[rllib] Native support for Dict and Tuple spaces; fix Tuple action spaces; add prev a, r to LSTM (#3051)

2026-07-28 11:25:04 +08:00 · 2018-10-20 15:21:22 -07:00
parent 9a2b5333ef
commit 59901a88a0
39 changed files with 921 additions and 185 deletions
@@ -88,7 +88,7 @@ class PPOAgent(Agent):
            self.optimizer = SyncSamplesOptimizer(
                self.local_evaluator, self.remote_evaluators, {
                    "num_sgd_iter": self.config["num_sgd_iter"],
-                    "train_batch_size": self.config["train_batch_size"]
+                    "train_batch_size": self.config["train_batch_size"],
                })
        else:
            self.optimizer = LocalMultiGPUOptimizer(