Added human player

tic_tac_toe.py

@@ -1,6 +1,7 @@
 import numpy as np
 from helper import create_empty_state, to_state_string
 from machine_player import MachinePlayer
+from human_player import HumanPlayer
 from player_provider import PlayerProvider
 import json
 
@@ -42,17 +43,13 @@ def play(player_provider: PlayerProvider, k_max=10000, with_debug=False):
                 if not has_moved:
                     if with_debug:
                         print(f"{player.mark}: No more moves")
-                    if isinstance(player, MachinePlayer):
                         player.reward(0.0)
-                    if isinstance(other_player, MachinePlayer):
                         other_player.reward(0.0)
                     run = False
                 if player.has_won(state):
                     if with_debug:
                         print(f"{player.mark}: Has won the game")
-                    if isinstance(player, MachinePlayer):
                         player.reward(1.0)
-                    if isinstance(other_player, MachinePlayer):
                         other_player.reward(0.0)
                     run = False
 
@@ -69,23 +66,21 @@ with open("values_x.json", "r") as fp:
 with open("values_o.json", "r") as fp:
     o_values = json.load(fp)
 
-px = MachinePlayer(mark='X', params=MachinePlayer.Params(p_exp=0.2, alpha=0.1))
-po = MachinePlayer(mark='O', params=MachinePlayer.Params(p_exp=0.2, alpha=0.1))
-
-px.assign_values(x_values)
-po.assign_values(o_values)
+#px = MachinePlayer(mark='X', params=MachinePlayer.Params(p_exp=0.2, alpha=0.1), values=x_values)
+px = HumanPlayer(mark='X', name="Jens")
+po = MachinePlayer(mark='O', params=MachinePlayer.Params(p_exp=0.2, alpha=0.1), values=o_values)
 
 player_provider = PlayerProvider(px, po)
 
 do_training = 1
 if do_training:
-    play(player_provider, 10000, False)
+    play(player_provider, 10000, True)
 
 play(player_provider, 10, True)
 
 # Convert and write JSON object to file
 with open("values_x.json", "w") as fp:
-    json.dump(px.values, fp, indent=0)
+    json.dump(x_values, fp, indent=0)
 
 with open("values_o.json", "w") as fp:
-    json.dump(po.values, fp, indent=0)
+    json.dump(o_values, fp, indent=0)