatrixD → MatriD
- def -=(u: VectoD): SparseMatrixD
From
thissparse matrix subtract in-place (row) vector 'u'.From
thissparse matrix subtract in-place (row) vector 'u'.- u
the vector to subtract
- Definition Classes
- SparseMatrixD → MatriD
- def -=(b: MatriD): SparseMatrixD
From 'this' sparse matrix subtract in-place matrix 'b'.
From 'this' sparse matrix subtract in-place matrix 'b'.
- b
the matrix to subtract (requires 'sameCrossDimensions')
- Definition Classes
- SparseMatrixD → MatriD
- def -=(b: SparseMatrixD): SparseMatrixD
From 'this' sparse matrix subtract in-place sparse matrix 'b'.
From 'this' sparse matrix subtract in-place sparse matrix 'b'.
- b
the sparse matrix to subtract (requires 'sameCrossDimensions')
- def /(x: Double): SparseMatrixD
Divide 'this' sparse matrix by scalar 'x'.
Divide 'this' sparse matrix by scalar 'x'.
- x
the scalar to divide by
- Definition Classes
- SparseMatrixD → MatriD
- def /=(x: Double): SparseMatrixD
Divide in-place 'this' sparse matrix by scalar 'x'.
Divide in-place 'this' sparse matrix by scalar 'x'.
- x
the scalar to divide by
- Definition Classes
- SparseMatrixD → MatriD
- def :+(u: VectoD): SparseMatrixD
Concatenate 'this' matrix and (row) vector 'u', i.e., append 'u' to 'this'.
Concatenate 'this' matrix and (row) vector 'u', i.e., append 'u' to 'this'.
- u
the vector to be appended as the new last row in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- def :^+(u: VectoD): SparseMatrixD
Concatenate 'this' matrix and (column) vector 'u', i.e., append 'u' to 'this'.
Concatenate 'this' matrix and (column) vector 'u', i.e., append 'u' to 'this'.
- u
the vector to be appended as the new last column in new matrix
- Definition Classes
- SparseMatrixD → MatriD
- final def ==(arg0: Any): Boolean
- Definition Classes
- AnyRef → Any
- def apply(ir: Range, jr: Range): SparseMatrixD
Get a slice this matrix row-wise on range 'ir' and column-wise on range 'jr'.
Get a slice this matrix row-wise on range 'ir' and column-wise on range 'jr'. Ex: b = a(2..4, 3..5)
- ir
the row range
- jr
the column range
- Definition Classes
- SparseMatrixD → MatriD
- def apply(i: Int): VectorD
Get 'this' sparse matrix's vector at the 'i'-th index position ('i'-th row).
Get 'this' sparse matrix's vector at the 'i'-th index position ('i'-th row).
- i
the row index
- Definition Classes
- SparseMatrixD → MatriD
- def apply(i: Int, j: Int): Double
Get 'this' sparse matrix's element at the 'i,j'-th index position.
Get 'this' sparse matrix's element at the 'i,j'-th index position.
- i
the row index
- j
the column index
- Definition Classes
- SparseMatrixD → MatriD
- def apply(iv: VectoI): MatriD
Get the rows indicated by the index vector 'iv' FIX - implement in all implementing classes
Get the rows indicated by the index vector 'iv' FIX - implement in all implementing classes
- iv
the vector of row indices
- Definition Classes
- MatriD
- def apply(i: Int, jr: Range): VectoD
Get a slice 'this' matrix row-wise at index 'i' and column-wise on range 'jr'.
Get a slice 'this' matrix row-wise at index 'i' and column-wise on range 'jr'. Ex: u = a(2, 3..5)
- i
the row index
- jr
the column range
- Definition Classes
- MatriD
- def apply(ir: Range, j: Int): VectoD
Get a slice 'this' matrix row-wise on range 'ir' and column-wise at index j.
Get a slice 'this' matrix row-wise on range 'ir' and column-wise at index j. Ex: u = a(2..4, 3)
- ir
the row range
- j
the column index
- Definition Classes
- MatriD
- final def asInstanceOf[T0]: T0
- Definition Classes
- Any
- def bsolve(y: VectoD): VectorD
Solve for 'x' using back substitution in the equation 'u*x = y' where 'this' matrix ('u') is upper triangular (see 'lud_npp' above).
Solve for 'x' using back substitution in the equation 'u*x = y' where 'this' matrix ('u') is upper triangular (see 'lud_npp' above).
- y
the constant vector
- Definition Classes
- SparseMatrixD → MatriD
- def clean(thres: Double, relative: Boolean = true): SparseMatrixD
Clean values in matrix at or below the threshold by setting them to zero.
