AdamOptimizer

Defined in: compat.js:105

Adam Optimizer (Adaptive Moment Estimation)

Extends

• Optimizer

Constructors

Constructor

new AdamOptimizer(options?): AdamOptimizer

Defined in: compat.js:106

Parameters

options?

Returns

AdamOptimizer

Overrides

Optimizer.constructor

Properties

beta1

beta1: any

Defined in: compat.js:108

---

beta2

beta2: any

Defined in: compat.js:109

---

epsilon

epsilon: any

Defined in: compat.js:110

---

learningRate

learningRate: any

Defined in: compat.js:17

Inherited from

GradientDescent.learningRate

---

maxIter

maxIter: any

Defined in: compat.js:18

Inherited from

GradientDescent.maxIter

---

tol

tol: any

Defined in: compat.js:19

Inherited from

GradientDescent.tol

---

verbose

verbose: any

Defined in: compat.js:20

Inherited from

GradientDescent.verbose

Methods

_options()

_options(options): object

Defined in: compat.js:34

Parameters

options

any

Returns

object

learningRate

learningRate: any

maxIter

maxIter: any

tol

tol: any

verbose

verbose: any

Inherited from

Optimizer._options

---

minimize()

minimize(lossFn, x0, options?): object

Defined in: compat.js:113

Minimize a loss function

Parameters

lossFn

any

Function that returns {loss, gradient}

x0

any

Initial parameters

options?

Additional options

Returns

object

{x, history}

history

history: object

history.gradNorm

gradNorm: any

history.loss

loss: any

x

x: any

Overrides

Optimizer.minimize