Class Matrix<T>

Matrix class

Type Parameters

  • T = number

Constructors

constructor

Accessors

cols dimension length rows sizes t value

Methods

[iterator] abs add addAt adjoint and andAt argmax argmin at balancing balancingSinkhornKnopp bidiag bidiagHouseholder bitand bitandAt bitnot bitor bitorAt bitxor bitxorAt block broadcastOperate ceil cholesky choleskyBanachiewicz choleskyCrout choleskyGaussian col concat convolute copy cov det diag div divAt dot eigen eigenInverseIteration eigenJacobi eigenPowerIteration eigenSimultaneousIteration eigenValues eigenValuesBiSection eigenValuesLR eigenValuesQR eigenVectors equals every exp fill flip floor forEach gram hessenberg hessenbergArnoldi idiv idivAt imod imodAt inv invLowerTriangular invLU invRowReduction invUpperTriangular isAlternating isDiag isHermitian isIdentity isLowerTriangular isNilpotent isNormal isOrthogonal isRegular isSkewHermitian isSquare isSymmetric isTriangular isub isubAt isUnitary isUpperTriangular isZero kron leftShift log lu map max mean median min mod modAt modifiedCholesky mult multAt negative norm normEntrywise normFrobenius normInduced normMax normNuclear normSchatten normSpectral not operateAt or orAt power prod pseudoInv pseudoInvBenIsraelCohen pseudoInvNaive pseudoInvQR pseudoInvSVD qr qrGramSchmidt qrHouseholder quantile rank rankFactorization reduce reducedRowEchelonForm remove removeIf repeat reshape resize round row sample schur schurQR set shuffle signedRightShift singularValuePowerIteration singularValues slice solve solveGaussSeidel solveJacobi solveLowerTriangular solveSOR solveUpperTriangular some sort spectralRadius sqrt std sub subAt sum svd svdEigen svdGolubKahan swap tDot toArray toScaler toString trace transpose tridiag tridiagHouseholder tridiagLanczos unique unsignedRightShift variance add and bitand bitor bitxor concat diag div eye fromArray map mod mult ones or randint randn random repeat resize sub zeros

Constructors

constructor

  • new Matrix<T = number>(
        rows: number,
        cols: number,
        values?: T | T[] | T[][],
    ): Matrix<T>

  • Type Parameters

    • T = number

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    • Optionalvalues: T | T[] | T[][]

      Initial values

    Returns Matrix<T>

  • new Matrix<T = number>(
        size: [number, number],
        values?: T | T[] | T[][],
    ): Matrix<T>

  • Type Parameters

    • T = number

    Parameters

    • size: [number, number]

      Sizes for each dimension

    • Optionalvalues: T | T[] | T[][]

      Initial values

    Returns Matrix<T>

Accessors

cols

  • get cols(): number

  • Number of columns of the matrix.

    Returns number

dimension

  • get dimension(): number

  • Dimension of the matrix.

    Returns number

length

  • get length(): number

  • Number of all elements in the matrix.

    Returns number

rows

  • get rows(): number

  • Number of rows of the matrix.

    Returns number

sizes

  • get sizes(): number[]

  • Sizes of the matrix.

    Returns number[]

t

  • get t(): Matrix<T>

  • Transpose matrix.

    Returns Matrix<T>

value

  • get value(): T[]

  • Elements in the matrix.

    Returns T[]

Methods

[iterator]

  • "[iterator]"(): Generator<T>

  • Iterate over the elements.

    Returns Generator<T>

abs

  • abs(): void

  • Set all elements to their absolute values.

    Returns void

add

  • add(o: number | Tensor | Matrix): void

  • Add a value or matrix.

    Parameters

    Returns void

addAt

  • addAt(r: number, c: number, v: number): number

  • Add a value to the position.

    Parameters

    • r: number

      Index of the row to add the value to

    • c: number

      Index of the column to add the value to

    • v: number

      Value to add

    Returns number

    Old value

adjoint

  • adjoint(): Matrix<T>

  • Returns adjoint matrix.

    Returns Matrix<T>

    Adjoint matrix

and

  • and(o: number | Tensor | Matrix<unknown>): void

  • Take a logical AND with a value or matrix.

    Parameters

    • o: number | Tensor | Matrix<unknown>

      Value to take a logical AND

    Returns void

andAt

  • andAt(r: number, c: number, v: number): T

  • Take logical AND with a value to the position.

    Parameters

    • r: number

      Index of the row to take a logical AND with

    • c: number

      Index of the column to take a logical AND with

    • v: number

      Value to take a logical AND

    Returns T

    Old value

argmax

  • argmax(axis: number): Matrix

  • Returns maximum indexes along the axis.

    Parameters

    • axis: number

      Axis to be reduced

    Returns Matrix

    Argmax values

argmin

  • argmin(axis: number): Matrix

  • Returns minimum indexes along the axis.

    Parameters

    • axis: number

      Axis to be reduced

    Returns Matrix

    Argmin values

at

  • at(r: number, c: number): T

  • Returns a value at the position.

    Parameters

    • r: number

      Row index

    • c: number

      Column index

    Returns T

    Value at the position

  • at(index: [number, number]): T

  • Returns a value at the position.

    Parameters

    • index: [number, number]

      Index values

    Returns T

    Value at the position

balancing

  • balancing(): [number[], Matrix, number[]]

  • Returns a doubly stochastic matrix.

    Returns [number[], Matrix, number[]]

    Doubly stochastic matrix

balancingSinkhornKnopp

  • balancingSinkhornKnopp(): [number[], Matrix, number[]]

  • Returns a doubly stochastic matrix by Sinkhorn-Knopp algorithm.

    Returns [number[], Matrix, number[]]

    Doubly stochastic matrix

bidiag

  • bidiag(): Matrix

  • Returns a bidiagonal matrix.

    Returns Matrix

    Bidiagonal matrix

bidiagHouseholder

  • bidiagHouseholder(return_uv?: false): Matrix

  • Returns a bidiagonal matrix by Householder method.

    Parameters

    • Optionalreturn_uv: false

      Returns orthogonal matrixes

    Returns Matrix

    Bidiagonal matrix, or Bidiagonal matrix and orthogonal matrixes

  • bidiagHouseholder(return_uv: true): [Matrix, Matrix, Matrix]

  • Returns a bidiagonal matrix by Householder method.

    Parameters

    • return_uv: true

      Returns orthogonal matrixes

    Returns [Matrix, Matrix, Matrix]

    Bidiagonal matrix, or Bidiagonal matrix and orthogonal matrixes

bitand

  • bitand(o: number | Tensor | Matrix): void

  • Take a bitwise AND with a value or matrix.

