@ai-on-browser/data-analysis-models
    Preparing search index...

    Class LadderNetwork

    Ladder network

    Index

    Constructors

    constructor

    Properties

    _activation _classes _epoch _hidden_sizes _lambdas _layers _model _noise_std _optimizer

    Accessors

    epoch

    Methods

    _build fit predict

    Constructors

    • Parameters

      • hidden_sizes: number[]

        Sizes of hidden layers

      • lambdas: number[]

        Regularization parameters

      • activation: string

        Activation name

      • optimizer: string

        Optimizer of the network

      Returns LadderNetwork

    Properties

    _activation

    _activation: string

    _classes

    _classes: any[]

    _epoch

    _epoch: number

    _hidden_sizes

    _hidden_sizes: number[]

    _lambdas

    _lambdas: number[]

    _layers

    _layers: (
        | {
            axis?: undefined;
            input?: undefined;
            name: string;
            size?: undefined;
            type: string;
            variance?: undefined;
        }
        | {
            axis?: undefined;
            input?: undefined;
            name: string;
            size: string;
            type: string;
            variance: number;
        }
        | {
            axis?: undefined;
            input: string[];
            name: string;
            size?: undefined;
            type: string;
            variance?: undefined;
        }
        | {
            axis?: undefined;
            input: string;
            name: string;
            size?: undefined;
            type: string;
            variance?: undefined;
        }
        | {
            axis: number;
            input: string[];
            name?: undefined;
            size?: undefined;
            type: string;
            variance?: undefined;
        }
    )[]

    _model

    _model: NeuralNetwork

    _noise_std

    _noise_std: any[]

    _optimizer

    _optimizer: string

    Accessors

    • get epoch(): number

      Epoch

      Returns number

    Methods

    _build

    • Fit model.

      Parameters

      • train_x: number[][]

        Training data

      • train_y: any[]

        Target values

      • iteration: number

        Iteration count

      • rate: number

        Learning rate

      • batch: number

        Batch size

      Returns { labeledLoss: number; unlabeledLoss: number }

      Loss value

    • Returns predicted values.

      Parameters

      • x: number[][]

        Sample data

      Returns any[]

      Predicted values

    Settings

    Member Visibility

    On This Page

    Constructors
    Properties
    Accessors
    Methods