JSDoc: Class: NDArray

new NDArray(shape, data, dtype, base, strides, offset, itemsize)

Parameters:

Name	Type	Default	Description
`shape`	Array.<number>
`data`	Array.<any>	null
`dtype`	Dtype	null
`base`	NDArray	null
`strides`	Array.<number>	null
`offset`	number	0
`itemsize`	number	1	always 1

Source:

ndarray.mjs, line 69

Classes

NDArray

Members

T :NDArray

Type:

NDArray

Source:

ndarray.mjs, line 123

base :NDArray|null

Type:

NDArray | null

Source:

ndarray.mjs, line 101

data :Array.<any>

Type:

Array.<any>

Source:

ndarray.mjs, line 89

dtype :Dtype

Type:

Dtype

Source:

ndarray.mjs, line 98

flat :Flatiter

Type:

Flatiter

Source:

ndarray.mjs, line 114

itemsize :number

Type:

number

Source:

ndarray.mjs, line 91

length :number

Type:

number

Source:

ndarray.mjs, line 105

ndim :number

Type:

number

Source:

ndarray.mjs, line 82

offset :number

Type:

number

Source:

ndarray.mjs, line 95

shape :Array.<number>

Type:

Array.<number>

Source:

ndarray.mjs, line 87

size :number

Type:

number

Source:

ndarray.mjs, line 84

strides :Array.<number>

Type:

Array.<number>

Source:

ndarray.mjs, line 93

Methods

add(x, outopt) → {NDArray}

Returns add(this, x, out)

Parameters:

Name	Type	Attributes	Default	Description
`x`	NDArray
`out`	null \| NDArray	<optional>	null

Source:

ndarray.mjs, line 385

Returns:

Type: NDArray

Example

x.add(a, x)
// in-place add

all(axis, out, keepdims, initial, return_scalar) → {NDArray.<boolean>|boolean}

Parameters:

Name	Type	Default
`axis`	number \| Array.<number> \| null	null
`out`	NDArray	null
`keepdims`	boolean	false
`initial`	boolean	true
`return_scalar`	boolean	true

Source:

ndarray.mjs, line 443

Returns:

Type: NDArray.<boolean> | boolean

any(axis, out, keepdims, initial, return_scalar) → {NDArray.<boolean>|boolean}

Parameters:

Name	Type	Default
`axis`	number \| Array.<number> \| null	null
`out`	NDArray	null
`keepdims`	boolean	false
`initial`	boolean	false
`return_scalar`	boolean	true

Source:

ndarray.mjs, line 455

Returns:

Type: NDArray.<boolean> | boolean

argmax(axisopt, outopt, keepdimsopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| null	<optional>	null
`out`	NDArray \| null	<optional>	null
`keepdims`	boolean	<optional>	false

Source:

ndarray.mjs, line 465

Returns:

Type: NDArray

argmin(axisopt, outopt, keepdimsopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| null	<optional>	null
`out`	NDArray \| null	<optional>	null
`keepdims`	boolean	<optional>	false

Source:

ndarray.mjs, line 475

Returns:

Type: NDArray

argsort(axis, keyopt) → {NDArray.<number>}

Parameters:

Name	Type	Attributes	Default	Description
`axis`	number
`key`	function	<optional>	null

Source:

ndarray.mjs, line 488

Returns:

Type: NDArray.<number>

array() → {Array.<any>|any}

Source:

ndarray.mjs, line 313

Returns:

Type: Array.<any> | any

as_strided(shapeopt, stridesopt, offsetopt) → {NDArray}

Returns a view with the given shape, strides, and offset. out.base = this.base ?? this

Parameters:

Name	Type	Attributes	Description
`shape`	Array.<number>	<optional>
`strides`	Array.<number>	<optional>
`offset`	number	<optional>	required as a ptr to the first item

Source:

ndarray.mjs, line 160

Returns:

Type: NDArray

astype(dtype, copyopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`dtype`	Dtype
`copy`	boolean	<optional>	true

Source:

ndarray.mjs, line 299

Returns:

Type: NDArray

at(…indices) → {NDArray}

`.at(0)` is is equivalent to `.get([0])`

Parameters:

Name	Type	Attributes	Description
`indices`	number \| Slice \| string \| null \| Array.<number> \| Array.<boolean> \| NDArray	<repeatable>

