alpha-thesis/interpret.py at master · davidschulte/alpha-thesis · 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
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
import os
from pickle import Pickler

def get_counts_player(file, games_played):
    places = np.zeros((3, 3))
    dataframe = pd.read_pickle(file)
    array = dataframe.to_numpy()
    for row in range(games_played):
        for column in range(3):
            if array[row, column+3] == 3:
                places[column, 0] += 1
            elif array[row, column+3] == 1:
                places[column, 1] += 1
            else:
                places[column, 2] += 1
    print(places)
    sum_places = sum(places)
    normalized = [x / sum_places for x in places]
    print(normalized)

    vector = np.transpose(normalized)
    return vector

def get_counts_all(file, games_played):
    places = [0] * 3
    dataframe = pd.read_pickle(file)
    print(dataframe.to_string())
    array = dataframe.to_numpy()
    print(array)
    for row in range(games_played):
        for column in range(3):
            if array[row, column] == 1:
                if array[row, column + 3] == 3:
                    places[0] += 1
                elif array[row, column + 3] == 1:
                    places[1] += 1
                else:
                    places[2] += 1
    print(places)
    sum_places = sum(places)
    normalized = [x / sum_places for x in places]
    print(normalized)

    vector = np.transpose(normalized)
    return vector

def get_counts_separate(file, alone, games_played):
    places = [0] * 3
    dataframe = pd.read_pickle(file)
    print(dataframe.to_string())
    array = dataframe.to_numpy()
    if alone:
        sum_players = 5
    else:
        sum_players = 4

    for row in range(games_played):
        if sum(array[row, 0:3]) == sum_players:
            if alone:
                for column in range(3):
                    if array[row, column] == 1:
                        if array[row, column + 3] == 3:
                            places[0] += 1
                        elif array[row, column + 3] == 1:
                            places[1] += 1
                        else:
                            places[2] += 1
            else:
                sum_positions = 0
                for column in range(3):
                    if array[row, column] == 1:
                        sum_positions += array[row, column + 3]

                print(sum_positions)
                if sum_positions == 4:
                    places[0] += 1
                elif sum_positions == 3:
                    places[1] += 1
                else:
                    places[2] += 1

    print(places)
    sum_places = sum(places)
    normalized = [x / sum_places for x in places]
    print(normalized)

    vector = np.transpose(normalized)
    return vector

#
versions = [1, 2, 3, 6, 8, 10, 11, 12, 13, 15, 17, 20, 22, 23, 27, 32, 34, 35, 37, 41]
#
# results_mcts_alone = np.zeros((3, len(versions)))
# results_mcts_team = np.zeros((3, len(versions)))
# results_nnet_alone = np.zeros((3, len(versions)))
# results_nnet_team = np.zeros((3, len(versions)))
#
# folder_mcts = "mcts vs nnet"
# folder_nnet = "nnet vs nnet"
# games_played_mcts = 30
# games_played_nnet = 120
#
# for v in range(len(versions)):
#     filename = os.path.join(folder_nnet, str(versions[v]) + ".pkl")
#     # results_mcts_alone[:, v] = get_counts_separate(filename, True, games_played_nnet)
#     # results_mcts_team[:, v] = get_counts_separate(filename, False, games_played_nnet)
#     #
#     # filename = os.path.join(folder_nnet, str(versions[v]) + ".pkl")
#     # results_nnet_alone[:, v] = get_counts_separate(filename, True, games_played_nnet)
#     # results_nnet_team[:, v] = get_counts_separate(filename, False, games_played_nnet)


x = np.array([x for x in range(1, len(versions)+1)])


fig = plt.figure()
fig.suptitle("Main Agent vs. Neural Net", fontsize=16)

plt.subplots_adjust(hspace=0.4)

ax1 = fig.add_subplot(211)
plt.xticks(np.arange(min(x), max(x)+1, 1.0))
plt.yticks(np.arange(0, 1.1, 0.1))
plt.ylim((0, 1))
ax1.set_title("1 vs. 2")
plt.xlabel("Version")
plt.ylabel("Relative frequency of score")
plt.grid(True)
ax2 = fig.add_subplot(212)
plt.xticks(np.arange(min(x), max(x)+1, 1.0))
plt.yticks(np.arange(0, 1.1, 0.1))
plt.ylim((0, 1))
ax2.set_title("2 vs. 1")
plt.xlabel("Version")
plt.ylabel("Relative frequency of score")
plt.grid(True)


for p in range(3):
    ax1.plot(x, results_mcts_alone[p,:], marker="o")
    ax2.plot(x, results_mcts_team[p,:], marker="o")


# plt.title("Neural Net without MCTS agains previous version")
ax1.legend(["3", "1", "0"])
ax2.legend(["4", "3", "1"])

plt.show()