    Parameters

    Returns void

bitandAt

  • bitandAt(r: number, c: number, v: number): T

  • Take bitwise AND with a value to the position.

    Parameters

    • r: number

      Index of the row to take a bitwise AND with

    • c: number

      Index of the column to take a bitwise AND with

    • v: number

      Value to take a bitwise AND

    Returns T

    Old value

bitnot

  • bitnot(): void

  • Set all elements to their bitwise NOT values.

    Returns void

bitor

  • bitor(o: number | Tensor | Matrix): void

  • Take a bitwise OR with a value or matrix.

    Parameters

    Returns void

bitorAt

  • bitorAt(r: number, c: number, v: number): T

  • Take bitwise OR with a value to the position.

    Parameters

    • r: number

      Index of the row to take a bitwise OR with

    • c: number

      Index of the column to take a bitwise OR with

    • v: number

      Value to take a bitwise OR

    Returns T

    Old value

bitxor

  • bitxor(o: number | Tensor | Matrix): void

  • Take a bitwise XOR with a value or matrix.

    Parameters

    Returns void

bitxorAt

  • bitxorAt(r: number, c: number, v: number): T

  • Take bitwise XOR with a value to the position.

    Parameters

    • r: number

      Index of the row to take a bitwise XOR with

    • c: number

      Index of the column to take a bitwise XOR with

    • v: number

      Value to take a bitwise XOR

    Returns T

    Old value

block

  • block(
        rows_from?: number,
        cols_from?: number,
        rows_to?: number,
        cols_to?: number,
    ): Matrix<T>

  • Returns the sub-matrix corresponding to position.

    Parameters

    • Optionalrows_from: number

      Start row index

    • Optionalcols_from: number

      Start column index

    • Optionalrows_to: number

      End row index(exclusive)

    • Optionalcols_to: number

      End column index(exclusive)

    Returns Matrix<T>

    Sub matrix

broadcastOperate

  • broadcastOperate<U_1>(
        o: Tensor | U_1 | Matrix<U_1>,
        fn: (arg0: T, arg1: U_1) => T,
    ): void

  • Apply function for all elements with broadcasting.

    Type Parameters

    • U_1

    Parameters

    Returns void

ceil

  • ceil(): void

  • Set all elements to their ceil values.

    Returns void

cholesky

  • cholesky(): Matrix

  • Returns a cholesky decomposition.

    Returns Matrix

    Cholesky decomposition matrix

choleskyBanachiewicz

  • choleskyBanachiewicz(): Matrix

  • Returns a cholesky decomposition by Banachiewicz algorithm.

    Returns Matrix

    Cholesky decomposition matrix

choleskyCrout

  • choleskyCrout(): Matrix

  • Returns a cholesky decomposition by Crout algorithm.

    Returns Matrix

    Cholesky decomposition matrix

choleskyGaussian

  • choleskyGaussian(): Matrix

  • Returns a cholesky decomposition by Gaussian algorithm.

    Returns Matrix

    Cholesky decomposition matrix

col

  • col(c: number | number[] | boolean[]): Matrix<T>

  • Returns a col matrix at c.

    Parameters

    • c: number | number[] | boolean[]

      Indexes of columns, or an array of boolean values where the column to be selected is true.

    Returns Matrix<T>

    Column selected matrix

concat

  • concat(m: Matrix<T>, axis?: number): void

  • Concatenate this and m.

    Parameters

    • m: Matrix<T>

      Concatenate matrix

    • Optionalaxis: number

      Axis to be concatenated

    Returns void

convolute

  • convolute(kernel: number[][], normalize?: boolean): void

  • Convoluted by a kernel.

    Parameters

    • kernel: number[][]

      Kernel matrix

    • Optionalnormalize: boolean

      Normalize kernel or not

    Returns void

copy

  • copy(dst?: Matrix<T>): Matrix<T>

  • Returns a copy of this matrix.

    Parameters

    • Optionaldst: Matrix<T>

      Destination matrix

    Returns Matrix<T>

    Copied matrix

cov

  • cov(ddof?: number): Matrix

  • Returns a covariance matrix.

    Parameters

    • Optionalddof: number

      Delta Degrees of Freedom

    Returns Matrix

    Covariance matrix

det

  • det(): number

  • Returns a determinant.

    Returns number

    Determinant value

diag

  • diag(): number[]

  • Returns diagonal elements.

    Returns number[]

    Diagonal values

div

  • div(o: number | Tensor | Matrix): void

  • Divides by a value element-wise.

    Parameters

    Returns void

divAt

  • divAt(r: number, c: number, v: number): number

  • Divides the value at the position by a value.

    Parameters

    • r: number

      Index of the row to divide the value by

    • c: number

      Index of the column to divide the value by

    • v: number

      Value to divide

    Returns number

    Old value

dot

  • dot(o: Matrix): Matrix

  • Returns a matrix product value.

    Parameters

    Returns Matrix

    Producted matrix

eigen

  • eigen(): [number[], Matrix]

  • Returns eigenvalues and eigenvectors.

    Returns [number[], Matrix]

    Eigenvalues and eigenvectors. Eigenvectors correspond to each column of the matrix.

eigenInverseIteration

  • eigenInverseIteration(ev?: number, maxIteration?: number): [number, Matrix]

  • Returns the nearest eigenvalue and its eigenvector to the specified value.

    Parameters

    • Optionalev: number

      Target value

    • OptionalmaxIteration: number

      Maximum iteration

    Returns [number, Matrix]

    Eigenvalue and eigenvector

eigenJacobi

  • eigenJacobi(maxIteration?: number): [number[], Matrix]

  • Returns eigenvalues and eigenvectors.

    Parameters

    • OptionalmaxIteration: number

      Maximum iteration

    Returns [number[], Matrix]

    Eigenvalues and eigenvectors. Eigenvectors correspond to each column of the matrix.

eigenPowerIteration

  • eigenPowerIteration(maxIteration?: number): [number, Matrix]

  • Returns the maximum eigenvalue and its eigenvector.

    Parameters

    • OptionalmaxIteration: number

      Maximum iteration

    Returns [number, Matrix]

    Maximum eigenvalue and its eigenvector

eigenSimultaneousIteration

  • eigenSimultaneousIteration(k: number, maxIteration?: number): [number[], Matrix]

  • Returns the k highest eigenvalues and its eigenvectors.

    Parameters

    • k: number

      Number of calculated eigenvalues

    • OptionalmaxIteration: number

      Maximum iteration

    Returns [number[], Matrix]

    The k highest eigenvalues and its eigenvectors. Eigenvectors correspond to each column of the matrix.

eigenValues

  • eigenValues(): number[]

  • Returns eigenvalues.