Source:

ndarray.mjs, line 223

Returns:

Type: NDArray

clip(a_min, a_max, out) → {NDArray}

Parameters:

Name	Type	Default
`a_min`	number \| NDArray
`a_max`	number \| NDArray
`out`	NDArray	null

Source:

ndarray.mjs, line 506

Returns:

Type: NDArray

compress(condition, axisopt, outopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`condition`	Array.<boolean>
`axis`	number	<optional>	null
`out`	NDArray	<optional>	null

Source:

ndarray.mjs, line 516

Returns:

Type: NDArray

copy() → {NDArray}

Source:

ndarray.mjs, line 531

Returns:

Type: NDArray

cumprod(axisopt, outopt) → {NDArray}

Parameters:

Name	Type	Attributes	Description
`axis`	number	<optional>
`out`	NDArray	<optional>

Source:

ndarray.mjs, line 540

Returns:

Type: NDArray

cumsum(axisopt, outopt) → {NDArray}

Parameters:

Name	Type	Attributes	Description
`axis`	number	<optional>
`out`	NDArray	<optional>

Source:

ndarray.mjs, line 549

Returns:

Type: NDArray

diagonal(offsetopt, axis1opt, axis2opt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`offset`	number	<optional>	0
`axis1`	number	<optional>	0
`axis2`	number	<optional>	1

Source:

ndarray.mjs, line 559

Returns:

Type: NDArray

div(x, outopt) → {NDArray}

Returns divide(this, x, out)

Parameters:

Name	Type	Attributes	Default	Description
`x`	NDArray
`out`	null \| NDArray	<optional>	null

Source:

ndarray.mjs, line 418

Returns:

Type: NDArray

fill(value) → {NDArray}

Parameters:

Name	Type	Description
`value`	any

Source:

ndarray.mjs, line 575

Returns:

this

Type: NDArray

flatten() → {NDArray}

Source:

ndarray.mjs, line 586

Returns:

Type: NDArray

get(indices, axisopt) → {NDArray}

`.get(indices)` is is equivalent to `.at(...indices)`

Parameters:

Name	Type	Attributes	Default	Description
`indices`	Array.<(number\|Slice\|string\|null\|Array.<number>\|Array.<boolean>\|NDArray)>
`axis`	number	<optional>	0

Source:

ndarray.mjs, line 243

Returns:

Type: NDArray

idx(index) → {number}

Converts NDArray index to .data[] index.

Parameters:

Name	Type	Description
`index`	number \| Array.<number> \| undefined

Source:

ndarray.mjs, line 189

Returns:

Type: number

item(index) → {any}

Returns an element of the array. if `index` is an integer, `index` is regarded as a flatten index. `index` can be undefined only if `a.size == 1`.

Parameters:

Name	Type	Description
`index`	number \| Array.<number> \| undefined

Source:

ndarray.mjs, line 277

Returns:

Type: any

itemset(index, scalar) → {NDArray}

Sets an element of the array. index can be undefined only if a.size == 1.

Parameters:

Name	Type	Description
`index`	number \| Array.<number> \| undefined
`scalar`	any

Source:

ndarray.mjs, line 289

Returns:

Type: NDArray

keys() → {NdoffsetIterator}

Returns `ndoffset(this.shape, this.strides, this.offset)`. Useful when iterating `this.data[i]` or `this.item(i)` in a `for (let i of this.keys())` loop.

Source:

ndarray.mjs, line 171

Returns:

Type: NdoffsetIterator

max(axisopt, outopt, keepdimsopt, initialopt, return_scalaropt) → {NDArray.<any>|any}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`keepdims`	boolean	<optional>	false
`initial`	any	<optional>	null
`return_scalar`	boolean	<optional>	true

Source:

ndarray.mjs, line 606

Returns:

Type: NDArray.<any> | any

min(axisopt, outopt, keepdimsopt, initialopt, return_scalaropt) → {NDArray|any}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`keepdims`	boolean	<optional>	false
`initial`	any	<optional>	null
`return_scalar`	boolean	<optional>	true

Source:

ndarray.mjs, line 622

Returns:

Type: NDArray | any

mod(x, outopt) → {NDArray}

Returns mod(this, x, out)

Parameters:

