Class Tensor<T>

Tensor class

Type Parameters

T extends unknown = number
Element type

Index

Constructors

constructor

Accessors

dimension length sizes value

Constructors

constructor

new Tensor<T extends unknown = number>(
size: number[],
value?: T | NestedArray<T>,
): Tensor<T>
Type Parameters
- T extends unknown = number
Parameters
- size: number[]
 Sizes for each dimension
- Optionalvalue: T | NestedArray<T>
 Initial values
Returns Tensor<T>
- Defined in lib/util/tensor.d.ts:65

Accessors

dimension

get dimension(): number
Dimension of the tensor.

Returns number
- Defined in lib/util/tensor.d.ts:78

length

get length(): number
Number of all elements in the tensor.

Returns number
- Defined in lib/util/tensor.d.ts:88

sizes

get sizes(): number[]
Sizes of the tensor.

Returns number[]
- Defined in lib/util/tensor.d.ts:83

value

get value(): T[]
Elements in the tensor.

Returns T[]
- Defined in lib/util/tensor.d.ts:93

Methods

_to_index

_to_index(p: any): any[]
Parameters
- p: any
Returns any[]
- Defined in lib/util/tensor.d.ts:116

_to_position

_to_position(...i: any[]): number
Parameters
- ...i: any[]
Returns number
- Defined in lib/util/tensor.d.ts:115

[iterator]

"[iterator]"(): Generator<any, void, unknown>
Iterate over the elements.

Returns Generator<any, void, unknown>
Yields
- Defined in lib/util/tensor.d.ts:318

at

at(...i: number[]): T
Returns value at the index position.
Parameters
- ...i: number[]
  Index values
Returns T
The value
- Defined in lib/util/tensor.d.ts:134
at(i: number[]): T
Returns value at the index position.
Parameters
- i: number[]
  Index values
Returns T
The value
- Defined in lib/util/tensor.d.ts:141

broadcastOperate

broadcastOperate(
o: U | Tensor | Matrix,
fn: (arg0: T, arg1: U) => T,
): void
Apply function for all elements with broadcasting.
Type Parameters
- U
Parameters
- o: U | Tensor | Matrix
 Applied value
- fn: (arg0: T, arg1: U) => T
 Applied function
Returns void
- Defined in lib/util/tensor.d.ts:300

concat

concat(t: Tensor<T>, axis?: number): void
Concatenate this and t.
Parameters
- t: Tensor<T>
 Concatenate tensor
- Optionalaxis: number
 Axis to be concatenated
Returns void
- Defined in lib/util/tensor.d.ts:253

copy

copy(): Tensor<T>
Returns a copy of this tensor.

Returns Tensor<T>
Copied tensor
- Defined in lib/util/tensor.d.ts:121

dot

dot(o: Matrix<number>): Tensor<number>
Returns a tensor product value.
Parameters
- o: Matrix<number>
 Right matrix
Returns Tensor<number>
Producted tensor
- Defined in lib/util/tensor.d.ts:313

equals

equals(other: any): boolean
Returns this tensor is equals to the others.
Parameters
- other: any
  Check tensor
Returns boolean
true if equal
- Defined in lib/util/tensor.d.ts:127

fill

fill(value: T): void
Fill in all the elements with the value.
Parameters
- value: T
  Filled value
Returns void
- Defined in lib/util/tensor.d.ts:181

flip

flip(axis?: number): void
Flip values along the axis.
Parameters
- Optionalaxis: number
  Axis to be flipped
Returns void
- Defined in lib/util/tensor.d.ts:211

forEach

forEach(cb: (arg0: T, arg1: number[], arg2: Tensor<T>) => void): void
Iterate over all the elements.
Parameters
- cb: (arg0: T, arg1: number[], arg2: Tensor<T>) => void
 Callback function
Returns void
- Defined in lib/util/tensor.d.ts:191

index

index(...i: number[]): Tensor<T>
Returns tensor at the index position.
Parameters
- ...i: number[]
 Index values
Returns Tensor<T>
Sub tensor
- Defined in lib/util/tensor.d.ts:148
index(i: number[]): Tensor<T>
Returns tensor at the index position.
Parameters
- i: number[]
 Index values
Returns Tensor<T>
Sub tensor
- Defined in lib/util/tensor.d.ts:155

map

map(cb: (arg0: T, arg1: number[], arg2: Tensor<T>) => T): void
Iterate over all the elements and replace the value.
Parameters
- cb: (arg0: T, arg1: number[], arg2: Tensor<T>) => T
 Mapping function
Returns void
- Defined in lib/util/tensor.d.ts:186