    Returns number[]

    Eigenvalues

eigenValuesBiSection

  • eigenValuesBiSection(): number[]

  • Returns eigenvalues by Bi-section.

    Returns number[]

    Eigenvalues

eigenValuesLR

  • eigenValuesLR(maxIteration?: number): number[]

  • Returns eigenvalues by LU decomposition.

    Parameters

    • OptionalmaxIteration: number

      Maximum iteration

    Returns number[]

    Eigenvalues

eigenValuesQR

  • eigenValuesQR(maxIteration?: number): number[]

  • Returns eigenvalues by QR decomposition.

    Parameters

    • OptionalmaxIteration: number

      Maximum iteration

    Returns number[]

    Eigenvalues

eigenVectors

  • eigenVectors(): Matrix

  • Returns eigenvectors.

    Returns Matrix

    Eigenvectors. Eigenvectors correspond to each column of the matrix.

equals

  • equals(other: any, tol?: number): boolean

  • Returns this matrix is equals to the others.

    Parameters

    • other: any

      Check tensor

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns boolean

    true if equal

every

  • every(cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean): boolean

  • Determines whether all the members of a matrix satisfy the specified test.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    Returns boolean

    Reduced value or matrix

  • every(
        cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean,
        axis: 0 | 1,
    ): Matrix<boolean>

  • Determines whether all the members of a matrix satisfy the specified test.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix<boolean>

    Reduced value or matrix

  • every(
        cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean,
        axis: number,
    ): boolean | Matrix<boolean>

  • Determines whether all the members of a matrix satisfy the specified test.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    • axis: number

      Axis to be reduced

    Returns boolean | Matrix<boolean>

    Reduced value or matrix

exp

  • exp(): Matrix

  • Returns a exponential matrix

    Returns Matrix

    Exponential matrix

fill

  • fill(value: T): void

  • Fill in all the elements with the value.

    Parameters

    • value: T

      Filled value

    Returns void

flip

  • flip(axis?: number): void

  • Flip values along the axis.

    Parameters

    • Optionalaxis: number

      Axis to be flipped

    Returns void

floor

  • floor(): void

  • Set all elements to their floored values.

    Returns void

forEach

  • forEach(cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => any): void

  • Iterate over all the elements.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => any

      Callback function

    Returns void

gram

  • gram(): Matrix

  • Returns a gram matrix.

    Returns Matrix

    Gram matrix

hessenberg

  • hessenberg(): Matrix

  • Returns a hessenberg matrix.

    Returns Matrix

    Hessenberg matrix

hessenbergArnoldi

  • hessenbergArnoldi(k?: number): Matrix

  • Returns a hessenberg matrix.

    Parameters

    • Optionalk: number

      Number of iterations

    Returns Matrix

    Hessenberg matrix

idiv

  • idiv(o: number | Tensor | Matrix): void

  • Divides a value by this matrix element-wise.

    Parameters

    Returns void

idivAt

  • idivAt(r: number, c: number, v: number): number

  • Divides a value by the value at the position.

    Parameters

    • r: number

      Index of the row whose value is to be divided

    • c: number

      Index of the column whose value is to be divided

    • v: number

      Value to be divided

    Returns number

    Old value

imod

  • imod(o: number | Tensor | Matrix): void

  • Take a remainder divided a value by this matrix element-wise.

    Parameters

    Returns void

imodAt

  • imodAt(r: number, c: number, v: number): number

  • Take a remainder divided a value by the value at the position.

    Parameters

    • r: number

      Index of the row whose value is to be divided

    • c: number

      Index of the column whose value is to be divided

    • v: number

      Value to be divided

    Returns number

    Old value

inv

  • inv(): Matrix

  • Returns a inverse matrix.

    Returns Matrix

    Inversed matrix

invLowerTriangular

  • invLowerTriangular(): Matrix

  • Returns a inverse matrix for lower triangular matrix.

    Returns Matrix

    Inversed matrix

invLU

  • invLU(): Matrix

  • Returns a inverse matrix by LU decompositioin.

    Returns Matrix

    Inversed matrix

invRowReduction

  • invRowReduction(): Matrix

  • Returns a inverse matrix by row reduction.

    Returns Matrix

    Inversed matrix

invUpperTriangular

  • invUpperTriangular(): Matrix

  • Returns a inverse matrix for upper triangular matrix.

    Returns Matrix

    Inversed matrix

isAlternating

  • isAlternating(tol?: number): boolean

  • Returns if this is alternating matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance within which sign-reversed values are recognized as the same

    Returns boolean

    true if this is alternating matrix

isDiag

  • isDiag(tol?: number): boolean

  • Returns if this is diagonal matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is diagonal matrix

isHermitian

  • isHermitian(tol?: number): boolean

  • Returns if this is hermitian matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns boolean

    true if this is hermitian matrix

isIdentity

  • isIdentity(tol?: number): boolean

  • Returns if this is identity matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0 or 1

    Returns boolean

    true if this is identity matrix

isLowerTriangular

  • isLowerTriangular(tol?: number): boolean

  • Returns if this is lower triangular matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is lower triangular matrix

isNilpotent

  • isNilpotent(tol?: number): boolean

  • Returns if this is nilpotent matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is nilpotent matrix

isNormal

  • isNormal(tol?: number): boolean

  • Returns if this is normal matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns boolean

    true if this is normal matrix

isOrthogonal

  • isOrthogonal(tol?: number): boolean

  • Returns if this is orthogonal matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0 or 1

    Returns boolean

    true if this is orthogonal matrix

isRegular

  • isRegular(tol?: number): boolean

  • Returns if this is regular matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to recognize the determinant as 0

    Returns boolean

    true if this is regular matrix

isSkewHermitian

  • isSkewHermitian(tol?: number): boolean

  • Returns if this is skew-hermitian matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance within which sign-reversed values are recognized as the same

    Returns boolean

    true if this is skew-hermitian matrix

isSquare

  • isSquare(): boolean

  • Returns if this is square matrix or not.

    Returns boolean

    true if this is square matrix

isSymmetric

  • isSymmetric(tol?: number): boolean

  • Returns if this is symmetric matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns boolean

    true if this is symmetric matrix

isTriangular

  • isTriangular(tol?: number): boolean

  • Returns if this is triangular matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is triangular matrix

isub

  • isub(o: number | Tensor | Matrix): void

  • Subtract this matrix from a value or matrix.

