Perhaps better naming

Rename networks.

main.py

@@ -4,7 +4,7 @@
 from q_learning.utils import read_config
 from q_learning.preprocess import greyscale
 from q_learning.environment import PreproWrapper, MaxPoolSkipEnv, EnvTest
-from q_learning.network import DQNLinear, DQNDeepMind, LinearExploration, LinearSchedule
+from q_learning.network import Linear, DQN, LinearExploration, LinearSchedule
 
 """
 This script lets us run deep Q network or linear approximation according to a custom config file.
@@ -26,7 +26,7 @@ def run():
     parser.add_argument(
         "--config_filename",
         help="The name of the config file in the config/ directory to be used for model training.",
-        default="test_dqn_linear.yml",
+        default="test_linear.yml",
     )
 
     args = parser.parse_args()
@@ -54,7 +54,7 @@ def run():
             )
 
             # train model
-            model = DQNDeepMind(env, config)
+            model = DQN(env, config)
             model.run(exp_schedule, lr_schedule)
 
         elif config["model"] == "linear":
@@ -76,7 +76,7 @@ def run():
             )
 
             # train model
-            model = DQNLinear(env, config)
+            model = Linear(env, config)
             model.run(exp_schedule, lr_schedule)
 
         else:
@@ -115,11 +115,11 @@ def run():
             )
 
             if config["model"] == "dqn":
-                model = DQNDeepMind(env, config)
+                model = DQN(env, config)
                 model.run(exp_schedule, lr_schedule)
 
             elif config["model"] == "linear":
-                model = DQNLinear(env, config)
+                model = Linear(env, config)
                 model.run(exp_schedule, lr_schedule)
 
             else:

q_learning/network/__init__.py

@@ -1,3 +1,3 @@
-from .dqn_linear_approximation import DQNLinear
-from .dqn_deepmind import DQNDeepMind
+from .linear import Linear
+from .dqn import DQN
 from .schedule import LinearExploration, LinearSchedule

q_learning/network/dqn_deepmind.py → q_learning/network/dqn.py

@@ -1,9 +1,9 @@
 import torch
 import torch.nn as nn
-from .dqn_abstract import DQN
+from .dqn_abstract import AbstractDQN
 
 
-class DQNDeepMind(DQN):
+class DQN(AbstractDQN):
     """
     Implementation of DeepMind's Nature paper:
     https://storage.googleapis.com/deepmind-data/assets/papers/DeepMindNature14236Paper.pdf

q_learning/network/dqn_abstract.py

@@ -11,7 +11,7 @@
 from q_learning.network.qn import QN
 
 
-class DQN(QN, ABC):
+class AbstractDQN(QN, ABC):
     """ Abstract class for Deep Q Network """
     def __init__(self, env, config, logger=None):
         self.q_network = None

q_learning/network/dqn_linear_approximation.py → q_learning/network/linear.py

@@ -1,10 +1,10 @@
 import torch
 import torch.nn as nn
 
-from .dqn_abstract import DQN
+from .dqn_abstract import AbstractDQN
 
 
-class DQNLinear(DQN):
+class Linear(AbstractDQN):
     """
     We represent Q function as linear approximation Q_\theta(s,a) = \thetaT*\delta(s,a)
        where [\delta(s,a)]_{s‘,a‘} = 1 iff s‘ = s, a‘ = a.

test/test_q_learning.py

@@ -3,7 +3,7 @@
 import torch.nn as nn
 
 from q_learning.environment import EnvTest
-from q_learning.network import DQNDeepMind, DQNLinear, LinearSchedule, LinearExploration
+from q_learning.network import DQN, Linear, LinearSchedule, LinearExploration
 from utils import read_config
 
 
@@ -34,11 +34,11 @@ def test_eps_out_of_range(self):
 class TestLinearDQN(unittest.TestCase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self.linear_config = read_config('test_dqn_linear.yml')
+        self.linear_config = read_config('test_linear.yml')
 
     def test_config(self):
         env = EnvTest((5, 5, 1))
-        model = DQNLinear(env, self.linear_config)
+        model = Linear(env, self.linear_config)
         state_shape = list(env.observation_space.shape)
         img_height, img_width, n_channels = state_shape
         num_actions = env.action_space.n
@@ -71,7 +71,7 @@ def test_config(self):
 
     def test_loss(self):
         env = EnvTest((5, 5, 1))
-        model = DQNLinear(env, self.linear_config)
+        model = Linear(env, self.linear_config)
         state_shape = list(env.observation_space.shape)
         img_height, img_width, n_channels = state_shape
         num_actions = env.action_space.n
@@ -105,7 +105,7 @@ def test_loss(self):
 
     def test_optimizer(self):
         env = EnvTest((5, 5, 1))
-        model = DQNLinear(env, self.linear_config)
+        model = Linear(env, self.linear_config)
         state_shape = list(env.observation_space.shape)
         img_height, img_width, n_channels = state_shape
         num_actions = env.action_space.n
@@ -139,18 +139,18 @@ def test_run(self):
         )
 
         # train model
-        model = DQNLinear(env, self.linear_config)
+        model = Linear(env, self.linear_config)
         model.run(exp_schedule, lr_schedule)
 
 
 class TestDeepMindDQN(unittest.TestCase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self.dqn_deepmind_config = read_config('test_dqn_deepmind.yml')
+        self.dqn_deepmind_config = read_config('test_dqn.yml')
 
     def test_input_output_shapes(self):
         env = EnvTest((80, 80, 1))
-        model = DQNDeepMind(env, self.dqn_deepmind_config)
+        model = DQN(env, self.dqn_deepmind_config)
 
         state_shape = list(env.observation_space.shape)
         img_height, img_width, n_channels = state_shape