Class A2CAgent

constructor

• new A2CAgent(
      env: RLEnvironmentBase,
      resolution: number,
      procs: number,
      layers: LayerObject[],
      optimizer: string,
  ): A2CAgent

  Parameters

  • env: RLEnvironmentBase

    Environment
  • resolution: number

    Resolution of actions
  • procs: number

    Number of processes
  • layers: LayerObject[]

    Network layers
  • optimizer: string

    Optimizer of the network

  Returns A2CAgent

_advanced_step

_env

_envs

_gamma

_net

_procs

_states

get_action

• get_action(state: any[]): any[]

  Returns a action.

  Parameters

  • state: any[]

    Current states

  Returns any[]

  Action

get_score

• get_score(): number[][][]

  Returns a score.

  Returns number[][][]

  Score values

terminate

• terminate(): void

  Returns void

update

• update(done: boolean, learning_rate: number, batch: number): number

  Update model.

  Parameters

  • done: boolean

    Done epoch or not
  • learning_rate: number

    Learning rate
  • batch: number

    Batch size

  Returns number

  Loss value