Update documentation

environment/mujoco/hopper/Hopper.go

@@ -23,19 +23,19 @@ import (
 // agent can control three of its joints to hop or move around. The
 // movable joints are the thigh joint, leg join, and foot joint.
 //
-// The state observation space is a vector of 11 components:
-// [
-// 		torso Z position
-//		torso Y orientation
-//		thigh -Y orientation
-//		leg -Y orientation
-//		foot -Y orientation
-//		torso X linear velocity
-//		torso Z linear velocity
-//		torso Y anglular velocity
-//		thigh Y angular velocity
-//		leg Y angular velocity
-//		foot Y angular velocity
+// The state observation space is a vector of 11 components consisting
+// of the following features in the following order
+//		1.  Torso Z position
+//		2.  Torso Y orientation
+//		3.  Thigh -Y orientation
+//		4.  Leg -Y orientation
+//		5.  Foot -Y orientation
+//		6.  Torso X linear velocity
+//		7.  Torso Z linear velocity
+//		8.  Torso Y anglular velocity
+//		9.  Thigh Y angular velocity
+//		10. Leg Y angular velocity
+//		11. Foot Y angular velocity
 // ]
 // where +/-Y denotes rotation about the positive or negative Y axis
 // respectively. All features of the state space are unbounded.

environment/mujoco/reacher/Reacher.go

@@ -26,20 +26,20 @@ import (
 // is denoted θ2.
 //
 // State observations are 11-dimensional vectors and consist of the
-// following features:
-// [
-// 	cos(θ1)
-// 	cos(θ2)
-// 	sin(θ1)
-// 	sin(θ2)
-//	target x
-//	target y
-//	θ1 angular velocity
-//	θ2 angular velocity
-//	x distance(fingertip, target)
-//	y distance(fingertip, target)
-//	z distance(fingertip, target)
-// ]
+// following features in the following order:
+//
+// 		1.  Cos(θ1)
+// 		2.  Cos(θ2)
+// 		3.  Sin(θ1)
+// 		4.  Sin(θ2)
+//		5.  Target x
+//		6.  Target y
+//		7.  θ1 angular velocity
+//		8.  θ2 angular velocity
+//		9.  X distance(fingertip, target)
+//		10. Y distance(fingertip, target)
+//		11. Z distance(fingertip, target)
+//
 //
 // Actions are 2-dimensional, continuous vectors consisting of the
 // rotation to apply to θ1 and θ2. Actions are clipped to stay between