Clean values in matrix at or below the threshold by setting them to zero. Iterative algorithms give approximate values and if very close to zero, may throw off other calculations, e.g., in computing eigenvectors.
- thres
the cutoff threshold (a small value)
- relative
whether to use relative or absolute cutoff
- Definition Classes
- SparseMatrixD → MatriD
- def clone(): AnyRef
- Attributes
- protected[lang]
- Definition Classes
- AnyRef
- Annotations
- @throws(classOf[java.lang.CloneNotSupportedException]) @native() @HotSpotIntrinsicCandidate()
- def col(col: Int, from: Int = 0): VectorD
Get column 'col' from the matrix, returning it as a vector.
Get column 'col' from the matrix, returning it as a vector.
- col
the column to extract from the matrix
- from
the position to start extracting from
- Definition Classes
- SparseMatrixD → MatriD
- def copy: SparseMatrixD
Create a clone of 'this' 'm-by-n' sparse matrix.
Create a clone of 'this' 'm-by-n' sparse matrix.
- Definition Classes
- SparseMatrixD → MatriD
- val d1: Int
- val d2: Int
- def det: Double
Compute the determinant of 'this' sparse matrix.
Compute the determinant of 'this' sparse matrix.
- Definition Classes
- SparseMatrixD → MatriD
- def diag(p: Int, q: Int): SparseMatrixD
Form a matrix '[Ip, this, Iq]' where 'Ir' is a 'r-by-r' identity matrix, by positioning the three matrices 'Ip', 'this' and 'Iq' along the diagonal.
Form a matrix '[Ip, this, Iq]' where 'Ir' is a 'r-by-r' identity matrix, by positioning the three matrices 'Ip', 'this' and 'Iq' along the diagonal.
- p
the size of identity matrix Ip
- q
the size of identity matrix Iq
- Definition Classes
- SparseMatrixD → MatriD
- def diag(b: MatriD): SparseMatrixD
Combine 'this' sparse matrix with matrix 'b', placing them along the diagonal and filling in the bottom left and top right regions with zeros: '[this, b]'.
Combine 'this' sparse matrix with matrix 'b', placing them along the diagonal and filling in the bottom left and top right regions with zeros: '[this, b]'.
- b
the matrix to combine with this matrix
- Definition Classes
- SparseMatrixD → MatriD
- lazy val dim1: Int
Dimension 1
Dimension 1
- Definition Classes
- SparseMatrixD → MatriD
- lazy val dim2: Int
Dimension 2
Dimension 2
- Definition Classes
- SparseMatrixD → MatriD
- def dot(b: MatriD): VectorD
Compute the dot product of 'this' matrix with matrix 'b' to produce a vector.
Compute the dot product of 'this' matrix with matrix 'b' to produce a vector.
- b
the second matrix of the dot product
- Definition Classes
- SparseMatrixD → MatriD
- def dot(u: VectoD): VectorD
Compute the dot product of 'this' matrix and vector 'u', by first transposing 'this' matrix and then multiplying by 'u' (i.e., 'a dot u = a.t * u').
Compute the dot product of 'this' matrix and vector 'u', by first transposing 'this' matrix and then multiplying by 'u' (i.e., 'a dot u = a.t * u').
- u
the vector to multiply by (requires same first dimensions)
- Definition Classes
- SparseMatrixD → MatriD
- final def eq(arg0: AnyRef): Boolean
- Definition Classes
- AnyRef
- def equals(arg0: AnyRef): Boolean
- Definition Classes
- AnyRef → Any
- val fString: String
Format string used for printing vector values (change using 'setFormat')
Format string used for printing vector values (change using 'setFormat')
- Attributes
- protected
- Definition Classes
- MatriD
- def flatten: VectoD
Flatten 'this' matrix in row-major fashion, returning a vector containing all the elements from the matrix.
Flatten 'this' matrix in row-major fashion, returning a vector containing all the elements from the matrix.
- Definition Classes
- MatriD
- final def flaw(method: String, message: String): Unit
Show the flaw by printing the error message.
Show the flaw by printing the error message.
- method
the method where the error occurred
- message
the error message
- Definition Classes
- Error
- def foreach[U](f: (Array[Double]) => U): Unit
Iterate over 'this' matrix row by row applying method 'f'.
Iterate over 'this' matrix row by row applying method 'f'.
- f
the function to apply
- Definition Classes
- MatriD
- final def getClass(): Class[_ <: AnyRef]
- Definition Classes
- AnyRef → Any
- Annotations
- @native() @HotSpotIntrinsicCandidate()
- def getDiag(k: Int = 0): VectorD
Get the 'k'th diagonal of this matrix.
