def decide(_status):
result_list = []
score_list = [0] * 6
for i in range(6):
if _status[i] == 0:
score_list[i] = -float("inf")
result_list.append({
"status": False,
"finish": False,
"again": False,
"robbery": False
})
continue
temp_status = _status.copy()
temp_result = mancala.move(temp_status, i)
if temp_result["again"] == True:
return i
score_list[i] = temp_result["status"][6] - _status[6]
result_list.append(temp_result)
if score_list.count(max(score_list)) == 1:
return score_list.index(max(score_list))
else:
max_diff = max(score_list)
for i in range(6):
if score_list[i] == max_diff:
score_list[i] = sum([
y - x for (x, y) in zip(result_list[i]["status"][7:14],
_status[7:14])
])
else:
score_list[i] = -float("inf")
return score_list.index(max(score_list))
def decide_sub(cls, _status, _original_status):
"""
ピッタリゴールの再帰処理をするために、
直前の状態と、元の状態を受け取って
今回の判断と、最終状態を返す
"""
result_list = [None] * 6
score_list = [0] * 6
for i in range(6):
if _status[i] == 0:
score_list[i] = -float("inf")
result_list.append({
"status": False,
"finish": False,
"again": False,
"robbery": False
})
continue
temp_status = _status.copy()
temp_result = mancala.move(temp_status, i)
if temp_result["again"] == True:
if temp_result["finish"] == True:
return i, temp_result
temp_decision, temp_result = AI.decide_sub(
temp_result["status"], _original_status)
score_list[i] = sum([
x - y for (x, y) in zip(temp_result["status"][0:6],
_original_status[0:6])
]) + (temp_result["status"][6] -
_original_status[6]) * AI.GOAL_RATIO
result_list[i] = temp_result
temp_decision = score_list.index(max(score_list))
return temp_decision, result_list[temp_decision]
def decide(_status):
result_list = []
score_list = [0]*6
for i in range(6):
if _status[i] == 0:
score_list[i] = -float("inf")
result_list.append({"status":False, "finish":False, "again":False, "robbery":False})
continue
temp_status = _status.copy()
temp_result = mancala.move(temp_status, i)
if temp_result["robbery"] == True or temp_result["again"] == True:
# 横取りやピッタリゴールが発生した時のみ、ゴールの玉の数の差異をスコアにする
score_list[i] = temp_result["status"][6]-_status[6]
else:
# それ以外は0点
score_list[i] = 0
result_list.append(temp_result)
# 最大値のものからランダムに選ぶ。特にピッタリゴールや横取りが無ければタダのランダム。
return random.choice([i for i, x in enumerate(score_list) if x == max(score_list)])
def decide_sub(_status, _original_status):
"""
ピッタリゴールの再帰処理をするために、
直前の状態と、元の状態を受け取って
今回の判断と、最終状態を返す
"""
result_list = [None] * 6
score_list = [0] * 6
for i in range(6):
if _status[i] == 0:
score_list[i] = -float("inf")
result_list.append({
"status": False,
"finish": False,
"again": False,
"robbery": False
})
continue
temp_status = _status.copy()
temp_result = mancala.move(temp_status, i)
if temp_result["again"] == True:
if temp_result["finish"] == True:
return i, temp_result
temp_decision, temp_result = AI.decide_sub(
temp_result["status"], _original_status)
score_list[i] = temp_result["status"][6] - _original_status[6]
result_list[i] = temp_result
if score_list.count(max(score_list)) == 1:
temp_decision = score_list.index(max(score_list))
return temp_decision, result_list[temp_decision]
else:
max_diff = max(score_list)
for i in range(6):
if score_list[i] == max_diff:
score_list[i] = sum([
y - x for (x, y) in zip(result_list[i]["status"][7:14],
_original_status[7:14])
])
else:
score_list[i] = -float("inf")
temp_decision = score_list.index(max(score_list))
return temp_decision, result_list[temp_decision]
def test_move_3():
b = mancala.Board()
bp, again = mancala.move(b, 0, 3)
assert again == False
assert bp.pits == [[4, 4, 4, 0, 5, 5], [5, 4, 4, 4, 4, 4]]
assert bp.score == [1, 0]
def test_move_2():
b = mancala.Board()
bp, again = mancala.move(b, 0, 2)
assert again == True
assert bp.pits == [[4, 4, 0, 5, 5, 5], [4, 4, 4, 4, 4, 4]]
assert bp.score == [1, 0]
def test_move_0():
b = mancala.Board()
bp, again = mancala.move(b, 0, 0)
assert again == False
assert bp.pits == [[0, 5, 5, 5, 5, 4], [4, 4, 4, 4, 4, 4]]
assert bp.score == [0, 0]