Fix num_envs=test_num (#653)

* fix num_envs=test_num

* fix mypy

examples/atari/atari_wrapper.py

@@ -279,7 +279,7 @@ def make_atari_env(task, seed, training_num, test_num, **kwargs):
         )
         test_envs = envpool.make_gym(
             task.replace("NoFrameskip-v4", "-v5"),
-            num_envs=training_num,
+            num_envs=test_num,
             seed=seed,
             episodic_life=False,
             reward_clip=False,

examples/mujoco/mujoco_env.py

@@ -19,7 +19,7 @@ def make_mujoco_env(task, seed, training_num, test_num, obs_norm):
     """
     if envpool is not None:
         train_envs = env = envpool.make_gym(task, num_envs=training_num, seed=seed)
-        test_envs = envpool.make_gym(task, num_envs=training_num, seed=seed)
+        test_envs = envpool.make_gym(task, num_envs=test_num, seed=seed)
     else:
         warnings.warn(
             "Recommend using envpool (pip install envpool) "

tianshou/data/buffer/base.py

@@ -347,7 +347,7 @@ class ReplayBuffer:
             indices = self.sample_indices(0) if index == slice(None) \
                 else self._indices[:len(self)][index]
         else:
-            indices = index
+            indices = index  # type: ignore
         # raise KeyError first instead of AttributeError,
         # to support np.array([ReplayBuffer()])
         obs = self.get(indices, "obs")

tianshou/data/buffer/prio.py

@@ -94,7 +94,7 @@ class PrioritizedReplayBuffer(ReplayBuffer):
             indices = self.sample_indices(0) if index == slice(None) \
                 else self._indices[:len(self)][index]
         else:
-            indices = index
+            indices = index  # type: ignore
         batch = super().__getitem__(indices)
         weight = self.get_weight(indices)
         # ref: https://github.com/Kaixhin/Rainbow/blob/master/memory.py L154