removed unnecessary state gather

ranpku · Sep 15, 2019 · a912c72 · a912c72
1 parent bc8e8a8
commit a912c72
Showing 1 changed file with 0 additions and 1 deletion.
diff --git a/agents/actor_critic_agents/SAC_Discrete.py b/agents/actor_critic_agents/SAC_Discrete.py
@@ -71,7 +71,6 @@ def calculate_critic_losses(self, state_batch, action_batch, reward_batch, next_
             min_qf_next_target = action_probabilities * (torch.min(qf1_next_target, qf2_next_target) - self.alpha * log_action_probabilities)
             min_qf_next_target = min_qf_next_target.mean(dim=1).unsqueeze(-1)
             next_q_value = reward_batch + (1.0 - mask_batch) * self.hyperparameters["discount_rate"] * (min_qf_next_target)
-            self.critic_target(next_state_batch).gather(1, next_state_action.unsqueeze(-1).long())
 
         qf1 = self.critic_local(state_batch).gather(1, action_batch.long())
         qf2 = self.critic_local_2(state_batch).gather(1, action_batch.long())