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

fix(rllib): allow explore=False with tuple action distributions (#10443)

Browse source
This commit is contained in:
Thomas Lecat 2020-09-11 00:03:02 +02:00 • committed by GitHub
parent 7aa55ca82c
commit 504da45e69
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
rllib/utils/exploration/stochastic_sampling.py
View file

@ -72,5 +72,5 @@ class StochasticSampling(Exploration):
logp = action_dist.sampled_action_logp()
else:
action = action_dist.deterministic_sample()
logp = torch.zeros((action.size()[0], ), dtype=torch.float32)
logp = torch.zeros_like(action_dist.sampled_action_logp())
return action, logp