Closed Sid-Bhatia-0 closed 4 years ago
Added rewards into the CollectGems env struct. The field r will maintain the step reward received from the immediately previous action.
CollectGems
r
Overloaded the default method for TURN_LEFT and TURN_RIGHT actions in order to update the env.r field on every action taken.
TURN_LEFT
TURN_RIGHT
env.r
Not re-implementing a time penalty for each step taken. Can incorporate this feature using RewardOverriddenEnv
Added rewards into the
CollectGemsenv struct. The fieldrwill maintain the step reward received from the immediately previous action.Overloaded the default method for
TURN_LEFTandTURN_RIGHTactions in order to update theenv.rfield on every action taken.Not re-implementing a time penalty for each step taken. Can incorporate this feature using RewardOverriddenEnv