reinforcement-learning - 在 A2C 中创建分离的演员和评论家模型，ray[rrlib]

Question

是否可以从配置文件中为 A2C 中的演员和批评者模型分离模型，如以下配置字典？

      # Model options for the actor.
      "Actor_model": {
          "fcnet_activation": "relu",
          "fcnet_hiddens": [256, 256],
      },
      # Model options for the critic.
      "critic_model": {
          "fcnet_activation": "relu",
          "fcnet_hiddens": [128, 128],
      },

标签： reinforcement-learningray