hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 10:31:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

Use observation filter in compute_action for PPO. (#884)

Browse source
This commit is contained in:
Robert Nishihara 2017-08-28 23:01:29 -07:00 • committed by Philipp Moritz
parent 5d72818ddc
commit 60d4d01d06
1 changed files with 1 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
python/ray/rllib/policy_gradient/policy_gradient.py
View file

@ -266,4 +266,5 @@ class PolicyGradient(Algorithm):
for (a, o) in zip(self.agents, extra_data[4])])
def compute_action(self, observation):
observation = self.model.observation_filter(observation)
return self.model.common_policy.compute([observation])[0][0]