CategoricalSlot | Wisty.js

A slot that stores a categorical value extracted using fuzzy string matching.

Hierarchy

Slot<CategoricalValue>
- CategoricalSlot

Index

Constructors

constructor

Properties

Methods

Constructors

constructor

new CategoricalSlot(__namedParameters: { categories: {}; dependantActions: string[]; invDependantActions: string[]; name: string; threshold: number }): CategoricalSlot

Overrides Slot.constructor
- Defined in slots/categorical_slot.ts:49
Parameters
- __namedParameters: { categories: {}; dependantActions: string[]; invDependantActions: string[]; name: string; threshold: number }
  - categories: {}
    - [category: string]: string[]
  - dependantActions: string[]
  - invDependantActions: string[]
  - name: string
  - threshold: number
Returns CategoricalSlot

Properties

Protected actions

actions: any[]

The list of every action the model can take.

Private categories

categories: Categories

Private categoryNames

categoryNames: string[]

Readonly id

id: string

Readonly size

size: number

Private threshold

threshold: number

Methods

export

export(): Promise<JSONSerializable>

Inherited from Featurizer.export
- Defined in featurizers/featurizer.ts:102
Export the featurizer's internal parameters to be serialized along the model.

Returns Promise<JSONSerializable>

getActionMask

getActionMask(): boolean[]

Inherited from Slot.getActionMask

Overrides Featurizer.getActionMask
- Defined in slots/slot.ts:35
Returns boolean[]

getOptimizableFeatures

getOptimizableFeatures(data: any): tf.Tensor1D

Inherited from Featurizer.getOptimizableFeatures
- Defined in featurizers/featurizer.ts:62
Turn the data returned by handleQuery into an embedding vector. This function is used to expose featurizer variables to the model optimizer for training.

Reimplementing this method is not necessary if your featurizer is not meant to be optimizable through gradient descent. In this case, just return the feature vector directly using the handleQuery method.

remarks

It's important to keep this function stateless, it should only depend of its tensor argument and of featurizer's variables.

Parameters
- data: any
Returns tf.Tensor1D

getValue

getValue(): CategoricalValue

Overrides Slot.getValue
- Defined in slots/categorical_slot.ts:129
Returns CategoricalValue

handleAction

handleAction(action: any): void

Inherited from Featurizer.handleAction
- Defined in featurizers/featurizer.ts:74
Let the featurizer know what action the model has taken.

Parameters
- action: any
Returns void

handleQuery

handleQuery(query: string): Promise<tf.Tensor1D>

Overrides Featurizer.handleQuery
- Defined in slots/categorical_slot.ts:82
Parameters
- query: string
Returns Promise<tf.Tensor1D>

init

init(actions: any[]): Promise<void>

Overrides Featurizer.init
- Defined in slots/categorical_slot.ts:68
Parameters
- actions: any[]
Returns Promise<void>

load

load(parameters: JSONSerializable): void

Inherited from Featurizer.load
- Defined in featurizers/featurizer.ts:96
Load parameters extracted from a JSON-like document.

Parameters
- parameters: JSONSerializable
Returns void

Private oneHotValue

oneHotValue(value: CategoricalValue): tf.Tensor1D

- Defined in slots/categorical_slot.ts:73
Parameters
- value: CategoricalValue
Returns tf.Tensor1D

resetDialog

resetDialog(): void

Inherited from Slot.resetDialog

Overrides Featurizer.resetDialog
- Defined in slots/slot.ts:46
Returns void

setValue

setValue(value: CategoricalValue): void

Inherited from Slot.setValue
- Defined in slots/slot.ts:60
Redefine a new value for the slot.

Parameters
- value: CategoricalValue
Returns void

Generated using TypeDoc