64 lines
1.4 KiB
Python
64 lines
1.4 KiB
Python
import numpy as np
|
|
|
|
|
|
class APlayer(object):
|
|
def __init__(self, mark):
|
|
self.mark = mark
|
|
self.other_mark = 'O'
|
|
if 'O' in mark:
|
|
self.other_mark = 'X'
|
|
|
|
self.state = None
|
|
self.state_last = None
|
|
|
|
def set_debug(self, with_debug):
|
|
pass
|
|
|
|
def move(self, state: np.array):
|
|
return state, False
|
|
|
|
def reward(self, value):
|
|
pass
|
|
|
|
def new_game(self):
|
|
pass
|
|
|
|
@staticmethod
|
|
def f_state_slices():
|
|
nr = 3
|
|
nc = 3
|
|
result = []
|
|
# Create row finishing states
|
|
d1 = ()
|
|
d2_r = ()
|
|
for r in range(0, nr):
|
|
d1 += (r,)
|
|
for c in range(0, nc):
|
|
d2 = (c,) * nc
|
|
result.append((d1, d2))
|
|
|
|
# Create column finishing states
|
|
for c in range(0, nc):
|
|
d2 = (c,) * nc
|
|
result.append((d2, d1))
|
|
for c in range(0, nc):
|
|
d2_r += (nc - c - 1,)
|
|
|
|
# Create diagonal finishing states #1
|
|
result.append((d1, d1))
|
|
|
|
# Create diagonal finishing states #2
|
|
result.append((d1, d2_r))
|
|
|
|
return result
|
|
|
|
def has_won(self, state: np.array):
|
|
result = False
|
|
ref = [self.mark, self.mark, self.mark]
|
|
for f_state_slice in APlayer.f_state_slices():
|
|
if np.all(state[f_state_slice] == ref):
|
|
result = True
|
|
break
|
|
|
|
return result
|