    Parameters

    Returns void

isubAt

  • isubAt(r: number, c: number, v: number): number

  • Subtract the value at the position from a value.

    Parameters

    • r: number

      Index of the row whose value is to be subtracted

    • c: number

      Index of the column whose value is to be subtracted

    • v: number

      Value to be subtracted

    Returns number

    Old value

isUnitary

  • isUnitary(tol?: number): boolean

  • Returns if this is unitary matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0 or 1

    Returns boolean

    true if this is unitary matrix

isUpperTriangular

  • isUpperTriangular(tol?: number): boolean

  • Returns if this is upper triangular matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is upper triangular matrix

isZero

  • isZero(tol?: number): boolean

  • Returns if this is zero matrix or not.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns boolean

    true if this is zero matrix

kron

  • kron(mat: Matrix): Matrix

  • Returns kronecker producted value.

    Parameters

    Returns Matrix

    Kronecker producted matrix

leftShift

  • leftShift(n: number): void

  • Set all elements to their left shift values.

    Parameters

    • n: number

      Shift amount

    Returns void

log

  • log(): Matrix

  • Returns a logarithm matrix

    Returns Matrix

    Logarithm matrix

lu

  • lu(): [Matrix, Matrix]

  • Returns a LU decomposition.

    Returns [Matrix, Matrix]

    Lower triangular matrix and upper triangular matrix

map

  • map(cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => T): void

  • Iterate over all the elements and replace the value.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => T

      Mapping function

    Returns void

max

  • max(): number

  • Returns maximum value of all element.

    Returns number

    Maximum value

  • max(axis: 0 | 1): Matrix

  • Returns maximum values along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Maximum values

  • max(axis: number): number | Matrix

  • Returns maximum values along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns the maximum value of the all element.

    Returns number | Matrix

    Maximum values

mean

  • mean(): number

  • Returns mean of all element.

    Returns number

    Mean value

  • mean(axis: 0 | 1): Matrix

  • Returns means along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Mean values

  • mean(axis: number): number | Matrix

  • Returns means along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a mean value of the all element.

    Returns number | Matrix

    Mean values

median

  • median(): number

  • Returns median of all element.

    Returns number

    Median value

  • median(axis: 0 | 1): Matrix

  • Returns medians along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Median values

  • median(axis: number): number | Matrix

  • Returns medians along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a median of the all element.

    Returns number | Matrix

    Median values

min

  • min(): number

  • Returns minimum value of all element.

    Returns number

    Minimum value

  • min(axis: 0 | 1): Matrix

  • Returns minimum values along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Minimum values

  • min(axis: number): number | Matrix

  • Returns minimum values along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns the minimum value of the all element.

    Returns number | Matrix

    Minimum values

mod

  • mod(o: number | Tensor | Matrix): void

  • Take a remainder divided by a value element-wise.

    Parameters

    Returns void

modAt

  • modAt(r: number, c: number, v: number): number

  • Take a remainder divided the value at the position by a value.

    Parameters

    • r: number

      Index of the row to divide the value by

    • c: number

      Index of the column to divide the value by

    • v: number

      Value to divide

    Returns number

    Old value

modifiedCholesky

  • modifiedCholesky(): [Matrix, number[]]

  • Returns a modified cholosky decomposition.

    Returns [Matrix, number[]]

    Cholesky decomposition matrix and diagonal matrix

mult

  • mult(o: number | Tensor | Matrix): void

  • Multiplies by a value element-wise.

    Parameters

    Returns void

multAt

  • multAt(r: number, c: number, v: number): number

  • Multiplies a value to the position.

    Parameters

    • r: number

      Index of the row to multiply the value by

    • c: number

      Index of the column to multiply the value by

    • v: number

      Value to multiply

    Returns number

    Old value

negative

  • negative(): void

  • Multiply all elements by -1 in-place.

    Returns void

norm

  • norm(p?: number): number

  • Returns a p-norm.

    Parameters

    • Optionalp: number

      p-norm

    Returns number

    Entry-wise norm

normEntrywise

  • normEntrywise(p?: number): number

  • Returns a entry-wise norm

    Parameters

    • Optionalp: number

      p-norm

    Returns number

    Entry-wise norm

normFrobenius

  • normFrobenius(): number

  • Returns frobenius norm.

    Returns number

    Frobenius norm

normInduced

  • normInduced(p?: number): number

  • Returns induced norm.

    Parameters

    • Optionalp: number

      p-norm

    Returns number

    Induced norm

normMax

  • normMax(): number

  • Returns max norm.

    Returns number

    Max norm

normNuclear

  • normNuclear(): number

  • Returns nuclear norm.

    Returns number

    Nuclear norm

normSchatten

  • normSchatten(p?: number): number

  • Returns schatten norm.

    Parameters

    • Optionalp: number

      p-norm

    Returns number

    Schatten norm

normSpectral

  • normSpectral(): number

  • Returns spectral norm.

    Returns number

    Spectral norm

not

  • not(): void

  • Set all elements to their logical NOT values.

    Returns void

operateAt

  • operateAt(r: number, c: number, fn?: (arg0: T) => T): T

  • Apply function to the position.

    Parameters

    • r: number

      Index of the row to apply function to

    • c: number

      Index of the column to apply function to

    • Optionalfn: (arg0: T) => T

      Applied function

    Returns T

    Old value

  • operateAt(index: [number, number], fn: (arg0: T) => T): T

  • Apply function to the position.

    Parameters

    • index: [number, number]

      Index to apply function to

    • fn: (arg0: T) => T

      Applied function

    Returns T

    Old value

or

  • or(o: number | Tensor | Matrix<unknown>): void

  • Take a logical OR with a value or matrix.

    Parameters

    • o: number | Tensor | Matrix<unknown>

      Value to take a logical OR

    Returns void

orAt

  • orAt(r: number, c: number, v: number): T

  • Take logical OR with a value to the position.

    Parameters

    • r: number

      Index of the row to take a logical OR with

    • c: number

      Index of the column to take a logical OR with

    • v: number

      Value to take a logical OR

    Returns T

    Old value

power

  • power(p: number): Matrix

  • Returns a power of this matrix.

    Parameters

    • p: number

      Power exponent value

    Returns Matrix

    Powered matrix

prod

  • prod(): number

  • Returns producted value of all element.

    Returns number

    Producted value

  • prod(axis: 0 | 1): Matrix

  • Returns producted values along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Producted values

  • prod(axis: number): number | Matrix

  • Returns producted values along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a producted value of the all element.

    Returns number | Matrix

    Producted values

pseudoInv

  • pseudoInv(): Matrix

  • Returns a pseudo inverse matrix.

