neuralnetwork.md

ReImproveJS

Globals / NeuralNetwork /

Class: NeuralNetwork

Hierarchy

• NeuralNetwork

  • ConvolutionalNeuralNetwork

Index

Constructors

• constructor

Properties

• inputShape
• neuralNetworkLayers

Accessors

• InputShape

Methods

• addNeuralNetworkLayer
• addNeuralNetworkLayers
• createLayers
• getLayers

Object literals

• DEFAULT_LAYER

Constructors

constructor

+ new NeuralNetwork(): NeuralNetwork

Defined in reimprove/networks.ts:62

Returns: NeuralNetwork

---

Properties

Protected inputShape

● inputShape: number[]

Defined in reimprove/networks.ts:55

---

Private neuralNetworkLayers

● neuralNetworkLayers: NeuralNetworkLayer[]

Defined in reimprove/networks.ts:56

---

Accessors

InputShape

● set InputShape(shape: number[]): void

Defined in reimprove/networks.ts:85

Parameters:

Name	Type
shape	number[]

Returns: void

---

Methods

addNeuralNetworkLayer

▸ addNeuralNetworkLayer(layer: number | NeuralNetworkLayer): void

Defined in reimprove/networks.ts:69

Parameters:

Name	Type
layer	number | NeuralNetworkLayer

Returns: void

---

addNeuralNetworkLayers

▸ addNeuralNetworkLayers(layers: Array<number | NeuralNetworkLayer>): void

Defined in reimprove/networks.ts:81

Parameters:

Name	Type
layers	Array<number | NeuralNetworkLayer>

Returns: void

---

createLayers

▸ createLayers(includeInputShape: boolean): Array<any>

Defined in reimprove/networks.ts:89

Parameters:

Name	Type	Default
includeInputShape	boolean	true

Returns: Array<any>

---

getLayers

▸ getLayers(): Layer[]

Defined in reimprove/networks.ts:99

Returns: Layer[]

---

Object literals

Static Private DEFAULT_LAYER

■ DEFAULT_LAYER: object

Defined in reimprove/networks.ts:58

activation

● activation: string = "relu"

Defined in reimprove/networks.ts:60

type

● type: "dense" = "dense"

Defined in reimprove/networks.ts:61

units

● units: number = 32

Defined in reimprove/networks.ts:59

---