typo in reinforcement_policy_trainer; fixes #161

dongbin · Sep 21, 2016 · 8cdd4aa · 8cdd4aa
1 parent d4f03c5
commit 8cdd4aa
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/AlphaGo/training/reinforcement_policy_trainer.py b/AlphaGo/training/reinforcement_policy_trainer.py
@@ -173,7 +173,7 @@ def run_n_games(optimizer, learner, opponent, num_games):
 		current, other = other, current
 
 	# Return the win ratio.
-	wins = sum(state.get_winner == pc for (state, pc) in zip(states, learner_color))
+	wins = sum(state.get_winner() == pc for (state, pc) in zip(states, learner_color))
 	return float(wins) / num_games