Class GBRBM

constructor

• new GBRBM(hiddenSize: number, lr?: number, fixSigma?: boolean): GBRBM

• Parameters

  • hiddenSize: number

    Size of hidden layer
  • Optionallr: number

    Learning rate
  • OptionalfixSigma: boolean

    Do not learn sigma or not

  Returns GBRBM

_b

_c

_fixSigma

_hidden

_k

_lr

_visible

_w

_z

_s

• get _s(): number[]

• Returns number[]

_h

• _h(v: any, sample?: boolean): any[][]

• Parameters

  • v: any
  • Optionalsample: boolean

  Returns any[][]

_randn

• _randn(): number

• Returns number

_v

• _v(h: any, sample?: boolean): any[][]

• Parameters

  • h: any
  • Optionalsample: boolean

  Returns any[][]

energy

• energy(v: number[], h: number[]): number

• Return a energy value of the data.

  Parameters

  • v: number[]

    Sample data
  • h: number[]

    Target values

  Returns number

  Energy value

fit

• fit(x: number[][]): void

• Fit model.

  Parameters

  • x: number[][]

    Training data

  Returns void

predict

• predict(x: number[][]): number[][]

• Returns predicted values.

  Parameters

  • x: number[][]

    Sample data

  Returns number[][]

  Predicted values