style(ai_trainer): use pep8 style guide for tainer

PythonBiellaGroup · Jul 31, 2022 · 1028285 · 1028285
1 parent 69325ad
commit 1028285
Show file tree

Hide file tree

Showing 4 changed files with 141 additions and 83 deletions.
diff --git a/ai_trainer/board.py b/ai_trainer/board.py
@@ -4,33 +4,34 @@
 HUNTER = 1
 BEAR = 2
 
+
 class Board:
     """
     Board abstraction
     """
-    adjacent = [[1,2,3], #0
-                [0,3,4],
-                [0,3,6], #2
-                [0,1,2,5],
-                [1,7,8], #4
-                [3,9,10,11],
-                [2,12,13], #6
-                [4,8,14],
-                [7,4,14,9], #8
-                [8, 10,5,15],
-                [5,9,11,15],#10
-                [5,10,15,12],
-                [11,6,16,13],#12
-                [6,12,16],
-                [7,8,18],#14
-                [9,10,11,17],
-                [12,13,19], #16
-                [15,18,19,20],
-                [14,17,20], #18
+    adjacent = [[1, 2, 3],  # 0
+                [0, 3, 4],
+                [0, 3, 6],  # 2
+                [0, 1, 2, 5],
+                [1, 7, 8],  # 4
+                [3, 9, 10, 11],
+                [2, 12, 13],  # 6
+                [4, 8, 14],
+                [7, 4, 14, 9],  # 8
+                [8, 10, 5, 15],
+                [5, 9, 11, 15],  # 10
+                [5, 10, 15, 12],
+                [11, 6, 16, 13],  # 12
+                [6, 12, 16],
+                [7, 8, 18],  # 14
+                [9, 10, 11, 17],
+                [12, 13, 19],  # 16
+                [15, 18, 19, 20],
+                [14, 17, 20],  # 18
                 [16, 17, 20],
                 [18, 17, 19]]
 
-    def __init__(self, size: int, default_char = '_'):
+    def __init__(self, size: int, default_char='_'):
         self._default_char = default_char
         self._cells = [default_char] * size
         self._last_action = None
@@ -51,7 +52,7 @@ def get_hash(self) -> str:
     def get_cells(self) -> list[str]:
         """ get cells """
         return self._cells
-    
+
     def set_cells(self, cells: list[str]) -> None:
         """ set cells """
         self._cells = cells
@@ -64,22 +65,23 @@ def get_default_char(self) -> str:
         """ get default character """
         return self._default_char
 
+
 class Game:
     """
     Game abstraction
     """
-    end_states = [  '2111_________________',
-                    '1_21__1______________',
-                    '__1___2_____11_______',
-                    '______1_____12__1____',
-                    '____________11__2__1_',
-                    '________________11_21',
-                    '_________________1112',
-                    '______________1__12_1',
-                    '_______11_____2___1__',
-                    '____1__21_____1______',
-                    '_1__2__11____________',
-                    '12_11________________']
+    end_states = ['2111_________________',
+                  '1_21__1______________',
+                  '__1___2_____11_______',
+                  '______1_____12__1____',
+                  '____________11__2__1_',
+                  '________________11_21',
+                  '_________________1112',
+                  '______________1__12_1',
+                  '_______11_____2___1__',
+                  '____1__21_____1______',
+                  '_1__2__11____________',
+                  '12_11________________']
 
     def __init__(self, board: Board, player1, player2, max_turns: int = 300):
         # pylint: disable=locally-disabled, import-outside-toplevel
@@ -93,8 +95,6 @@ def __init__(self, board: Board, player1, player2, max_turns: int = 300):
         self._max_turns: int = max_turns
         self._winner: 0 | 1 | 2 = None
 
-
-
     def has_ended(self) -> bool:
         """
         Check if the game has ended
@@ -118,7 +118,7 @@ def play(self) -> int:
         """
         Play the game
         """
-        
+
         while True:
             self._player_1.move(self._board)
             self._player_1.add_state(self._board.get_hash())
@@ -147,16 +147,28 @@ def train(self, n_times: int = 100) -> None:
             self.apply_reward()
             self.reset()
 
