gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-17 11:10:20 +01:00
Code Issues Projects Releases Wiki Activity

fix qr_dqn_agent

Browse Source
This commit is contained in:
Zach Dwiel
2018-02-16 20:32:50 -05:00
parent e1ad86417f
commit d8f5a35013
2 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
agents/actor_critic_agent.py
View File

@@ -101,9 +101,7 @@ class ActorCriticAgent(PolicyOptimizationAgent):
actions = np.expand_dims(actions, -1)
# train
inputs = copy.copy(current_states)
inputs['output_1_0'] = actions
result = self.main_network.online_network.accumulate_gradients(inputs,
result = self.main_network.online_network.accumulate_gradients({**current_states, 'output_1_0': actions},
[state_value_head_targets, action_advantages])
# logging