operateAt

operateAt(i: number | number[], fn?: (arg0: T) => T): T
Apply function to the position.
Parameters
- i: number | number[]
  Index values
- Optionalfn: (arg0: T) => T
  Applied function
Returns T
Old value
- Defined in lib/util/tensor.d.ts:307

reduce

reduce(
cb: (arg0: T, arg1: T, arg2: number[], arg3: Tensor<T>) => T,
init?: null,
): T
Returns a tensor reduced along all element with the callback function.
Parameters
- cb: (arg0: T, arg1: T, arg2: number[], arg3: Tensor<T>) => T
 Reducing function
- Optionalinit: null
 Initial value
Returns T
Reduced tensor or value
- Defined in lib/util/tensor.d.ts:261
reduce(
cb: (arg0: U, arg1: T, arg2: number[], arg3: Tensor<T>) => U,
init: U,
): U
Returns a tensor reduced along all element with the callback function.
Type Parameters
- U
Parameters
- cb: (arg0: U, arg1: T, arg2: number[], arg3: Tensor<T>) => U
 Reducing function
- init: U
 Initial value
Returns U
Reduced tensor or value
- Defined in lib/util/tensor.d.ts:270
reduce<F extends boolean>(
 cb: (arg0: T, arg1: T, arg2: number[], arg3: Tensor<T>) => T,
 init: null,
 axis: number | number[],
 keepdims?: F,
): Tensor<T> | (F extends true ? never : T)
Returns a tensor reduced along the axis with the callback function.
Type Parameters
- F extends boolean
Parameters
- cb: (arg0: T, arg1: T, arg2: number[], arg3: Tensor<T>) => T
 Reducing function
- init: null
 Initial value
- axis: number | number[]
 Axis to be reduced. If negative, reduce along all elements.
- Optionalkeepdims: F
 Keep dimensions or not.
Returns Tensor<T> | (F extends true ? never : T)
Reduced tensor or value
- Defined in lib/util/tensor.d.ts:281
reduce<U, F extends boolean>(
 cb: (arg0: U, arg1: T, arg2: number[], arg3: Tensor<T>) => U,
 init: U,
 axis: number | number[],
 keepdims?: F,
): Tensor | (F extends true ? never : U)
Returns a tensor reduced along the axis with the callback function.
Type Parameters
- U
- F extends boolean
Parameters
- cb: (arg0: U, arg1: T, arg2: number[], arg3: Tensor<T>) => U
 Reducing function
- init: U
 Initial value
- axis: number | number[]
 Axis to be reduced. If negative, reduce along all elements.
- Optionalkeepdims: F
 Keep dimensions or not.
Returns Tensor | (F extends true ? never : U)
Reduced tensor or value
- Defined in lib/util/tensor.d.ts:293

repeat

repeat(n: number, axis?: number): any
Repeat the elements n times along the axis this.
Parameters
- n: number
  Repeated count
- Optionalaxis: number
  Axis to be repeated
Returns any
- Defined in lib/util/tensor.d.ts:241
repeat(n: number[]): any
Repeat the elements n times along the axis this.
Parameters
- n: number[]
  Repeated counts of each axis
Returns any
- Defined in lib/util/tensor.d.ts:247

reshape

reshape(...sizes: number[]): any
Reshape this as the sizes.
Parameters
- ...sizes: number[]
  New sizes for each dimension
Returns any
- Defined in lib/util/tensor.d.ts:228
reshape(sizes: number[]): any
Reshape this as the sizes.
Parameters
- sizes: number[]
  New sizes for each dimension
Returns any
- Defined in lib/util/tensor.d.ts:234

resize

resize(sizes: number[], init?: number): void
Resize this tensor.
Parameters
- sizes: number[]
  New sizes
- Optionalinit: number
  Value of the extended region
Returns void
- Defined in lib/util/tensor.d.ts:222

select

select(idx: number | number[], axis?: number): Tensor<T>
Returns the sub-tensor corresponding to position i in the first dimension of this.
Parameters
- idx: number | number[]
 Select index value(s)
- Optionalaxis: number
 Axis
Returns Tensor<T>
Selected tensor
- Defined in lib/util/tensor.d.ts:168

set

set(i: number | number[], value: T): void
Set the value at the specific position.
Parameters
- i: number | number[]
  Index values
- value: T
  Set value
Returns void
- Defined in lib/util/tensor.d.ts:161

shuffle

shuffle(axis?: number): void
Shuffle along the axis.
Parameters
- Optionalaxis: number
  Axis
Returns void
- Defined in lib/util/tensor.d.ts:216

slice

slice(from: number, to: number, axis?: number): Tensor<T>
Returns a tensor sliced by first dimension.
Parameters
- from: number
 Start index
- to: number
 End index
- Optionalaxis: number
 Axis
Returns Tensor<T>
Sliced tensor
- Defined in lib/util/tensor.d.ts:176

toArray

toArray(): NestedArray<T>
Returns a nested array represented this tensor.

