DeepMind Control Suite: dm_control 1.0.13

dm_control: DeepMind Infrastructure for Physics-Based Simulation

DeepMind的软件堆栈，用于基于物理的模拟和强化学习环境，使用MuJoCo物理。

1、基准任务

from dm_control import suite
import numpy as np

domain_name = 'walker'
task_name = 'walk'
env = suite.load(domain_name, task_name)

action_spec = env.action_spec()
time_step = env.reset()

action = np.random.uniform(action_spec.minimum,
                           action_spec.maximum,
                           size=action_spec.shape)
time_step = env.step(action)
print(time_step.last(), time_step.reward, time_step.discount, time_step.observation)

• acrobot swingup/swingup_sparse
  • orientations: array(4)
    
  • velocity: array(2)
    
• ball_in_cup catch
  • position: array(4)
  • velocity: array(4)
• cartpole balance/balance_sparse/swingup/swingup_sparse
  • position: array(3)
  • velocity: array(2)
• cheetah run
  • position: array(8)
  • velocity: array(9)
• finger spin
  • position: array(4)
  • velocity: array(3)
  • touch: array(2)
• finger turn_easy/turn_hard
  • position: array(4)
  • velocity: array(3)
  • touch: array(2)
  • target_position: array(2)
  • dist_to_target: float
• fish upright
  • joint_angles: array(7)
    
  • upright: float
  • velocity: array(13)
• fish swim
  • joint_angles: array(7)
    
  • upright: float
  • target: array(3)
  • velocity: array(13)
• hopper stand/hop
  • position: array(6)
    
  • velocity: array(7)
  • touch: array(2)
• humanoid stand/walk/run
  • joint_angles: array(21)
    
  • head_height: float
  • extremities: array(12)
  • torso_vertical: array(3)
  • com_velocity: array(3)
  • velocity: array(27)
• manipulator bring_ball
  • arm_pos: array(16)
  • arm_vel: array(8)
  • touch: array(5)
  • hand_pos: array(4)
  • object_pos: array(4)
  • object_vel: array(3)
  • target_pos: array(4)
• pendulum swingup
  • orientation: array(2)
  • velocity: array(1)
• point_mass easy
  • position: array(2)
  • velocity: array(2)
• reacher easy/hard
  • position: array(2)
  • to_target: array(2)
  • velocity: array(2)
• swimmer swimmer6
  • joints: array(5)
  • to_target: array(2)
  • body_velocities: array(18)
• swimmer swimmer15
  • joints: array(14)
  • to_target: array(2)
  • body_velocities: arrray(45)
• walker stand/walk/run
  • orientations: array(14)
  • height: float
  • velocity: array(9)