-        self._player_1.save_policy(n_times, self._player_2.get_state_info(n_times))
-        self._player_2.save_policy(n_times, self._player_1.get_state_info(n_times))
+        self._player_1.save_policy(
+            n_times,
+            self._player_2.get_state_info(n_times)
+        )
+        self._player_2.save_policy(
+            n_times,
+            self._player_1.get_state_info(n_times)
+        )
         print("Saved policy")
 
     def apply_reward(self) -> None:
         """
         Apply reward to the players
         """
-        bear = self._player_1 if self._player_1.get_symbol() == '2' else self._player_2
-        hunter = self._player_2 if self._player_1.get_symbol() == '2' else self._player_1
+        bear = (
+            self._player_1 if self._player_1.get_symbol() == '2' else
+            self._player_2
+        )
+        hunter = (
+            self._player_2 if self._player_1.get_symbol() == '2' else
+            self._player_1
+        )
         if self._winner == HUNTER:
             bear.feed_reward(has_won=False)
             hunter.feed_reward(has_won=True)

diff --git a/ai_trainer/controller.py b/ai_trainer/controller.py
@@ -2,17 +2,21 @@
 This module handles all the logic to start the game
 """
 import argparse
+import random
 from board import Board, Game
 from player import AIPlayer
-import random 
+
 
 DEFAULT_NO_PLAYER = 'random'
 DEFAULT_HUNTER_POSITION = [0, 1, 2]
 DEFAULT_BEAR_POSITION = [20]
+
+
 def inizialize_board() -> Board:
     """
     Initialize the board
     """
+
     board = Board(21)
     board[0] = '1'
     board[1] = '1'
@@ -30,6 +34,7 @@ def _parse_arguments():
     parser.add_argument('--seed', type=int, default=None)
     return parser.parse_args()
 
+
 if __name__ == '__main__':
     args = _parse_arguments()
     hunter_player_file = args.hunter_player
@@ -39,9 +44,18 @@ def _parse_arguments():
     if args.seed is not None:
         random.seed(args.seed)
 
-    hunter_player = AIPlayer(DEFAULT_HUNTER_POSITION, 'hunter', '1', loss_reward=-5, win_reward=100)
-    bear_player = AIPlayer(DEFAULT_BEAR_POSITION, 'bear', '2', loss_reward=-5, win_reward=1)
-
+    hunter_player = AIPlayer(
+        DEFAULT_HUNTER_POSITION,
+        'hunter',
+        '1',
+        loss_reward=-5, win_reward=100
+    )
+    bear_player = AIPlayer(
+        DEFAULT_BEAR_POSITION,
+        'bear',
+        '2',
+        loss_reward=-5, win_reward=1
+    )
 
     if hunter_player_file != DEFAULT_NO_PLAYER:
         hunter_player.load_policy(hunter_player_file)

diff --git a/ai_trainer/dump_pickle.py b/ai_trainer/dump_pickle.py
@@ -1,28 +1,41 @@
-import argparse 
+import argparse
 import pickle
 
-parser = argparse.ArgumentParser() 
+
+parser = argparse.ArgumentParser()
 parser.add_argument('-i', '--input', help='input file', required=True)
-parser.add_argument('-d', '--dump', type=bool, help='dump the state', default=False)
-parser.add_argument('--reverse', type=bool, help='reverse the input file', default=False)
-parser.add_argument('--state', help='display valueof the state', default = '')
+parser.add_argument(
+    '-d', '--dump',
+    type=bool,
+    help='dump the state', default=False
+)
+parser.add_argument(
+    '--reverse',
+    type=bool,
+    help='reverse the input file',
+    default=False
+)
+parser.add_argument('--state', help='display valueof the state', default='')
 
 args = parser.parse_args()
-input = args.input 
-state = args.state 
-with open(input, 'rb') as f: 
+input = args.input
+state = args.state
+with open(input, 'rb') as f:
     data: dict[str, float] = pickle.load(f)
-    # sort my data dict by value 
-
 
     states = data['states_value'].copy()
-    data['states_value'] = None 
+    data['states_value'] = None
 
-    print(data) 
+    print(data)
     if state == '' and args.dump:
-        states = {k: v for k, v in sorted(states.items(), key=lambda item: item[1], reverse = args.reverse)}
-        for key, value in states.items(): 
+        states = {
+            k: v for k, v in sorted(
+                states.items(), key=lambda item: item[1],
+                reverse=args.reverse
+            )
+        }
+        for key, value in states.items():
             print(key, value)
-    else: 
+    else:
         print("the value of the state is:")
-        print(f"{state}: {data[state]}")
+        print(f"{state}: {data[state]}")