Class RBM

constructor

• new RBM(hiddenSize: number, lr?: number): RBM

  Parameters

  • hiddenSize: number

    Size of hidden layer
  • Optionallr: number

    Learning rate

  Returns RBM

_a

_b

_hidden

_k

_lr

_visible

_w

_h

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

  Parameters

  • v: any
  • Optionalsample: boolean

  Returns any[][]

_normalize

• _normalize(x: any): void

  Parameters

  • x: any

  Returns void

_sgm

• _sgm(x: any): number

  Parameters

  • x: any

  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[]): (0 | 1)[]

  Returns predicted values.

  Parameters

  • x: number[]

    Sample data

  Returns (0 | 1)[]

  Predicted values