Auto-upgrade syntax via pyupgrade (#158)

* Auto-upgrade syntax via `pyupgrade`

* Add changes

Author: vwxyzjn

.pre-commit-config.yaml

@@ -1,4 +1,10 @@
 repos:
+  - repo: https://github.com/asottile/pyupgrade
+    rev: v2.31.1
+    hooks:
+      - id: pyupgrade
+        args: 
+          - --py37-plus
   - repo: https://github.com/PyCQA/isort
     rev: 5.10.1
     hooks:

cleanrl/c51.py

@@ -89,7 +89,7 @@ def thunk():
 # ALGO LOGIC: initialize agent here:
 class QNetwork(nn.Module):
     def __init__(self, env, n_atoms=101, v_min=-100, v_max=100):
-        super(QNetwork, self).__init__()
+        super().__init__()
         self.env = env
         self.n_atoms = n_atoms
         self.register_buffer("atoms", torch.linspace(v_min, v_max, steps=n_atoms))

cleanrl/c51_atari.py

@@ -105,7 +105,7 @@ def thunk():
 # ALGO LOGIC: initialize agent here:
 class QNetwork(nn.Module):
     def __init__(self, env, n_atoms=101, v_min=-100, v_max=100):
-        super(QNetwork, self).__init__()
+        super().__init__()
         self.env = env
         self.n_atoms = n_atoms
         self.register_buffer("atoms", torch.linspace(v_min, v_max, steps=n_atoms))

cleanrl/ddpg_continuous_action.py

@@ -86,7 +86,7 @@ def thunk():
 # ALGO LOGIC: initialize agent here:
 class QNetwork(nn.Module):
     def __init__(self, env):
-        super(QNetwork, self).__init__()
+        super().__init__()
         self.fc1 = nn.Linear(np.array(env.single_observation_space.shape).prod() + np.prod(env.single_action_space.shape), 256)
         self.fc2 = nn.Linear(256, 256)
         self.fc3 = nn.Linear(256, 1)
@@ -101,7 +101,7 @@ def forward(self, x, a):
 
 class Actor(nn.Module):
     def __init__(self, env):
-        super(Actor, self).__init__()
+        super().__init__()
         self.fc1 = nn.Linear(np.array(env.single_observation_space.shape).prod(), 256)
         self.fc2 = nn.Linear(256, 256)
         self.fc_mu = nn.Linear(256, np.prod(env.single_action_space.shape))

cleanrl/dqn.py

@@ -84,7 +84,7 @@ def thunk():
 # ALGO LOGIC: initialize agent here:
 class QNetwork(nn.Module):
     def __init__(self, env):
-        super(QNetwork, self).__init__()
+        super().__init__()
         self.network = nn.Sequential(
             nn.Linear(np.array(env.single_observation_space.shape).prod(), 64),
             nn.Tanh(),

cleanrl/dqn_atari.py

@@ -100,7 +100,7 @@ def thunk():
 # ALGO LOGIC: initialize agent here:
 class QNetwork(nn.Module):
     def __init__(self, env):
-        super(QNetwork, self).__init__()
+        super().__init__()
         self.network = nn.Sequential(
             nn.Conv2d(4, 32, 8, stride=4),
             nn.ReLU(),

cleanrl/ppg_procgen.py

@@ -142,7 +142,7 @@ def get_output_shape(self):
 
 class Agent(nn.Module):
     def __init__(self, envs):
-        super(Agent, self).__init__()
+        super().__init__()
         h, w, c = envs.single_observation_space.shape
         shape = (c, h, w)
         conv_seqs = []

cleanrl/ppo.py

@@ -103,7 +103,7 @@ def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
 
 class Agent(nn.Module):
     def __init__(self, envs):
-        super(Agent, self).__init__()
+        super().__init__()
         self.critic = nn.Sequential(
             layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),
             nn.Tanh(),

cleanrl/ppo_atari.py

@@ -117,7 +117,7 @@ def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
 
 class Agent(nn.Module):
     def __init__(self, envs):
-        super(Agent, self).__init__()
+        super().__init__()
         self.network = nn.Sequential(
             layer_init(nn.Conv2d(4, 32, 8, stride=4)),
             nn.ReLU(),

cleanrl/ppo_atari_envpool.py

@@ -81,7 +81,7 @@ def parse_args():
 
 class RecordEpisodeStatistics(gym.Wrapper):
     def __init__(self, env, deque_size=100):
-        super(RecordEpisodeStatistics, self).__init__(env)
+        super().__init__(env)
         self.num_envs = getattr(env, "num_envs", 1)
         self.episode_returns = None
         self.episode_lengths = None
@@ -94,7 +94,7 @@ def __init__(self, env, deque_size=100):
             print("env has lives")
 
     def reset(self, **kwargs):
-        observations = super(RecordEpisodeStatistics, self).reset(**kwargs)
+        observations = super().reset(**kwargs)
         self.episode_returns = np.zeros(self.num_envs, dtype=np.float32)
         self.episode_lengths = np.zeros(self.num_envs, dtype=np.int32)
         self.lives = np.zeros(self.num_envs, dtype=np.int32)
@@ -103,7 +103,7 @@ def reset(self, **kwargs):
         return observations
 
     def step(self, action):
-        observations, rewards, dones, infos = super(RecordEpisodeStatistics, self).step(action)
+        observations, rewards, dones, infos = super().step(action)
         self.episode_returns += infos["reward"]
         self.episode_lengths += 1
         self.returned_episode_returns[:] = self.episode_returns
@@ -133,7 +133,7 @@ def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
 
 class Agent(nn.Module):
     def __init__(self, envs):
-        super(Agent, self).__init__()
+        super().__init__()
         self.network = nn.Sequential(
             layer_init(nn.Conv2d(4, 32, 8, stride=4)),
             nn.ReLU(),