    Returns Matrix

    pseudo inverse matrix

pseudoInvBenIsraelCohen

  • pseudoInvBenIsraelCohen(): Matrix

  • Returns a Moore–Penrose inverse matrix by Ben-Israel and Cohen iterative method.

    Returns Matrix

    Moore–Penrose inverse matrix

pseudoInvNaive

  • pseudoInvNaive(): Matrix

  • Returns a pseudo inverse matrix.

    Returns Matrix

    pseudo inverse matrix

pseudoInvQR

  • pseudoInvQR(): Matrix

  • Returns a Moore–Penrose inverse matrix by QR decomposition.

    Returns Matrix

    Moore–Penrose inverse matrix

pseudoInvSVD

  • pseudoInvSVD(): Matrix

  • Returns a Moore–Penrose inverse matrix by SVD decomposition.

    Returns Matrix

    Moore–Penrose inverse matrix

qr

  • qr(): [Matrix, Matrix]

  • Returns a QR decomposition.

    Returns [Matrix, Matrix]

    Orthogonal matrix and upper triangular matrix

qrGramSchmidt

  • qrGramSchmidt(): [Matrix, Matrix]

  • Returns a QR decomposition by Gram-Schmidt method.

    Returns [Matrix, Matrix]

    Orthogonal matrix and upper triangular matrix

qrHouseholder

  • qrHouseholder(): [Matrix, Matrix]

  • Returns a QR decomposition by Householder method.

    Returns [Matrix, Matrix]

    Orthogonal matrix and upper triangular matrix

quantile

  • quantile(q: number): number

  • Returns quantile value of all element.

    Parameters

    • q: number

      Partition rate

    Returns number

    Quantile value

  • quantile(q: number, axis: 0 | 1): Matrix

  • Returns quantile values along the axis.

    Parameters

    • q: number

      Partition rate

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Quantile values

  • quantile(q: number, axis: number): number | Matrix

  • Returns quantile values along the axis.

    Parameters

    • q: number

      Partition rate

    • axis: number

      Axis to be reduced. If negative, returns the quantile value of the all element.

    Returns number | Matrix

    Quantile values

rank

  • rank(tol?: number): number

  • Returns a rank of this matrix.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns number

    Rank of this matrix

rankFactorization

  • rankFactorization(): [Matrix, Matrix]

  • Returns rank factorization.

    Returns [Matrix, Matrix]

    Rank factorization matrix

reduce

  • reduce<U>(
        cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U,
        init?: U,
    ): U

  • Returns a matrix reduced along all element with the callback function.

    Type Parameters

    • U

    Parameters

    • cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U

      Reducing function

    • Optionalinit: U

      Initial value

    Returns U

    Reduced value

  • reduce<U>(
        cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U,
        init: U,
        axis: 0 | 1,
        keepdims?: boolean,
    ): Matrix<U>

  • Returns a matrix reduced along the axis with the callback function.

    Type Parameters

    • U

    Parameters

    • cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U

      Reducing function

    • init: U

      Initial value

    • axis: 0 | 1

      Axis to be reduced

    • Optionalkeepdims: boolean

      Keep dimensions or not. If null, negative axis retuns number and other axis returns Matrix.

    Returns Matrix<U>

    Reduced matrix

  • reduce<U>(
        cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U,
        init: U,
        axis: number | number[],
        keepdims?: boolean,
    ): U | Matrix<U>

  • Returns a matrix reduced along the axis with the callback function.

    Type Parameters

    • U

    Parameters

    • cb: (arg0: U, arg1: T, arg2: number[], arg3: Matrix<T>) => U

      Reducing function

    • init: U

      Initial value

    • axis: number | number[]

      Axis to be reduced. If negative, reduce along all elements.

    • Optionalkeepdims: boolean

      Keep dimensions or not. If null, negative axis retuns number and other axis returns Matrix.

    Returns U | Matrix<U>

    Reduced matrix or value

reducedRowEchelonForm

  • reducedRowEchelonForm(tol?: number): void

  • Calculate reduced row echelon form in-place.

    Parameters

    • Optionaltol: number

      Tolerance to be recognized as 0

    Returns void

remove

  • remove(idx: number | number[], axis?: number): void

  • Remove specified indexes.

    Parameters

    • idx: number | number[]

      Remove index

    • Optionalaxis: number

      Axis to be removed

    Returns void

removeIf

  • removeIf(cond: (arg0: Matrix<T>) => boolean, axis?: number): void

  • Remove specified indexes.

    Parameters

    • cond: (arg0: Matrix<T>) => boolean

      Remove condition function. Remove if it returns true

    • Optionalaxis: number

      Axis to be removed

    Returns void

repeat

  • repeat(n: number, axis?: number): void

  • Repeat the elements n times along the axis this.

    Parameters

    • n: number

      Repeated count

    • Optionalaxis: number

      Axis to be repeated

    Returns void

    No return

  • repeat(n: number[]): void

  • Repeat the elements n times along the axis this.

    Parameters

    • n: number[]

      Repeated counts for each axis

    Returns void

    No return

reshape

  • reshape(rows: number, cols: number): any

  • Reshape this.

    Parameters

    • rows: number

      New row size

    • cols: number

      New column size

    Returns any

  • reshape(sizes: [number, number]): any

  • Reshape this.

    Parameters

    • sizes: [number, number]

      New sizes for each dimension

    Returns any

resize

  • resize(rows: number, cols: number, init?: T): any

  • Resize this matrix.

    Parameters

    • rows: number

      New row size

    • cols: number

      New column size

    • Optionalinit: T

      Value of the extended region

    Returns any

  • resize(size: [number, number], init?: T): any

  • Resize this matrix.

    Parameters

    • size: [number, number]

      New sizes for each dimension

    • Optionalinit: T

      Value of the extended region

    Returns any

round

  • round(): void

  • Set all elements to their rounded values.

    Returns void

row

  • row(r: number | number[] | boolean[]): Matrix<T>

  • Returns a row matrix at r.

    Parameters

    • r: number | number[] | boolean[]

      Indexes of rows, or an array of boolean values where the row to be selected is true.

    Returns Matrix<T>

    Row selected matrix

sample

  • sample(n: number, axis?: number, duplicate?: boolean): [Matrix<T>, number[]]

  • Returns a matrix that sampled along the axis.

    Parameters

    • n: number

      Sampled size

    • Optionalaxis: number

      Axis to be sampled

    • Optionalduplicate: boolean

      Allow duplicate index or not

    Returns [Matrix<T>, number[]]

    Sampled matrix and its original indexes

schur

  • schur(): [Matrix, Matrix]

  • Returns schur decomposition.

