wassname/DeepRL
1
0
Fork 0
mirror of https://github.com/wassname/DeepRL.git synced 2026-09-09 11:13:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactor DDPG

Browse Source
This commit is contained in:
Shangtong Zhang
2017-10-06 11:16:21 -06:00
parent 8b9fd8d24f
commit c5cbc10f94
7 changed files with 167 additions and 110 deletions
Download Patch File Download Diff File Expand all files Collapse all files
async_worker/continuous_actor_critic.py
-9
View File
@@ -46,15 +46,6 @@ class ContinuousAdvantageActorCritic:
terminal = (terminal or (config.max_episode_length and steps >= config.max_episode_length))
next_state = self.state_normalizer(next_state)
# if deterministic:
# self.config.logger.scalar_summary('reward', reward, self.counter)
# self.config.logger.histo_summary('std', std.data.numpy(), self.counter)
# self.config.logger.histo_summary('mean', mean.data.numpy(), self.counter)
# self.config.logger.histo_summary('action', action, self.counter)
# self.config.logger.scalar_summary('steps', steps, self.counter)
# self.config.logger.histo_summary('states', state, self.counter)
# self.counter += 1
steps += 1
total_reward += reward
reward = np.asscalar(self.reward_normalizer(np.array([reward])))