Get the 'k'th diagonal of this matrix. Assumes 'dim2 >= dim1'.
- k
how far above the main diagonal, e.g., (-1, 0, 1) for (sub, main, super)
- Definition Classes
- SparseMatrixD → MatriD
- def hashCode(): Int
- Definition Classes
- AnyRef → Any
- Annotations
- @native() @HotSpotIntrinsicCandidate()
- def inverse: SparseMatrixD
Invert 'this' sparse matrix (requires a 'squareMatrix') using partial pivoting.
Invert 'this' sparse matrix (requires a 'squareMatrix') using partial pivoting.
- Definition Classes
- SparseMatrixD → MatriD
- def inverse_ip(): SparseMatrixD
Invert in-place 'this' sparse matrix (requires a 'squareMatrix').
Invert in-place 'this' sparse matrix (requires a 'squareMatrix'). This version uses partial pivoting.
- Definition Classes
- SparseMatrixD → MatriD
- def inverse_npp: SparseMatrixD
Invert 'this' sparse matrix (requires a 'squareMatrix') not using partial pivoting.
- def isBidiagonal: Boolean
Check whether 'this' matrix is bidiagonal (has non-zero elements only in main diagonal and super-diagonal).
Check whether 'this' matrix is bidiagonal (has non-zero elements only in main diagonal and super-diagonal). The method may be overriding for efficiency.
- Definition Classes
- MatriD
- final def isInstanceOf[T0]: Boolean
- Definition Classes
- Any
- def isNonnegative: Boolean
Check whether 'this' sparse matrix is nonnegative (has no negative elements).
Check whether 'this' sparse matrix is nonnegative (has no negative elements).
- Definition Classes
- SparseMatrixD → MatriD
- def isRectangular: Boolean
Check whether 'this' sparse matrix is rectangular (all rows have the same number of columns).
Check whether 'this' sparse matrix is rectangular (all rows have the same number of columns).
- Definition Classes
- SparseMatrixD → MatriD
- def isSquare: Boolean
Check whether 'this' matrix is square (same row and column dimensions).
Check whether 'this' matrix is square (same row and column dimensions).
- Definition Classes
- MatriD
- def isSymmetric: Boolean
Check whether 'this' matrix is symmetric.
Check whether 'this' matrix is symmetric.
- Definition Classes
- MatriD
- def isTridiagonal: Boolean
Check whether 'this' matrix is bidiagonal (has non-zero elements only in main diagonal and super-diagonal).
Check whether 'this' matrix is bidiagonal (has non-zero elements only in main diagonal and super-diagonal). The method may be overriding for efficiency.
- Definition Classes
- MatriD
- def leDimensions(b: MatriD): Boolean
Check whether 'this' matrix dimensions are less than or equal to 'le' those of the other matrix 'b'.
Check whether 'this' matrix dimensions are less than or equal to 'le' those of the other matrix 'b'.
- b
the other matrix
- Definition Classes
- MatriD
- def lowerT: SparseMatrixD
Return the lower triangular of 'this' matrix (rest are zero).
Return the lower triangular of 'this' matrix (rest are zero).
- Definition Classes
- SparseMatrixD → MatriD
- def lud_ip(): (SparseMatrixD, SparseMatrixD)
Factor in-place 'this' sparse matrix into the product of lower and upper triangular matrices '(l, u)' using the 'LU' Decomposition algorithm.
Factor in-place 'this' sparse matrix into the product of lower and upper triangular matrices '(l, u)' using the 'LU' Decomposition algorithm.
- Definition Classes
- SparseMatrixD → MatriD
- def lud_npp: (SparseMatrixD, SparseMatrixD)
Factor 'this' sparse matrix into the product of lower and upper triangular matrices '(l, u)' using the 'LU' Decomposition algorithm.
Factor 'this' sparse matrix into the product of lower and upper triangular matrices '(l, u)' using the 'LU' Decomposition algorithm.
- Definition Classes
- SparseMatrixD → MatriD
- def mag: Double
Find the magnitude of 'this' matrix, the element value farthest from zero.
Find the magnitude of 'this' matrix, the element value farthest from zero.
- Definition Classes
- MatriD
- def map(f: (VectoD) => VectoD): MatriD
Map the elements of 'this' matrix by applying the mapping function 'f'.
Map the elements of 'this' matrix by applying the mapping function 'f'. FIX - remove ??? and implement in all implementing classes
- f
the function to apply
- Definition Classes
- MatriD
- def max(e: Int = dim1): Double
Find the maximum element in 'this' sparse matrix.
