wassname/DeepRL
1
0 Fork 0
mirror of https://github.com/wassname/DeepRL.git synced 2026-10-29 12:40:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

Specify a gpu for a network

Browse Source
This commit is contained in:
Shangtong Zhang committed 2018-02-03 10:53:14 -07:00
1 parent fef04b6bf6
commit e91ec3be45
12 files changed
+93 -103
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

async_worker/dpg.py
+2 -2
View File
@@ -78,8 +78,8 @@ class DeterministicPolicyGradient:
experiences = self.replay.sample()
states, actions, rewards, next_states, terminals = experiences
q_next = target_critic.predict(next_states, target_actor.predict(next_states))
terminals = critic.to_torch_variable(terminals).unsqueeze(1)
rewards = critic.to_torch_variable(rewards).unsqueeze(1)
terminals = critic.variable(terminals).unsqueeze(1)
rewards = critic.variable(rewards).unsqueeze(1)
q_next = config.discount * q_next * (1 - terminals)
q_next.add_(rewards)
q_next = q_next.detach()