Class VAE

constructor

• new VAE(
      in_size: number,
      noise_dim: number,
      enc_layers: LayerObject[],
      dec_layers: LayerObject[],
      optimizer: string,
      class_size: number,
      type: "" | "conditional",
  ): VAE

• Parameters

  • in_size: number

    Input size
  • noise_dim: number

    Number of noise dimension
  • enc_layers: LayerObject[]

    Layers of encoder
  • dec_layers: LayerObject[]

    Layers of decoder
  • optimizer: string

    Optimizer of the network
  • class_size: number

    Class size for conditional type
  • type: "" | "conditional"

    Type name

  Returns VAE

_aeNet

_decodeNet

_epoch

_reconstruct_rate

_type

epoch

• get epoch(): number

• Epoch

  Returns number

fit

• fit(
      x: number[][],
      y: number[][],
      iteration: number,
      rate: number,
      batch: number,
  ): number

• Fit model.

  Parameters

  • x: number[][]

    Training data
  • y: number[][]

    Conditional values
  • iteration: number

    Iteration count
  • rate: number

    Learning rate
  • batch: number

    Batch size

  Returns number

  Loss value

predict

• predict(x: number[][], y: number[][]): number[][]

• Returns predicted values.

  Parameters

  • x: number[][]

    Sample data
  • y: number[][]

    Conditional values

  Returns number[][]

  Predicted values

reduce

• reduce(x: number[][], y: number[][]): number[][]

• Returns predicted values.

  Parameters

  • x: number[][]

    Sample data
  • y: number[][]

    Conditional values

  Returns number[][]

  Predicted values