Find the maximum element in 'this' sparse matrix.
- e
the ending row index (exclusive) for the search
- Definition Classes
- SparseMatrixD → MatriD
- def mdot(b: MatriD): SparseMatrixD
Compute the matrix dot product of 'this' matrix and matrix 'b', by first transposing 'this' matrix and then multiplying by 'b' (i.e., 'a dot b = a.t * b').
Compute the matrix dot product of 'this' matrix and matrix 'b', by first transposing 'this' matrix and then multiplying by 'b' (i.e., 'a dot b = a.t * b').
- b
the matrix to multiply by (requires same first dimensions)
- Definition Classes
- SparseMatrixD → MatriD
- def mean: VectoD
Compute the column means of 'this' matrix.
Compute the column means of 'this' matrix.
- Definition Classes
- MatriD
- def meanNZ: VectoD
Compute the column means of 'this' matrix ignoring zero elements (e.g., a zero may indicate a missing value as in recommender systems).
Compute the column means of 'this' matrix ignoring zero elements (e.g., a zero may indicate a missing value as in recommender systems).
- Definition Classes
- MatriD
- def meanR: VectoD
Compute the row means of 'this' matrix.
Compute the row means of 'this' matrix.
- Definition Classes
- MatriD
- def meanRNZ: VectoD
Compute the row means of 'this' matrix ignoring zero elements (e.g., a zero may indicate a missing value as in recommender systems).
Compute the row means of 'this' matrix ignoring zero elements (e.g., a zero may indicate a missing value as in recommender systems).
- Definition Classes
- MatriD
- def min(e: Int = dim1): Double
Find the minimum element in 'this' sparse matrix.
Find the minimum element in 'this' sparse matrix.
- e
the ending row index (exclusive) for the search
- Definition Classes
- SparseMatrixD → MatriD
- final def ne(arg0: AnyRef): Boolean
- Definition Classes
- AnyRef
- def norm1: Double
Compute the 1-norm of 'this' matrix, i.e., the maximum 1-norm of the column vectors.
Compute the 1-norm of 'this' matrix, i.e., the maximum 1-norm of the column vectors. This is useful for comparing matrices '(a - b).norm1'.
- Definition Classes
- MatriD
- See also
en.wikipedia.org/wiki/Matrix_norm
- def normF: Double
Compute the Frobenius-norm of 'this' matrix, i.e., the square root of the sum of the squared values over all the elements (sqrt (sse)).
Compute the Frobenius-norm of 'this' matrix, i.e., the square root of the sum of the squared values over all the elements (sqrt (sse)). FIX: for
MatriCshould take absolute values, first.- Definition Classes
- MatriD
- See also
en.wikipedia.org/wiki/Matrix_norm#Frobenius_norm
- def normFSq: Double
Compute the sqaure of the Frobenius-norm of 'this' matrix, i.e., the sum of the squared values over all the elements (sse).
Compute the sqaure of the Frobenius-norm of 'this' matrix, i.e., the sum of the squared values over all the elements (sse). FIX: for
MatriCshould take absolute values, first.- Definition Classes
- MatriD
- See also
en.wikipedia.org/wiki/Matrix_norm#Frobenius_norm
- def normINF: Double
Compute the (infinity) INF-norm of 'this' matrix, i.e., the maximum 1-norm of the row vectors.
Compute the (infinity) INF-norm of 'this' matrix, i.e., the maximum 1-norm of the row vectors.
- Definition Classes
- MatriD
- See also
en.wikipedia.org/wiki/Matrix_norm
- final def notify(): Unit
- Definition Classes
- AnyRef
- Annotations
- @native() @HotSpotIntrinsicCandidate()
- final def notifyAll(): Unit
- Definition Classes
- AnyRef
- Annotations
- @native() @HotSpotIntrinsicCandidate()
- def nullspace: VectorD
Compute the (right) nullspace of 'this' 'm-by-n' matrix (requires 'n = m+1') by performing Gauss-Jordan reduction and extracting the negation of the last column augmented by 1.
Compute the (right) nullspace of 'this' 'm-by-n' matrix (requires 'n = m+1') by performing Gauss-Jordan reduction and extracting the negation of the last column augmented by 1.
nullspace (a) = set of orthogonal vectors v s.t. a * v = 0
The left nullspace of matrix 'a' is the same as the right nullspace of 'a.t'. FIX: need a more robust algorithm for computing nullspace (@see Fac_QR.scala). FIX: remove the 'n = m+1' restriction.