Name	Type	Attributes	Default	Description
`x`	NDArray
`out`	null \| NDArray	<optional>	null

Source:

ndarray.mjs, line 429

Returns:

Type: NDArray

mul(x, outopt) → {NDArray}

Returns multiply(this, x, out)

Parameters:

Name	Type	Attributes	Default	Description
`x`	NDArray
`out`	null \| NDArray	<optional>	null

Source:

ndarray.mjs, line 407

Returns:

Type: NDArray

nonzero() → {Array.<NDArray>}

Source:

ndarray.mjs, line 633

Returns:

Type: Array.<NDArray>

prod(axisopt, outopt, keepdimsopt, initialopt, return_scalaropt) → {NDArray|any}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`keepdims`	boolean	<optional>	false
`initial`	any	<optional>	0
`return_scalar`	boolean	<optional>	true

Source:

ndarray.mjs, line 649

Returns:

Type: NDArray | any

ptp(axisopt, outopt, keepdimsopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`keepdims`	boolean	<optional>	false

Source:

ndarray.mjs, line 659

Returns:

Type: NDArray

put(indices, values, modeopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`indices`	Array.<number>
`values`	Array.<any>
`mode`	string	<optional>	raise

Source:

ndarray.mjs, line 669

Returns:

this

Type: NDArray

ravel() → {NDArray}

Source:

ndarray.mjs, line 677

Returns:

Type: NDArray

repeat(repeats, axisopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`repeats`	number \| Array.<number>
`axis`	number	<optional>	null

Source:

ndarray.mjs, line 686

Returns:

Type: NDArray

reshape(…shape) → {NDArray}

`.reshape(1, 2)` is equivalent to `.reshape([1, 2])`

Parameters:

Name	Type	Attributes	Description
`shape`	number \| Array.<number>	<repeatable>

Source:

ndarray.mjs, line 366

Returns:

Type: NDArray

resize(new_shape) → {NDArray}

inplace resize

Parameters:

Name	Type	Description
`new_shape`	Array.<number>

Source:

ndarray.mjs, line 697

Returns:

this

Type: NDArray

round(decimalsopt, outopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`decimals`	number	<optional>	0
`out`	NDArray	<optional>	null

Source:

ndarray.mjs, line 718

Returns:

Type: NDArray

searchsorted(v, sideopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`v`	Array.<any>
`side`	string	<optional>	left

Source:

ndarray.mjs, line 727

Returns:

Type: NDArray

set(indices, valueopt) → {NDArray}

`.set(value)` is equivalent to `.set(['...'], value)`, but faster

Parameters:

Name	Type	Attributes	Description
`indices`	Array.<(number\|Slice\|string\|null\|Array.<number>\|Array.<boolean>\|NDArray)>
`value`	any	<optional>

Source:

ndarray.mjs, line 256

Returns:

Type: NDArray

slice(startopt, endopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`start`	number \| null	<optional>	null
`end`	number \| null	<optional>	null

Source:

ndarray.mjs, line 233

Returns:

Type: NDArray

sort(axisopt, keyopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default	Description
`axis`	number	<optional>
`key`	function	<optional>	null

Source:

ndarray.mjs, line 744

Returns:

this

Type: NDArray

squeeze(axis) → {NDArray}

Parameters:

Name	Type	Default	Description
`axis`	number \| Array.<number>	null

Source:

ndarray.mjs, line 753

Returns:

Type: NDArray

std(axisopt, outopt, ddofopt, keepdimsopt) → {NDArray}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`ddof`	number	<optional>	0
`keepdims`	boolean	<optional>	false

Source:

ndarray.mjs, line 764

Returns:

Type: NDArray

sub(x, outopt) → {NDArray}

Returns subtract(this, x, out)

Parameters:

Name	Type	Attributes	Default	Description
`x`	NDArray
`out`	null \| NDArray	<optional>	null

Source:

ndarray.mjs, line 396

Returns:

Type: NDArray

sum(axisopt, outopt, keepdimsopt, initialopt, return_scalaropt) → {NDArray|any}

Parameters:

Name	Type	Attributes	Default
`axis`	number \| Array.<number>	<optional>	null
`out`	NDArray	<optional>	null
`keepdims`	boolean	<optional>	false
`initial`	number	<optional>	0
`return_scalar`	boolean	<optional>	true