yet more formatting

examples/human/leduc_holdem_human.py

@@ -1,5 +1,4 @@
-"""A toy example of playing against pretrianed AI on Leduc Hold'em
-"""
+"""A toy example of playing against pretrianed AI on Leduc Hold'em"""
 
 import rlcard
 from rlcard import models

examples/human/limit_holdem_human.py

@@ -1,5 +1,4 @@
-"""A toy example of playing against a random agent on Limit Hold'em
-"""
+"""A toy example of playing against a random agent on Limit Hold'em"""
 
 import rlcard
 from rlcard.agents import LimitholdemHumanAgent as HumanAgent

examples/run_dmc.py

@@ -7,8 +7,8 @@
 import rlcard
 from rlcard.agents.dmc_agent import DMCTrainer
 
-def train(args):
 
+def train(args):
     # Make the environment
     env = rlcard.make(args.env)
 
@@ -28,6 +28,7 @@ def train(args):
     # Train DMC Agents
     trainer.start()
 
+
 if __name__ == '__main__':
     parser = argparse.ArgumentParser("DMC example in RLCard")
     parser.add_argument(
@@ -94,4 +95,3 @@ def train(args):
 
     os.environ["CUDA_VISIBLE_DEVICES"] = args.cuda
     train(args)
-

examples/run_rl.py

@@ -15,11 +15,11 @@
     plot_curve,
 )
 
-def train(args):
 
+def train(args):
     # Check whether gpu is available
     device = get_device()
-        
+
     # Seed numpy, torch, random
     set_seed(args.seed)
 
@@ -40,7 +40,7 @@ def train(args):
             agent = DQNAgent(
                 num_actions=env.num_actions,
                 state_shape=env.state_shape[0],
-                mlp_layers=[64,64],
+                mlp_layers=[64, 64],
                 device=device,
                 save_path=args.log_dir,
                 save_every=args.save_every
@@ -54,8 +54,8 @@ def train(args):
             agent = NFSPAgent(
                 num_actions=env.num_actions,
                 state_shape=env.state_shape[0],
-                hidden_layers_sizes=[64,64],
-                q_mlp_layers=[64,64],
+                hidden_layers_sizes=[64, 64],
+                q_mlp_layers=[64, 64],
                 device=device,
                 save_path=args.log_dir,
                 save_every=args.save_every
@@ -105,6 +105,7 @@ def train(args):
     torch.save(agent, save_path)
     print('Model saved in', save_path)
 
+
 if __name__ == '__main__':
     parser = argparse.ArgumentParser("DQN/NFSP example in RLCard")
     parser.add_argument(
@@ -162,13 +163,13 @@ def train(args):
         type=str,
         default='experiments/leduc_holdem_dqn_result/',
     )
-    
+
     parser.add_argument(
         "--load_checkpoint_path",
         type=str,
         default="",
     )
-    
+
     parser.add_argument(
         "--save_every",
         type=int,
@@ -178,4 +179,3 @@ def train(args):
 
     os.environ["CUDA_VISIBLE_DEVICES"] = args.cuda
     train(args)
-

rlcard/agents/dmc_agent/model.py

@@ -18,21 +18,17 @@
 import torch
 from torch import nn
 
+
 class DMCNet(nn.Module):
-    def __init__(
-        self,
-        state_shape,
-        action_shape,
-            mlp_layers=None
-    ):
+    def __init__(self, state_shape, action_shape, mlp_layers=None):
         super().__init__()
         if mlp_layers is None:
             mlp_layers = [512, 512, 512, 512, 512]
         input_dim = np.prod(state_shape) + np.prod(action_shape)
         layer_dims = [input_dim] + mlp_layers
         fc = []
-        for i in range(len(layer_dims)-1):
-            fc.append(nn.Linear(layer_dims[i], layer_dims[i+1]))
+        for i in range(len(layer_dims) - 1):
+            fc.append(nn.Linear(layer_dims[i], layer_dims[i + 1]))
             fc.append(nn.ReLU())
         fc.append(nn.Linear(layer_dims[-1], 1))
         self.fc_layers = nn.Sequential(*fc)
@@ -44,19 +40,13 @@ def forward(self, obs, actions):
         values = self.fc_layers(x).flatten()
         return values
 
+
 class DMCAgent:
-    def __init__(
-        self,
-        state_shape,
-        action_shape,
-            mlp_layers=None,
-        exp_epsilon=0.01,
-        device="0",
-    ):
+    def __init__(self, state_shape, action_shape, mlp_layers=None, exp_epsilon=0.01, device="0"):
         if mlp_layers is None:
             mlp_layers = [512, 512, 512, 512, 512]
         self.use_raw = False
-        self.device = 'cuda:'+device if device != "cpu" else "cpu"
+        self.device = 'cuda:' + device if device != "cpu" else "cpu"
         self.net = DMCNet(state_shape, action_shape, mlp_layers).to(self.device)
         self.exp_epsilon = exp_epsilon
         self.action_shape = action_shape
@@ -78,8 +68,7 @@ def eval_step(self, state):
         action_idx = np.argmax(values)
         action = action_keys[action_idx]
 
-        info = {}
-        info['values'] = {state['raw_legal_actions'][i]: float(values[i]) for i in range(len(action_keys))}
+        info = {'values': {state['raw_legal_actions'][i]: float(values[i]) for i in range(len(action_keys))}}
 
         return action, info
 
@@ -125,15 +114,9 @@ def state_dict(self):
     def set_device(self, device):
         self.device = device
 
+
 class DMCModel:
-    def __init__(
-        self,
-        state_shape,
-        action_shape,
-            mlp_layers=None,
-        exp_epsilon=0.01,
-        device=0
-    ):
+    def __init__(self, state_shape, action_shape, mlp_layers=None, exp_epsilon=0.01, device=0):
         if mlp_layers is None:
             mlp_layers = [512, 512, 512, 512, 512]
         self.agents = []

rlcard/agents/dmc_agent/pettingzoo_model.py

@@ -20,13 +20,7 @@ def feed(self, ts):
 
 
 class DMCModelPettingZoo:
-    def __init__(
-        self,
-        env,
-        mlp_layers=None,
-        exp_epsilon=0.01,
-        device="0"
-    ):
+    def __init__(self, env, mlp_layers=None, exp_epsilon=0.01, device="0"):
         if mlp_layers is None:
             mlp_layers = [512, 512, 512, 512, 512]