Class KOPLS

constructor

• new KOPLS(
      options: {
          kernel?: object;
          orthogonalComponents?: number;
          predictiveComponents?: number;
      },
      model: object,
  ): KOPLS

  Constructor for Kernel-based Orthogonal Projections to Latent Structures (K-OPLS)

  Parameters

  • options: {
        kernel?: object;
        orthogonalComponents?: number;
        predictiveComponents?: number;
    }

    constructor options.

    • Optionalkernel?: object

      Kernel object to apply, see ml-kernel.

    • OptionalorthogonalComponents?: number

      Number of Y-Orthogonal components.

    • OptionalpredictiveComponents?: number

      Number of predictive components to use.

  • model: object

    for load purposes.

  Returns KOPLS

kernel

kernelX

orthogonalComp

predictiveComp

predScoreMat

SigmaPow

toNorm

trainingSet

TURegressionCoeff

YLoadingMat

YOrthEigen

YOrthLoadingVec

YOrthScoreMat

YScoreMat

predict

• predict(
      toPredict: any[] | Matrix,
  ): { predScoreMat: Matrix[]; predYOrthVectors: Matrix[]; y: Matrix }

  Predicts the output given the matrix to predict.

  Parameters

  • toPredict: any[] | Matrix

    matrix of features to predict.

  Returns { predScoreMat: Matrix[]; predYOrthVectors: Matrix[]; y: Matrix }

  predictions

toJSON

• toJSON(): object

  Export the current model to JSON.

  Returns object

  • Current model.

train

• train(trainingSet: any[] | Matrix, trainingValues: any[] | Matrix): void

  Train the K-OPLS model with the given training set and labels.

  Parameters

  • trainingSet: any[] | Matrix

    matrix of features.

  • trainingValues: any[] | Matrix

    matrix of labels.

  Returns void

Staticload

• load(model: object, kernel: object): KOPLS

  Load a K-OPLS with the given model.

  Parameters

  • model: object

    the serialized model to load.

  • kernel: object

    kernel used on the model, see ml-kernel.

  Returns KOPLS

  the loaded K-OPLS model.