dm_control: environments
task list
timestep object
step_type
reward
discount
observation: OrderedDict
Environments (<domain_name>-<task_name>)
table: acrobot-swingup
orientations 4
velocity 2
table: ball_in_cup-catch
position 4
velocity 4
table: cartpole-balance | swingup
position 3
velocity 2
table:cheetah-run
position 8
velocity 9
table: finger-spin
position 4
velocity 3
touch 2
table: fish-swim
joint_angles 7
upright 1
target 3
velocity 13
table: fish-upright
joint_angles 7
upright 1
velocity 13
table: hopper-hop | stand
position 6
velocity 7
touch 2
table: humanoid-stand | walk | run
joint_angles 21
head_height 1
extremities 12
torso_vertical 3
com_velocity 3
velocity 27
table: humanoid_CMU-stand
joint_angles 48
head_height 1
extremities 12
torso_vertical 3
com_velocity 3
velocity 62
table: pendulum-swingup
orientation 2
velocity 1
table:point_mass-easy
position 2
velocity 2
table:quadruped-walk
egocentric_state 44
torso_velocity 3
torso_upright 1
imu 6
force_torque 24
table:reacher-easy
position 2
to_target 2
velocity 2
table:swimmer-swimmer6
joints 5
to_target 2
body_velocities 18
table: walker-run | stand | walk
orientations 14
height 1
velocity 9