- Definition Classes
- SparseMatrixD → MatriD
- See also
http://ocw.mit.edu/courses/mathematics/18-06sc-linear-algebra-fall-2011/ax-b-and-the-four-subspaces
/solving-ax-0-pivot-variables-special-solutions/MIT18_06SCF11_Ses1.7sum.pdf
- def nullspace_ip(): VectorD
Compute in-place the (right) nullspace of 'this' 'm-by-n' matrix (requires 'n = m+1') by performing Gauss-Jordan reduction and extracting the negation of the last column augmented by 1.
Compute in-place the (right) nullspace of 'this' 'm-by-n' matrix (requires 'n = m+1') by performing Gauss-Jordan reduction and extracting the negation of the last column augmented by 1.
nullspace (a) = set of orthogonal vectors v s.t. a * v = 0
The left nullspace of matrix 'a' is the same as the right nullspace of 'a.t'. FIX: need a more robust algorithm for computing nullspace (@see Fac_QR.scala). FIX: remove the 'n = m+1' restriction.
- Definition Classes
- SparseMatrixD → MatriD
- See also
http://ocw.mit.edu/courses/mathematics/18-06sc-linear-algebra-fall-2011/ax-b-and-the-four-subspaces
/solving-ax-0-pivot-variables-special-solutions/MIT18_06SCF11_Ses1.7sum.pdf
- val range1: Range
Range for the storage array on dimension 1 (rows)
Range for the storage array on dimension 1 (rows)
- Definition Classes
- MatriD
- val range2: Range
Range for the storage array on dimension 2 (columns)
Range for the storage array on dimension 2 (columns)
- Definition Classes
- MatriD
- def reduce: SparseMatrixD
Use Gauss-Jordan reduction on 'this' sparse matrix to make the left part embed an identity matrix.
Use Gauss-Jordan reduction on 'this' sparse matrix to make the left part embed an identity matrix. A constraint on this m by n matrix is that n >= m. It can be used to solve 'a * x = b': augment 'a' with 'b' and call reduce. Takes '[a | b]' to '[I | x]'.
- Definition Classes
- SparseMatrixD → MatriD
- def reduce_ip(): SparseMatrixD
Use Gauss-Jordan reduction in-place on 'this' sparse matrix to make the left part embed an identity matrix.
Use Gauss-Jordan reduction in-place on 'this' sparse matrix to make the left part embed an identity matrix. A constraint on this m by n matrix is that n >= m. It can be used to solve 'a * x = b': augment 'a' with 'b' and call reduce. Takes '[a | b]' to '[I | x]'.
- Definition Classes
- SparseMatrixD → MatriD
- def sameCrossDimensions(b: MatriD): Boolean
Check whether 'this' matrix and the other matrix 'b' have the same cross dimensions.
Check whether 'this' matrix and the other matrix 'b' have the same cross dimensions.
- b
the other matrix
- Definition Classes
- MatriD
- def sameDimensions(b: MatriD): Boolean
Check whether 'this' matrix and the other matrix 'b' have the same dimensions.
Check whether 'this' matrix and the other matrix 'b' have the same dimensions.
- b
the other matrix
- Definition Classes
- MatriD
- def selectCols(colIndex: Array[Int]): SparseMatrixD
Select columns from this matrix according to the given index/basis.
Select columns from this matrix according to the given index/basis. Ex: Can be used to divide a matrix into a basis and a non-basis.
- colIndex
the column index positions (e.g., (0, 2, 5))
- Definition Classes
- SparseMatrixD → MatriD
- def selectRows(rowIndex: Array[Int]): SparseMatrixD
Select rows from this matrix according to the given index/basis.
Select rows from this matrix according to the given index/basis.
- rowIndex
the row index positions (e.g., (0, 2, 5))
- Definition Classes
- SparseMatrixD → MatriD
- def selectRows(rowIndex: VectoI): MatriD
Select rows from 'this' matrix according to the given index/basis 'rowIndex'.
Select rows from 'this' matrix according to the given index/basis 'rowIndex'.
- rowIndex
the row index positions (e.g., (0, 2, 5))
- Definition Classes
- MatriD
- def selectRowsEx(rowIndex: VectoI): MatriD
Select all rows from 'this' matrix excluding the rows from the given 'rowIndex'.
Select all rows from 'this' matrix excluding the rows from the given 'rowIndex'.
- rowIndex
the row indices to exclude
- Definition Classes
- MatriD
- def selectRowsEx(rowIndex: Array[Int]): MatriD
Select all rows from 'this' matrix excluding the rows from the given 'rowIndex'.
Select all rows from 'this' matrix excluding the rows from the given 'rowIndex'.
