Intro2AI/train_table.py at master · vovavvk/Intro2AI · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
import random
import time

import numpy as np

from entities.card import Card
from entities.hand import Hand
from entities.qtabletrainer import QTableTrainer
from gym_poker.envs.poker_env import PokerEnv


def generate_tp(val=15, val2=15, val3=15):
    """Generate two pairs 5 cards hand"""
    suits = ["Spade", "Club", "Diamond", "Heart"]
    h = Hand()
    # First Pair
    choice = random.choice(suits)
    choice2 = random.choice(suits)
    while choice == choice2:
        choice2 = random.choice(suits)
    if val == 15:
        val = random.choice(range(2, 15))
    for su in suits:
        if not (su.__eq__(choice) or su.__eq__(choice2)):
            h.add_card(Card(su, val))
    # Second Pair
    choice = random.choice(suits)
    choice2 = random.choice(suits)
    while choice == choice2:
        choice2 = random.choice(suits)
    while val2 == 15 or val2 == val:
        val2 = random.choice(range(2, 15))
    for su in suits:
        if not (su.__eq__(choice) or su.__eq__(choice2)):
            h.add_card(Card(su, val2))
    # Last Card
    while val3 == 15 or val3 == val or val3 == val2:
        val3 = random.choice(range(2, 15))
    h.add_card(Card(random.choice(suits), val3))
    return h


def generate_tok(val=15, val2=15, val3=15):
    """Generate three of a kind 5 cards hand"""
    suits = ["Spade", "Club", "Diamond", "Heart"]
    choice = random.choice(suits)
    if val == 15:
        val = random.choice(range(2, 15))
    h = Hand()
    for su in suits:
        if not su.__eq__(choice):
            h.add_card(Card(su, val))
    while val2 == 15 or val2 == val:
        val2 = random.choice(range(2, 15))
    h.add_card(Card(random.choice(suits), val2))
    while val3 == 15 or val3 == val or val3 == val2:
        val3 = random.choice(range(2, 15))
    h.add_card(Card(random.choice(suits), val3))
    return h


def generate_fh(val=15, val2=15):
    """Generate Full house 5 cards hand"""
    suits = ["Spade", "Club", "Diamond", "Heart"]
    choice = random.choice(suits)
    if val == 15:
        val = random.choice(range(2, 15))
    h = Hand()
    for su in suits:
        if not su.__eq__(choice):
            h.add_card(Card(su, val))
    choice, choice2 = random.choice(suits), random.choice(suits)
    while choice2 == choice:
        choice2 = random.choice(suits)
    while val2 == 15 or val2 == val:
        val2 = random.choice(range(2, 15))
    for su in suits:
        if not (su.__eq__(choice) or su.__eq__(choice2)):
            h.add_card(Card(su, val2))
    return h


def train_agent(crds, chips):
    trainer = QTableTrainer(chips)
    trainer.train_agent(crds)


def print_qt(num, ch):
    _c = False
    _qt = None
    try:
        # Load Qtable from file
        with np.load('Qtable/qtablenpc.npz') as data:
            _qt = data['qtable']
        _c = True
    except IOError:
        print("error loading qtable")
    if not _c:
        exit(0)
    _h = Hand()
    _res = np.zeros(8, dtype=int)
    for i in range(num):
        print("Cards  --  big blind  --  money ranges  --  small blind  -- money ranges")
        for j in range(i + 1, num):
            _h.add_card(Card.decode(i))
            _h.add_card(Card.decode(j))
            for i1 in range(2):
                for j1 in range(4):
                    _res[i1 * 4 + j1] = PokerEnv.encode(_h, i1, 1 + (j1 * (ch / 2)), ch)
            _s = _h.__str__() + " - "
            for i2 in range(8):
                _s += str.format('{:.2f}', _qt[_res[i2]][0])
                _s += "|"
                _s += str.format('{:.2f}', _qt[_res[i2]][1])
                _s += " -- "
            print(_s)
            _h.clear_hand()


if __name__ == "__main__":
    strt = time.time()
    cards = 52
    starting_chips = 20
    penalty = 25
    train_agent(cards, starting_chips)
    print_qt(cards, starting_chips)
    elapsed = int(time.time() - strt)
    print("Time elapsed: {0:d}:{1:02d}".format(elapsed // 60, elapsed % 60))