    Returns [Matrix, Matrix]

    Schur decomposition matrix

schurQR

  • schurQR(shift?: "no" | "single"): [Matrix, Matrix]

  • Returns schur decomposition by QR decomposition.

    Parameters

    • Optionalshift: "no" | "single"

      Shifting type

    Returns [Matrix, Matrix]

    Schur decomposition matrix

set

  • set(r: number, c: number, value: Matrix<T> | T): T

  • Set a value at the position.

    Parameters

    • r: number

      Row index

    • c: number

      Column index

    • value: Matrix<T> | T

      The value to be set

    Returns T

    Old value

  • set(r: [number, number], value: Matrix<T> | T): T

  • Set a value at the position.

    Parameters

    • r: [number, number]

      Index values

    • value: Matrix<T> | T

      The value to be set

    Returns T

    Old value

shuffle

  • shuffle(axis?: number): number[]

  • Shuffle values along the axis.

    Parameters

    • Optionalaxis: number

      Axis to be shuffled

    Returns number[]

    Original index.

signedRightShift

  • signedRightShift(n: number): void

  • Set all elements to their right shift values.

    Parameters

    • n: number

      Shift amount

    Returns void

singularValuePowerIteration

  • singularValuePowerIteration(maxIteration?: number): number

  • Returns the maximum singular value.

    Parameters

    • OptionalmaxIteration: number

      Maximum iteration

    Returns number

    Maximum singular value

singularValues

  • singularValues(): number[]

  • Returns singular values.

    Returns number[]

    Singular values

slice

  • slice(from: number, to: number, axis?: number): Matrix<T>

  • Returns sliced matrix.

    Parameters

    • from: number

      Start index

    • to: number

      End index

    • Optionalaxis: number

      Axis to be sliced

    Returns Matrix<T>

    Sliced matrix

solve

  • solve(b: Matrix): Matrix

  • Returns a solved value A of a equation XA=B.

    Parameters

    • b: Matrix

      Dependent variable values

    Returns Matrix

    Solved matrix

solveGaussSeidel

  • solveGaussSeidel(b: Matrix, maxIteration?: number): Matrix

  • Returns a solved value with Gauss-Seidel method.

    Parameters

    • b: Matrix

      Dependent variable values

    • OptionalmaxIteration: number

      Maximum iteration

    Returns Matrix

    Solved matrix

solveJacobi

  • solveJacobi(b: Matrix, maxIteration?: number): Matrix

  • Returns a solved value with Jacobi method.

    Parameters

    • b: Matrix

      Dependent variable values

    • OptionalmaxIteration: number

      Maximum iteration

    Returns Matrix

    Solved matrix

solveLowerTriangular

  • solveLowerTriangular(b: Matrix): Matrix

  • Returns a solved value for lower triangular matrix.

    Parameters

    • b: Matrix

      Dependent variable values

    Returns Matrix

    Solved matrix

solveSOR

  • solveSOR(b: Matrix, w: Matrix, maxIteration?: number): Matrix

  • Returns a solved value with Successive Over-Relaxation method.

    Parameters

    • b: Matrix

      Dependent variable values

    • w: Matrix

      Relaxation factor

    • OptionalmaxIteration: number

      Maximum iteration

    Returns Matrix

    Solved matrix

solveUpperTriangular

  • solveUpperTriangular(b: Matrix): Matrix

  • Returns a solved value for upper triangular matrix.

    Parameters

    • b: Matrix

      Dependent variable values

    Returns Matrix

    Solved matrix

some

  • some(cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean): boolean

  • Determines whether the specified callback function returns true for any element of a matrix.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    Returns boolean

    Reduced value or matrix

  • some(
        cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean,
        axis: 0 | 1,
    ): Matrix<boolean>

  • Determines whether the specified callback function returns true for any element of a matrix.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix<boolean>

    Reduced value or matrix

  • some(
        cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean,
        axis: number,
    ): boolean | Matrix<boolean>

  • Determines whether the specified callback function returns true for any element of a matrix.

    Parameters

    • cb: (arg0: T, arg1: number[], arg2: Matrix<T>) => boolean

      Check function

    • axis: number

      Axis to be reduced

    Returns boolean | Matrix<boolean>

    Reduced value or matrix

sort

  • sort(axis?: number): number[]

  • Sort values along the axis.

    Parameters

    • Optionalaxis: number

      Axis to be sorted

    Returns number[]

    Original index.

spectralRadius

  • spectralRadius(): number

  • Returns a spectral radius.

    Returns number

    Spectral radius

sqrt

  • sqrt(): Matrix

  • Returns a square root of this matrix.

    Returns Matrix

    Squared matrix

std

  • std(): number

  • Returns standard deviation of all element.

    Returns number

    Standard deviation value

  • std(axis: 0 | 1, ddof?: number): Matrix

  • Returns standard deviations along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    • Optionalddof: number

      Delta Degrees of Freedom

    Returns Matrix

    Standard deviation values

  • std(axis: number, ddof?: number): number | Matrix

  • Returns standard deviations along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a standard deviation of the all element.

    • Optionalddof: number

      Delta Degrees of Freedom

    Returns number | Matrix

    Standard deviation values

sub

  • sub(o: number | Tensor | Matrix): void

  • Subtract a value or matrix.

    Parameters

    Returns void

subAt

  • subAt(r: number, c: number, v: number): number

  • Subtract a value from the value at the position.

    Parameters

    • r: number

      Index of the row to subtract the value to

    • c: number

      Index of the column to subtract the value to

    • v: number

      Value to subtract

    Returns number

    Old value

sum

  • sum(): number

  • Returns summation value of all element.

    Returns number

    Summation value

  • sum(axis: 0 | 1): Matrix

  • Returns summation values along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced

    Returns Matrix

    Summation values

  • sum(axis: number): number | Matrix

  • Returns summation values along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a summation value of the all element.

    Returns number | Matrix

    Summation values

svd

  • svd(): [Matrix, number[], Matrix]

  • Returns a singular value decomposition.

    Returns [Matrix, number[], Matrix]

    Unitary matrix and singular values

svdEigen

  • svdEigen(): [Matrix, number[], Matrix]

  • Returns a singular value decomposition by eigen decomposition.

    Returns [Matrix, number[], Matrix]

    Unitary matrix and singular values

svdGolubKahan

  • svdGolubKahan(): [Matrix, number[], Matrix]

  • Returns a singular value decomposition by Golub-Kahan method.