- rowIndex
the row indices to exclude
- Definition Classes
- MatriD
- def set(i: Int, u: VectoD, j: Int = 0): Unit
Set this matrix's 'i'th row starting at column 'j' to the vector 'u'.
Set this matrix's 'i'th row starting at column 'j' to the vector 'u'.
- i
the row index
- u
the vector value to assign
- j
the starting column index
- Definition Classes
- SparseMatrixD → MatriD
- def set(u: MatriD): Unit
Set the values in 'this' matrix as copies of the values in matrix 'u'.
Set the values in 'this' matrix as copies of the values in matrix 'u'.
- u
the matrix of values to assign
- Definition Classes
- SparseMatrixD → MatriD
- def set(u: Array[Array[Double]]): Unit
Set all the values in this matrix as copies of the values in 2D array 'u'.
Set all the values in this matrix as copies of the values in 2D array 'u'.
- u
the 2D array of values to assign
- Definition Classes
- SparseMatrixD → MatriD
- def set(x: Double): Unit
Set all the elements in this matrix to the scalar 'x'.
Set all the elements in this matrix to the scalar 'x'.
- x
the scalar value to assign
- Definition Classes
- SparseMatrixD → MatriD
- def setCol(col: Int, u: VectoD): Unit
Set column 'col' of the matrix to a vector.
Set column 'col' of the matrix to a vector.
- col
the column to set
- u
the vector to assign to the column
- Definition Classes
- SparseMatrixD → MatriD
- def setDiag(x: Double): Unit
Set the main diagonal of this matrix to the scalar 'x'.
Set the main diagonal of this matrix to the scalar 'x'. Assumes 'dim2 >= dim1'.
- x
the scalar to set the diagonal to
- Definition Classes
- SparseMatrixD → MatriD
- def setDiag(u: VectoD, k: Int = 0): Unit
Set the 'k'th diagonal of this matrix to the vector 'u'.
Set the 'k'th diagonal of this matrix to the vector 'u'. Assumes 'dim2 >= dim1'.
- u
the vector to set the diagonal to
- k
how far above the main diagonal, e.g., (-1, 0, 1) for (sub, main, super)
- Definition Classes
- SparseMatrixD → MatriD
- def setFormat(newFormat: String): Unit
Set the format to the 'newFormat'.
- def showAll(): Unit
Show all elements in 'this' sparse matrix.
- def slice(r_from: Int, r_end: Int, c_from: Int, c_end: Int): SparseMatrixD
Slice 'this' sparse matrix row-wise 'r_from' to 'r_end' and column-wise 'c_from' to 'c_end'.
Slice 'this' sparse matrix row-wise 'r_from' to 'r_end' and column-wise 'c_from' to 'c_end'.
- r_from
the start of the row slice
- r_end
the end of the row slice
- c_from
the start of the column slice
- c_end
the end of the column slice
- Definition Classes
- SparseMatrixD → MatriD
- def slice(from: Int, end: Int): SparseMatrixD
Slice 'this' sparse matrix row-wise 'from' to 'end'.
Slice 'this' sparse matrix row-wise 'from' to 'end'.
- from
the start row of the slice
- end
the end row of the slice
- Definition Classes
- SparseMatrixD → MatriD
- def slice(rg: Range): MatriD
Slice 'this' matrix row-wise over the given range 'rg'.
Slice 'this' matrix row-wise over the given range 'rg'.
- rg
the range specifying the slice
- Definition Classes
- MatriD
- def sliceCol(from: Int, end: Int): SparseMatrixD
Slice 'this' sparse matrix column-wise 'from' to 'end'.
Slice 'this' sparse matrix column-wise 'from' to 'end'.
- from
the start column of the slice (inclusive)
- end
the end column of the slice (exclusive)
- Definition Classes
- SparseMatrixD → MatriD
- def sliceEx(row: Int, col: Int): SparseMatrixD
Slice 'this' sparse matrix excluding the given row and column.
Slice 'this' sparse matrix excluding the given row and column.
- row
the row to exclude
- col
the column to exclude
- Definition Classes
- SparseMatrixD → MatriD
- def sliceEx(rg: Range): MatriD
Slice 'this' matrix row-wise excluding the given range 'rg'.
Slice 'this' matrix row-wise excluding the given range 'rg'.
- rg
the excluded range of the slice
- Definition Classes
- MatriD
- def solve(b: VectoD): VectoD
Solve for 'x' in the equation 'a*x = b' where 'a' is 'this' matrix.
Solve for 'x' in the equation 'a*x = b' where 'a' is 'this' matrix.
- b
the constant vector.
- Definition Classes
- SparseMatrixD → MatriD
- def solve(l: MatriD, u: MatriD, b: VectoD): VectoD
Solve for 'x' in the equation 'l*u*x = b' (see 'lud_npp' above).
