Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Impala CNN + ProcGen #298

Open
wants to merge 9 commits into
base: master
Choose a base branch
from
Open

Impala CNN + ProcGen #298

Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
5c34b0e
Imapala CNN with PPO shared value-function+policy layers
burgalon Oct 12, 2020
16965f5
tqdm auto
burgalon Oct 12, 2020
c4de480
fix batch sampling
burgalon Oct 12, 2020
4f40e62
gpu
burgalon Oct 12, 2020
3f7abe3
gpu fix
burgalon Oct 12, 2020
a5945bf
gpu fix
burgalon Oct 12, 2020
e404b39
gpu fix
burgalon Oct 12, 2020
447cd16
do not quit on KL divergence before 1 step
burgalon Oct 12, 2020
a67e163
stop KL for debugging
burgalon Oct 12, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
gpu fix
  • Loading branch information
@burgalon
burgalon committed Oct 12, 2020
commit a5945bfe42a7666380c5d38a893e810715e09755
2 changes: 1 addition & 1 deletion spinup/algos/pytorch/ppo_shared/impala_cnn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,7 +117,7 @@ def step(self, obs):
a = pi.sample()
logp_a = self.pi._log_prob_from_distribution(pi, a)
v = self.v(obs)
return a.numpy(), v.numpy(), logp_a.numpy()
return a.cpu().numpy(), v.cpu().numpy(), logp_a.cpu().numpy()

def act(self, obs):
return self.step(obs)[0]