gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-17 19:20:19 +01:00
Code Issues Projects Releases Wiki Activity

remove unused parameter scale_external_reward_by_intrinsic_reward_value

Browse Source
This commit is contained in:
zach dwiel
2019-04-04 11:54:18 -04:00
committed by Zach Dwiel
parent 881f78f45a
commit 7d79433c05
3 changed files with 3 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
rl_coach/agents/agent.py
View File

@@ -895,10 +895,7 @@ class Agent(AgentInterface):
transition = self.update_transition_before_adding_to_replay_buffer(transition)
# merge the intrinsic reward in
if self.ap.algorithm.scale_external_reward_by_intrinsic_reward_value:
transition.reward = transition.reward * (1 + self.last_action_info.action_intrinsic_reward)
else:
transition.reward = transition.reward + self.last_action_info.action_intrinsic_reward
transition.reward = transition.reward + self.last_action_info.action_intrinsic_reward
# sum up the total shaped reward
self.total_shaped_reward_in_current_episode += transition.reward
@@ -1026,7 +1023,7 @@ class Agent(AgentInterface):
self.total_reward_in_current_episode += transition.reward
self.shaped_reward.add_sample(transition.reward)
self.reward.add_sample(transition.reward)
# create and store the transition
if self.phase in [RunPhase.TRAIN, RunPhase.HEATUP]:
# for episodic memories we keep the transitions in a local buffer until the episode is ended.