    Returns [Matrix, number[], Matrix]

    Unitary matrix and singular values

swap

  • swap(a: number, b: number, axis?: number): void

  • Swap the index a and b along the axis.

    Parameters

    • a: number

      First index

    • b: number

      Second index

    • Optionalaxis: number

      Axis to be swapped

    Returns void

tDot

  • tDot(o: Matrix): Matrix

  • Returns a matrix product of the transposed matrix of this and input.

    Parameters

    Returns Matrix

    Producted matrix

toArray

  • toArray(): T[][]

  • Returns a nested array represented this matrix.

    Returns T[][]

    Nested array

toScaler

  • toScaler(): T

  • Returns the only element.

    Returns T

    The only element

toString

  • toString(): string

  • Returns a string represented this matrix.

    Returns string

    String represented this matrix

trace

  • trace(): number

  • Returns a trace.

    Returns number

    Trace value

transpose

  • transpose(): Matrix<T>

  • Returns transpose matrix.

    Returns Matrix<T>

    Transposed matrix

tridiag

  • tridiag(): Matrix

  • Returns a tridiagonal matrix.

    Returns Matrix

    Tridiagonal matrix

tridiagHouseholder

  • tridiagHouseholder(return_u?: false): Matrix

  • Returns a tridiagonal matrix.

    Parameters

    • Optionalreturn_u: false

      Returns orthogonal matrix

    Returns Matrix

    Tridiagonal matrix, or Tridiagonal matrix and orthogonal matrix

  • tridiagHouseholder(return_u: true): [Matrix, Matrix]

  • Returns a tridiagonal matrix.

    Parameters

    • return_u: true

      Returns orthogonal matrix

    Returns [Matrix, Matrix]

    Tridiagonal matrix, or Tridiagonal matrix and orthogonal matrix

tridiagLanczos

  • tridiagLanczos(k?: number): Matrix

  • Returns a tridiagonal matrix.

    Parameters

    • Optionalk: number

      Number of iterations

    Returns Matrix

    Tridiagonal matrix

unique

  • unique(axis?: number, tol?: number): number[]

  • Make it unique in the specified axis.

    Parameters

    • Optionalaxis: number

      Axis to be uniqued

    • Optionaltol: number

      Tolerance to be recognized as the same

    Returns number[]

    Selected indexes

unsignedRightShift

  • unsignedRightShift(n: number): void

  • Set all elements to their unsigned right shift values.

    Parameters

    • n: number

      Shift amount

    Returns void

variance

  • variance(): number

  • Returns variance of all element.

    Returns number

    Variance value

  • variance(axis: 0 | 1, ddof?: number): Matrix

  • Returns variances along the axis.

    Parameters

    • axis: 0 | 1

      Axis to be reduced.

    • Optionalddof: number

      Delta Degrees of Freedom

    Returns Matrix

    Variance values

  • variance(axis: number, ddof?: number): number | Matrix

  • Returns variances along the axis.

    Parameters

    • axis: number

      Axis to be reduced. If negative, returns a variance of the all element.

    • Optionalddof: number

      Delta Degrees of Freedom

    Returns number | Matrix

    Variance values

Staticadd

  • add(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that add two values.

    Parameters

    Returns Matrix

    Added matrix

Staticand

  • and(a: number | Matrix<unknown>, b: number | Matrix<unknown>): Matrix

  • Returns a matrix that takes logical AND two values.

    Parameters

    • a: number | Matrix<unknown>

      Left value

    • b: number | Matrix<unknown>

      Right value

    Returns Matrix

    Logical AND matrix

Staticbitand

  • bitand(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that takes bitwise AND two values.

    Parameters

    Returns Matrix

    Bitwise AND matrix

Staticbitor

  • bitor(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that takes bitwise OR two values.

    Parameters

    Returns Matrix

    Bitwise OR matrix

Staticbitxor

  • bitxor(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that takes bitwise XOR two values.

    Parameters

    Returns Matrix

    Bitwise XOR matrix

Staticconcat

  • concat<T_6, U_3>(
        a: Matrix<T_6>,
        b: Matrix<U_3>,
        axis?: number,
    ): Matrix<T_6 | U_3>

  • Returns a matrix concatenated this and m.

    Type Parameters

    • T_6
    • U_3

    Parameters

    • a: Matrix<T_6>

      Original matrix

    • b: Matrix<U_3>

      Concatenate matrix

    • Optionalaxis: number

      Axis to be concatenated

    Returns Matrix<T_6 | U_3>

    Concatenated matrix

Staticdiag

Staticdiv

  • div(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that divides by two values element-wise.

    Parameters

    Returns Matrix

    Divided matrix

Staticeye

  • eye(rows: number, cols: number, init?: number): Matrix

  • Returns a identity matrix.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    • Optionalinit: number

      Diagonal values

    Returns Matrix

    Identity matrix

  • eye(size: [number, number], init?: number): Matrix

  • Returns a identity matrix.

    Parameters

    • size: [number, number]

      Sizes for each dimension

    • Optionalinit: number

      Diagonal values

    Returns Matrix

    Identity matrix

StaticfromArray

Staticmap

  • map<T_3, U_2>(
        mat: Matrix<T_3>,
        cb: (arg0: T_3, arg1: number[], arg2: Matrix<T_3>) => U_2,
    ): Matrix<U_2>

  • Returns a matrix that replace all the elements.

    Type Parameters

    • T_3
    • U_2

    Parameters

    Returns Matrix<U_2>

    Mapped matrix

Staticmod

  • mod(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that takes a remainder divided by two values element-wise.

    Parameters

    Returns Matrix

    Remainder matrix

Staticmult

  • mult(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that multiplies by two values element-wise.

    Parameters

    Returns Matrix

    Multiplied matrix

Staticones

  • ones(rows: number, cols: number): Matrix

  • Returns a matrix filled with 1.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    Returns Matrix

    Matrix filled with 1

  • ones(size: [number, number]): Matrix

  • Returns a matrix filled with 1.

    Parameters

    • size: [number, number]

      Sizes for each dimension

    Returns Matrix

    Matrix filled with 1

Staticor

  • or(a: number | Matrix<unknown>, b: number | Matrix<unknown>): Matrix

  • Returns a matrix that takes logical OR two values.

    Parameters

    • a: number | Matrix<unknown>

      Left value

    • b: number | Matrix<unknown>

      Right value

    Returns Matrix

    Logical OR matrix

Staticrandint

  • randint(rows: number, cols: number, min?: number, max?: number): Matrix

  • Returns a matrix initialized uniform random integer values.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    • Optionalmin: number

      Minimum value of the Matrix (include)

    • Optionalmax: number

      Maximum value of the Matrix (include)

    Returns Matrix

    Matrix initialized uniform random values

  • randint(size: [number, number], min?: number, max?: number): Matrix

  • Returns a matrix initialized uniform random integer values.