Returns NestedArray<T>
Nested array
- Defined in lib/util/tensor.d.ts:98

toMatrix

toMatrix(): Matrix<T>
Returns a Matrix if the dimension of this tensor is 2.

Returns Matrix<T>
Matrix

Throws
If the dimension of this tensor is not 2.
- Defined in lib/util/tensor.d.ts:109

toScaler

toScaler(): T
Returns the only element.

Returns T
The only element
- Defined in lib/util/tensor.d.ts:114

toString

toString(): string
Returns a string represented this tensor.

Returns string
String represented this tensor
- Defined in lib/util/tensor.d.ts:103

transpose

transpose(...axises: (number | number[])[]): Tensor<T>
Parameters
- ...axises: (number | number[])[]
 Selected axises
Returns Tensor<T>
Transposed tensor
- Defined in lib/util/tensor.d.ts:206

`Static`fromArray

fromArray<T_1>(
arr: T_1 | Tensor<T_1> | Matrix<T_1> | NestedArray<T_1>,
): Tensor<T_1>
Returns a tensor from some value.
Type Parameters
- T_1
Parameters
- arr: T_1 | Tensor<T_1> | Matrix<T_1> | NestedArray<T_1>
 Original values
Returns Tensor<T_1>
Tensor from some value
- Defined in lib/util/tensor.d.ts:60

`Static`ones

ones(...size: number[]): Tensor<number>
Returns a tensor filled with 1.
Parameters
- ...size: number[]
 Sizes for each dimension
Returns Tensor<number>
Tensor filled with 1
- Defined in lib/util/tensor.d.ts:30
ones(size: number[]): Tensor<number>
Returns a tensor filled with 1.
Parameters
- size: number[]
 Sizes for each dimension
Returns Tensor<number>
Tensor filled with 1
- Defined in lib/util/tensor.d.ts:37

`Static`randn

randn(size: number[], myu?: number, sigma?: number): Tensor<number>
Returns a tensor initialized normal random values.
Parameters
- size: number[]
 Sizes for each dimension
- Optionalmyu: number
 Mean value of the Tensor
- Optionalsigma: number
 Variance value of the Tensor
Returns Tensor<number>
Tensor initialized normal random values
- Defined in lib/util/tensor.d.ts:53

`Static`random

random(size: number[], min?: number, max?: number): Tensor<number>
Returns a tensor initialized uniform random values.
Parameters
- size: number[]
 Sizes for each dimension
- Optionalmin: number
 Minimum value of the Tensor
- Optionalmax: number
 Maximum value of the Tensor
Returns Tensor<number>
Tensor initialized uniform random values
- Defined in lib/util/tensor.d.ts:45

`Static`zeros

zeros(...size: number[]): Tensor<number>
Returns a tensor filled with 0.
Parameters
- ...size: number[]
 Sizes for each dimension
Returns Tensor<number>
Tensor filled with 0
- Defined in lib/util/tensor.d.ts:16
zeros(size: number[]): Tensor<number>
Returns a tensor filled with 0.
Parameters
- size: number[]
 Sizes for each dimension
Returns Tensor<number>
Tensor filled with 0
- Defined in lib/util/tensor.d.ts:23

Class Tensor<T>

Type Parameters

Index

Constructors

Accessors

Methods

Constructors

constructor

Type Parameters

Parameters

Returns Tensor<T>

Accessors

dimension

Returns number

length

Returns number

sizes

Returns number[]

value

Returns T[]

Methods

_to_index

Parameters

Returns any[]

_to_position

Parameters

Returns number

[iterator]

Returns Generator<any, void, unknown>

Yields

at

Parameters

Returns T

Parameters

Returns T

broadcastOperate

Type Parameters

Parameters

Returns void

concat

Parameters

Returns void

copy

Returns Tensor<T>

dot

Parameters

Returns Tensor<number>

equals

Parameters

Returns boolean

fill

Parameters

Returns void

flip

Parameters

Returns void

forEach

Parameters

Returns void

index

Parameters

Returns Tensor<T>

Parameters

Returns Tensor<T>

map

Parameters

Returns void

operateAt

Parameters

Returns T

reduce

Parameters

Returns T

Type Parameters

Parameters

Returns U

Type Parameters

Parameters

Returns Tensor<T> | (F extends true ? never : T)

Type Parameters

`Static`fromArray

`Static`ones

`Static`randn

`Static`random

`Static`zeros