Class GPLVM

constructor

• new GPLVM(
      rd: number,
      alpha: number,
      ez?: number,
      ea?: number,
      ep?: number,
      kernel?: "gaussian" | { a?: number; b?: number; name: "gaussian" },
  ): GPLVM

  Parameters

  • rd: number

    Reduced dimension
  • alpha: number

    Precision parameter
  • Optionalez: number

    Learning rate for z
  • Optionalea: number

    Learning rate for alpha
  • Optionalep: number

    Learning rate for kernel
  • Optionalkernel: "gaussian" | { a?: number; b?: number; name: "gaussian" }

    Kernel name

  Returns GPLVM

_alpha

_ea

_ez

_kernel

_rd

_s

_x

_z

fit

• fit(): void

  Fit model.

  Returns void

init

• init(x: number[][]): void

  Initialize model.

  Parameters

  • x: number[][]

    Training data

  Returns void

llh

• llh(): number

  Returns log likelihood.

  Returns number

  Log likelihood

predict

• predict(): number[][]

  Returns reduced datas.

  Returns number[][]

  Predicted values

reconstruct

• reconstruct(z: number[][]): number[][]

  Returns reconstruct datas.

  Parameters

  • z: number[][]

    Sample data

  Returns number[][]

  Predicted values