Solve for 'x' in the equation 'l*u*x = b' (see 'lud_npp' above).
- l
the lower triangular matrix
- u
the upper triangular matrix
- b
the constant vector
- Definition Classes
- SparseMatrixD → MatriD
- def solve(lu: (MatriD, MatriD), b: VectoD): VectoD
Solve for 'x' in the equation 'l*u*x = b' (see 'lud' above).
Solve for 'x' in the equation 'l*u*x = b' (see 'lud' above).
- lu
the lower and upper triangular matrices
- b
the constant vector
- Definition Classes
- MatriD
- def splitRows(rowIndex: VectoI): (MatriD, MatriD)
Split the rows from 'this' matrix to form two matrices, one from the rows in 'rowIndex' and the other from rows not in 'rowIndex'.
Split the rows from 'this' matrix to form two matrices, one from the rows in 'rowIndex' and the other from rows not in 'rowIndex'.
- rowIndex
the row indices to include/exclude
- Definition Classes
- MatriD
- def splitRows(rowIndex: Array[Int]): (MatriD, MatriD)
Split the rows from 'this' matrix to form two matrices, one from the rows in 'rowIndex' and the other from rows not in 'rowIndex'.
Split the rows from 'this' matrix to form two matrices, one from the rows in 'rowIndex' and the other from rows not in 'rowIndex'.
- rowIndex
the row indices to include/exclude
- Definition Classes
- MatriD
- def sum: Double
Compute the sum of 'this' sparse matrix, i.e., the sum of its elements.
Compute the sum of 'this' sparse matrix, i.e., the sum of its elements.
- Definition Classes
- SparseMatrixD → MatriD
- def sumAbs: Double
Compute the 'abs' sum of this matrix, i.e., the sum of the absolute value of its elements.
Compute the 'abs' sum of this matrix, i.e., the sum of the absolute value of its elements. This is useful for comparing matrices '(a - b).sumAbs'.
- Definition Classes
- SparseMatrixD → MatriD
- def sumLower: Double
Compute the sum of the lower triangular region of 'this' sparse matrix.
Compute the sum of the lower triangular region of 'this' sparse matrix.
- Definition Classes
- SparseMatrixD → MatriD
- def swap(i: Int, k: Int, col: Int = 0): Unit
Swap the elements in rows 'i' and 'k' starting from column 'col'.
Swap the elements in rows 'i' and 'k' starting from column 'col'.
- i
the first row in the swap
- k
the second row in the swap
- col
the starting column for the swap (default 0 => whole row)
- Definition Classes
- MatriD
- def swapCol(j: Int, l: Int, row: Int = 0): Unit
Swap the elements in columns 'j' and 'l' starting from row 'row'.
Swap the elements in columns 'j' and 'l' starting from row 'row'.
- j
the first column in the swap
- l
the second column in the swap
- row
the starting row for the swap (default 0 => whole column)
- Definition Classes
- MatriD
- final def synchronized[T0](arg0: => T0): T0
- Definition Classes
- AnyRef
- def t: SparseMatrixD
Transpose 'this' sparse matrix (rows => columns).
Transpose 'this' sparse matrix (rows => columns).
- Definition Classes
- SparseMatrixD → MatriD
- def times_s(b: SparseMatrixD): SparseMatrixD
Multiply 'this' sparse matrix by sparse matrix 'b' using the Strassen matrix multiplication algorithm.
Multiply 'this' sparse matrix by sparse matrix 'b' using the Strassen matrix multiplication algorithm. Both matrices ('this' and 'b') must be square. Although the algorithm is faster than the traditional cubic algorithm, its requires more memory and is often less stable (due to round-off errors). FIX: could be make more efficient using a virtual slice 'vslice' method.
- b
the matrix to multiply by (it has to be a square matrix)
- See also
http://en.wikipedia.org/wiki/Strassen_algorithm
- def toDense: MatrixD
Convert this sparse matrix to a dense matrix.
Convert this sparse matrix to a dense matrix. FIX - new builder
- Definition Classes
- SparseMatrixD → MatriD
- def toDouble: MatrixD
Convert 'this'
SparseMatrixDinto a dense double matrixMatrixD.Convert 'this'
SparseMatrixDinto a dense double matrixMatrixD.- Definition Classes
- SparseMatrixD → MatriD
- def toInt: MatrixI
Convert 'this'
SparseMatrixDinto a dense integer matrixMatrixI.Convert 'this'
SparseMatrixDinto a dense integer matrixMatrixI.- Definition Classes
- SparseMatrixD → MatriD
- def toString(): String
Show the non-zero elements in 'this' sparse matrix.
