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

Remove unnecessary conversion

Browse Source
This commit is contained in:
Shangtong Zhang
2017-11-14 19:14:42 -07:00
parent 65eea71bdb
commit 4697d1d3e9
1 changed files with 2 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
async_worker/ppo.py
+2 -4
View File
@@ -95,12 +95,10 @@ class ProximalPolicyOptimization:
values.append(actor_net.to_torch_variable(R))
A = actor_net.to_torch_variable(torch.zeros((1, 1)))
discount = actor_net.to_torch_variable([self.config.discount])
gae_tau = actor_net.to_torch_variable([self.config.gae_tau])
for i in reversed(range(len(rewards))):
R = actor_net.to_torch_variable([[rewards[i]]])
ret = R + discount * values[i + 1]
A = ret - values[i] + discount * gae_tau * A
ret = R + self.config.discount * values[i + 1]
A = ret - values[i] + self.config.discount * self.config.gae_tau * A
advantages.append(A.detach())
returns.append(ret.detach())
advantages = list(reversed(advantages))