Class DDPGAgent

constructor

• new DDPGAgent(
      env: RLEnvironmentBase,
      resolution: number,
      layers: LayerObject[],
      optimizer: string,
  ): DDPGAgent

  Parameters

  • env: RLEnvironmentBase

    Environment

  • resolution: number

    Resolution of actions

  • layers: LayerObject[]

    Network layers

  • optimizer: string

    Optimizer of the network

  Returns DDPGAgent

_env

_net

get_action

• get_action(state: any[], factor: number): any[]

  Returns a action.

  Parameters

  • state: any[]

    Current states

  • factor: number

    Random factor

  Returns any[]

  Action

get_score

• get_score(): number[][][]

  Returns a score.

  Returns number[][][]

  Score values

terminate

• terminate(): void

  Returns void

update

• update(
      action: any[],
      state: any[],
      next_state: any[],
      reward: number,
      learning_rate: number,
      batch: number,
  ): number | undefined

  Update model.

  Parameters

  • action: any[]

    Action

  • state: any[]

    Current states

  • next_state: any[]

    Next states

  • reward: number

    Reward

  • learning_rate: number

    Learning rate

  • batch: number

    Batch size

  Returns number | undefined

  Loss value