    Parameters

    • size: [number, number]

      Sizes for each dimension

    • Optionalmin: number

      Minimum value of the Matrix (include)

    • Optionalmax: number

      Maximum value of the Matrix (include)

    Returns Matrix

    Matrix initialized uniform random values

Staticrandn

  • randn(
        rows: number,
        cols: number,
        myu?: number | number[],
        sigma?: number | number[][],
    ): Matrix

  • Returns a matrix initialized normal random values.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    • Optionalmyu: number | number[]

      Mean value(s) of each columns

    • Optionalsigma: number | number[][]

      Variance value or covariance matrix of each columns

    Returns Matrix

    Matrix initialized normal random values

  • randn(
        rows: [number, number],
        myu?: number | number[],
        sigma?: number | number[][],
    ): Matrix

  • Returns a matrix initialized normal random values.

    Parameters

    • rows: [number, number]

      Sizes for each dimension

    • Optionalmyu: number | number[]

      Mean value(s) of each columns

    • Optionalsigma: number | number[][]

      Variance value or covariance matrix of each columns

    Returns Matrix

    Matrix initialized normal random values

Staticrandom

  • random(rows: number, cols: number, min?: number, max?: number): Matrix

  • Returns a matrix initialized uniform random values.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    • Optionalmin: number

      Minimum value of the Matrix (include)

    • Optionalmax: number

      Maximum value of the Matrix (exclude)

    Returns Matrix

    Matrix initialized uniform random values

  • random(size: [number, number], min?: number, max?: number): Matrix

  • Returns a matrix initialized uniform random values.

    Parameters

    • size: [number, number]

      Sizes for each dimension

    • Optionalmin: number

      Minimum value of the Matrix (include)

    • Optionalmax: number

      Maximum value of the Matrix (exclude)

    Returns Matrix

    Matrix initialized uniform random values

Staticrepeat

  • repeat<T_5>(mat: Matrix<T_5>, n: number, axis?: number): Matrix<T_5>

  • Returns a matrix that repeat the elements n times along the axis.

    Type Parameters

    • T_5

    Parameters

    • mat: Matrix<T_5>

      Original matrix

    • n: number

      Repeated count

    • Optionalaxis: number

      Axis to be repeated

    Returns Matrix<T_5>

    Repeated matrix

  • repeat<T_5>(mat: Matrix<T_5>, n: number[]): Matrix<T_5>

  • Returns a matrix that repeat the elements n times along the axis.

    Type Parameters

    • T_5

    Parameters

    • mat: Matrix<T_5>

      Original matrix

    • n: number[]

      Repeated counts for each axis

    Returns Matrix<T_5>

    Repeated matrix

Staticresize

  • resize<T_4>(
        mat: Matrix<T_4>,
        rows: number,
        cols: number,
        init?: T_4,
    ): Matrix<T_4>

  • Return resized matrix.

    Type Parameters

    • T_4

    Parameters

    • mat: Matrix<T_4>

      Original matrix

    • rows: number

      New row size

    • cols: number

      New column size

    • Optionalinit: T_4

      Value of the extended region

    Returns Matrix<T_4>

    Resized matrix

  • resize<T_4>(mat: Matrix<T_4>, rows: [number, number], init?: T_4): Matrix<T_4>

  • Return resized matrix.

    Type Parameters

    • T_4

    Parameters

    • mat: Matrix<T_4>

      Original matrix

    • rows: [number, number]

      New sizes for each dimension

    • Optionalinit: T_4

      Value of the extended region

    Returns Matrix<T_4>

    Resized matrix

Staticsub

  • sub(a: number | Matrix, b: number | Matrix): Matrix

  • Returns a matrix that subtract two values.

    Parameters

    Returns Matrix

    Subtracted matrix

Staticzeros

  • zeros(rows: number, cols: number): Matrix

  • Returns a matrix filled with 0.

    Parameters

    • rows: number

      Number of rows

    • cols: number

      Number of columns

    Returns Matrix

    Matrix filled with 0

  • zeros(size: [number, number]): Matrix

  • Returns a matrix filled with 0.

    Parameters

    • size: [number, number]

      Sizes for each dimension

    Returns Matrix

    Matrix filled with 0

Settings

Member Visibility

On This Page

Constructors
Accessors
Methods
[iterator]absaddaddAtadjointandandAtargmaxargminatbalancingbalancingSinkhornKnoppbidiagbidiagHouseholderbitandbitandAtbitnotbitorbitorAtbitxorbitxorAtblockbroadcastOperateceilcholeskycholeskyBanachiewiczcholeskyCroutcholeskyGaussiancolconcatconvolutecopycovdetdiagdivdivAtdoteigeneigenInverseIterationeigenJacobieigenPowerIterationeigenSimultaneousIterationeigenValueseigenValuesBiSectioneigenValuesLReigenValuesQReigenVectorsequalseveryexpfillflipfloorforEachgramhessenberghessenbergArnoldiidividivAtimodimodAtinvinvLowerTriangularinvLUinvRowReductioninvUpperTriangularisAlternatingisDiagisHermitianisIdentityisLowerTriangularisNilpotentisNormalisOrthogonalisRegularisSkewHermitianisSquareisSymmetricisTriangularisubisubAtisUnitaryisUpperTriangularisZerokronleftShiftloglumapmaxmeanmedianminmodmodAtmodifiedCholeskymultmultAtnegativenormnormEntrywisenormFrobeniusnormInducednormMaxnormNuclearnormSchattennormSpectralnotoperateAtororAtpowerprodpseudoInvpseudoInvBenIsraelCohenpseudoInvNaivepseudoInvQRpseudoInvSVDqrqrGramSchmidtqrHouseholderquantilerankrankFactorizationreducereducedRowEchelonFormremoveremoveIfrepeatreshaperesizeroundrowsampleschurschurQRsetshufflesignedRightShiftsingularValuePowerIterationsingularValuesslicesolvesolveGaussSeidelsolveJacobisolveLowerTriangularsolveSORsolveUpperTriangularsomesortspectralRadiussqrtstdsubsubAtsumsvdsvdEigensvdGolubKahanswaptDottoArraytoScalertoStringtracetransposetridiagtridiagHouseholdertridiagLanczosuniqueunsignedRightShiftvarianceaddandbitandbitorbitxorconcatdiagdiveyefromArraymapmodmultonesorrandintrandnrandomrepeatresizesubzeros