@saehrimnir/druidjs / TSNE

Class: TSNE<T>

Defined in: dimred/TSNE.js:39

t-SNE (t-Distributed Stochastic Neighbor Embedding)

A nonlinear dimensionality reduction technique particularly well-suited for visualizing high-dimensional data in 2D or 3D. Preserves local structure while revealing global patterns.

Template

T

See

• Paper
• UMAP for faster alternative with similar results

Example

import * as druid from "@saehrimnir/druidjs";

const X = [[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12]];
const tsne = new druid.TSNE(X, {
    perplexity: 30,
    epsilon: 10,
    d: 2,
    seed: 42
});

const Y = tsne.transform(500); // 500 iterations
// [[x1, y1], [x2, y2], [x3, y3]]

Extends

• DR

Type Parameters

Type Parameter	Description
T extends InputType

Constructors

Constructor

new TSNE<T>(X: T, parameters?: Partial<ParametersTSNE>): TSNE<T>;

Defined in: dimred/TSNE.js:49

Parameters

Parameter	Type	Description
X	T	The high-dimensional data.
parameters?	Partial<ParametersTSNE>	Object containing parameterization of the DR method.

Returns

TSNE<T>

Overrides

DR.constructor

Properties

Property	Type	Inherited from	Defined in
__input	T	DR.__input	dimred/DR.js:46
_D	number	DR._D	dimred/DR.js:28
_gains	Matrix | undefined	-	dimred/TSNE.js:93
_is_initialized	boolean	DR._is_initialized	dimred/DR.js:34
_iter	number	-	dimred/TSNE.js:62
_N	number	DR._N	dimred/DR.js:30
_P	Matrix | undefined	-	dimred/TSNE.js:145
_parameters	ParametersTSNE	DR._parameters	dimred/DR.js:49
_randomizer	Randomizer	DR._randomizer	dimred/DR.js:32
_type	"array" | "matrix" | "typed"	DR._type	dimred/DR.js:54
_ystep	Matrix | undefined	-	dimred/TSNE.js:92
X	Matrix	DR.X	dimred/DR.js:56
Y	Matrix	DR.Y	dimred/TSNE.js:64

Accessors

projection

Get Signature

get projection(): T;

Defined in: dimred/DR.js:219

Returns

T

The projection in the type of input X.

Inherited from

DR.projection

Methods

[dispose]()

dispose: void;

Defined in: dimred/DR.js:320

Alias of release for the using declaration, so a hand-driven run frees its buffers when the block exits. Note that transform and a completed or break-ed generator() already release on their own — this only matters for a generator abandoned part way.

using tsne = new TSNE(X, { d: 2 });
const steps = tsne.generator(500);
steps.next(); // buffers released when the enclosing block exits

Returns

void

Inherited from

DR.[dispose]

---

check_init()

check_init(): DR<T, ParametersTSNE>;

Defined in: dimred/DR.js:210

If the respective DR method has an init function, call it before transform.

Returns

DR<T, ParametersTSNE>

Inherited from

DR.check_init

---

generator()

generator(iterations?: number): Generator<T, T, void>;

Defined in: dimred/TSNE.js:169

Parameters

Parameter	Type	Default value	Description
iterations?	number	500	Number of iterations. Default is 500

Returns

Generator<T, T, void>

• The projection.

Overrides

DR.generator

---

init()

abstract init(): TSNE<T>;

Defined in: dimred/TSNE.js:67

Returns

TSNE<T>

Overrides

DR.init

---

parameter()

Call Signature

parameter(): ParametersTSNE;

Defined in: dimred/DR.js:82

Get all Parameters.

Returns

ParametersTSNE

Inherited from

DR.parameter

Call Signature

parameter<K>(name: K): ParametersTSNE[K];

Defined in: dimred/DR.js:88

Get value of given parameter.

Type Parameters

Type Parameter	Description
K extends keyof ParametersTSNE

Parameters

Parameter	Type	Description
name	K	Name of the parameter.

Returns

ParametersTSNE[K]

Inherited from

DR.parameter

Call Signature

parameter<K>(name: K, value: ParametersTSNE[K]): TSNE<T>;

Defined in: dimred/DR.js:95

Set value of given parameter.

Type Parameters

Type Parameter	Description
K extends keyof ParametersTSNE

Parameters

Parameter	Type	Description
name	K	Name of the parameter.
value	ParametersTSNE[K]	Value of the parameter to set.

Returns

TSNE<T>

Inherited from

DR.parameter

---

release()

release(): TSNE<T>;

Defined in: dimred/DR.js:302

Hands back the WASM buffers this instance is holding.

Only needed after driving generator() by hand and stopping early — a plain transform(), or a for…of over generator() (including one you break), already releases when it ends. It frees only this method's buffers, never another running instance's, and the next run simply reallocates, so it is safe to call at any time, more than once, and while other instances are mid-run.

Returns

TSNE<T>

Example

const tsne = new TSNE(X, { d: 2 });
const steps = tsne.generator(500);
steps.next();
tsne.release(); // stop early and give the buffers back

Inherited from

DR.release

---

transform()

transform(iterations?: number): T;

Defined in: dimred/TSNE.js:153

Parameters

Parameter	Type	Default value	Description
iterations?	number	500	Number of iterations. Default is 500

Returns

T

The projection.

Overrides

DR.transform

---

transform_async()

transform_async(...args: unknown[]): Promise<T>;

Defined in: dimred/DR.js:241

Computes the projection.

Parameters

Parameter	Type	Description
...args	unknown[]	Arguments the transform method of the respective DR method takes.

Returns

Promise<T>

The dimensionality reduced dataset.

Inherited from

DR.transform_async

---

generator()

static generator<T>(X: T, parameters?: Partial<ParametersTSNE>): Generator<T, T, void>;

Defined in: dimred/TSNE.js:298

Type Parameters

Type Parameter	Description
T extends InputType

Parameters

Parameter	Type	Description
X	T	-
parameters?	Partial<ParametersTSNE>	-

Returns

Generator<T, T, void>

Overrides

DR.generator

---

transform()

static transform<T>(X: T, parameters?: Partial<ParametersTSNE>): T;

Defined in: dimred/TSNE.js:287

Type Parameters

Type Parameter	Description
T extends InputType

Parameters

Parameter	Type	Description
X	T	-
parameters?	Partial<ParametersTSNE>	-

Returns

T

Overrides

DR.transform

---

transform_async()

static transform_async<T>(X: T, parameters?: Partial<ParametersTSNE>): Promise<T>;

Defined in: dimred/TSNE.js:310

Type Parameters

Type Parameter	Description
T extends InputType

Parameters

Parameter	Type	Description
X	T	-
parameters?	Partial<ParametersTSNE>	-

Returns

Promise<T>

Overrides

DR.transform_async