Show the non-zero elements in 'this' sparse matrix.
- Definition Classes
- SparseMatrixD → AnyRef → Any
- def trace: Double
Compute the trace of 'this' sparse matrix, i.e., the sum of the elements on the main diagonal.
Compute the trace of 'this' sparse matrix, i.e., the sum of the elements on the main diagonal. Should also equal the sum of the eigenvalues.
- Definition Classes
- SparseMatrixD → MatriD
- See also
Eigen.scala
- def update(ir: Range, jr: Range, b: MatriD): Unit
Set a slice 'this' sparse matrix row-wise on range 'ir' and column-wise on range 'jr'.
Set a slice 'this' sparse matrix row-wise on range 'ir' and column-wise on range 'jr'. Ex: a(2..4, 3..5) = b
- ir
the row range
- jr
the column range
- b
the matrix to assign
- Definition Classes
- SparseMatrixD → MatriD
- def update(i: Int, u: TreeMap[Int, Double]): Unit
Set 'this' sparse matrix's row at the 'i'-th index position to the sorted-linked-map 'u'.
Set 'this' sparse matrix's row at the 'i'-th index position to the sorted-linked-map 'u'.
- i
the row index
- u
the sorted-linked-map of non-zero values to assign
- def update(i: Int, u: VectoD): Unit
Set 'this' sparse matrix's row at the i-th index position to the vector 'u'.
Set 'this' sparse matrix's row at the i-th index position to the vector 'u'.
- i
the row index
- u
the vector value to assign
- Definition Classes
- SparseMatrixD → MatriD
- def update(i: Int, j: Int, x: Double): Unit
Set 'this' sparse matrix's element at the 'i,j'-th index position to the scalar 'x'.
Set 'this' sparse matrix's element at the 'i,j'-th index position to the scalar 'x'. Only store 'x' if it is non-zero.
- i
the row index
- j
the column index
- x
the scalar value to assign
- Definition Classes
- SparseMatrixD → MatriD
- def update(i: Int, jr: Range, u: VectoD): Unit
Set a slice of 'this' matrix row-wise at index 'i' and column-wise on range 'jr' to vector 'u'.
Set a slice of 'this' matrix row-wise at index 'i' and column-wise on range 'jr' to vector 'u'. Ex: a(2, 3..5) = u
- i
the row index
- jr
the column range
- u
the vector to assign
- Definition Classes
- MatriD
- def update(ir: Range, j: Int, u: VectoD): Unit
Set a slice of 'this' matrix row-wise on range 'ir' and column-wise at index 'j' to vector 'u'.
Set a slice of 'this' matrix row-wise on range 'ir' and column-wise at index 'j' to vector 'u'. Ex: a(2..4, 3) = u
- ir
the row range
- j
the column index
- u
the vector to assign
- Definition Classes
- MatriD
- def upperT: SparseMatrixD
Return the upper triangular of 'this' matrix (rest are zero).
Return the upper triangular of 'this' matrix (rest are zero).
- Definition Classes
- SparseMatrixD → MatriD
- final def wait(arg0: Long, arg1: Int): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws(classOf[java.lang.InterruptedException])
- final def wait(arg0: Long): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws(classOf[java.lang.InterruptedException]) @native()
- final def wait(): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws(classOf[java.lang.InterruptedException])
- def write(fileName: String): Unit
Write 'this' matrix to a CSV-formatted text file with name 'fileName'.
Write 'this' matrix to a CSV-formatted text file with name 'fileName'.
- fileName
the name of file to hold the data
- Definition Classes
- SparseMatrixD → MatriD
- def zero(m: Int = dim1, n: Int = dim2): SparseMatrixD
Create an 'm-by-n' sparse matrix with all elements initialized to zero.
Create an 'm-by-n' sparse matrix with all elements initialized to zero.
- m
the number of rows
- n
the number of columns
- Definition Classes
- SparseMatrixD → MatriD
- def ~^(p: Int): SparseMatrixD
Raise 'this' sparse matrix to the 'p'th power (for some integer 'p' >= 2).
Raise 'this' sparse matrix to the 'p'th power (for some integer 'p' >= 2). Caveat: should be replace by a divide and conquer algorithm.
- p
the power to raise this matrix to
- Definition Classes
- SparseMatrixD → MatriD
Deprecated Value Members
- def finalize(): Unit
- Attributes
- protected[lang]
- Definition Classes
- AnyRef
- Annotations
- @throws(classOf[java.lang.Throwable]) @Deprecated
- Deprecated