NanoChef/virtual_space_image.py at main · KIST-CSRC/NanoChef · 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
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
import numpy as np
import pandas as pd
import scipy.stats
import matplotlib.pyplot as plt
from itertools import combinations

from olympus.noises.noise_gaussian_noise import GaussianNoise
from olympus.surfaces import SurfaceLoader

surface_loader=SurfaceLoader()

# noise=GaussianNoise(scale=0.5)
# surface=Dejong(param_dim=2, noise=noise)
# print(surface.run([[0.5, 0.5], [1, 1]]))
# print(surface.run([1, 1]))
# print(surface.kind)

# surfaces = [
#     "Dejong", "HyperEllipsoid", "Zakharov",
#     "AckleyPath", "Branin", "Levy", "Michalewicz", "Rastrigin", "Rosenbrock", "Schwefel", "StyblinskiTang",
#     "LinearFunnel", "NarrowFunnel",
#     "GaussianMixture", "Denali", "Everest", "K2", "Kilimanjaro", "Matterhorn", "MontBlanc"
#     ]

surfaces = [
    "Dejong", "HyperEllipsoid", "AckleyPath", "Rastrigin", "LinearFunnel", "Levy",
    "Branin", "Michalewicz", "Rosenbrock", "Schwefel", "StyblinskiTang", "GaussianMixture",
    "Denali", "Everest", "K2", "Kilimanjaro", "Matterhorn", "MontBlanc"
    ]

# for surface_index, surface_name in enumerate(surfaces):

#     surface = surface_loader.import_surface(surface_name)()

#     domain = np.linspace(0, 1, 100)
#     X, Y = np.meshgrid(domain, domain)
#     Z = np.zeros((len(domain), len(domain)))
#     for x_index, x in enumerate(domain):
#         for y_index, y in enumerate(domain):
#             value = surface.run(params=[x, y])
#             Z[x_index, y_index] = value[0][0]
#     max_z=Z.max()
#     min_z=Z.min()
#     # syntax for 3-D projection
#     fig = plt.figure()
#     ax = plt.axes(projection ='3d')

#     levels = np.linspace(min_z, max_z, 200)
#     pc = ax.contourf(X, Y, Z, cmap = "coolwarm", levels=levels)
#     cbar = plt.colorbar(pc)

#     plt.tight_layout()
#     plt.savefig('surface_{}.png'.format(surface_name),dpi=300)
#     plt.close()

selected_surfaces_list=list(combinations(surfaces, 2))
print(selected_surfaces_list)
# print(len(surfaces))
# print(selected_surfaces_list)
# print(len(selected_surfaces_list))

def calculate_spearman_rank_correlation(data1, data2):
    # 데이터를 순위로 변환
    data1=data1.ravel()
    data2=data2.ravel()
    rank_data1 = np.argsort(np.argsort(data1))
    rank_data2 = np.argsort(np.argsort(data2))

    # 중복된 원소 찾기
    unique_elements_1, counts_1 = np.unique(data1, return_counts=True)
    unique_elements_2, counts_2 = np.unique(data2, return_counts=True)

    # 중복된 원소의 평균 순위 계산
    average_ranks_1 = {element: np.mean(rank_data1[data1 == element]) for element in unique_elements_1}
    average_ranks_2 = {element: np.mean(rank_data2[data2 == element]) for element in unique_elements_2}

    # 평균 순위를 사용하여 covert_rank_arr 생성
    covert_rank_arr_1 = np.array([average_ranks_1[element] for element in data1])
    covert_rank_arr_2 = np.array([average_ranks_2[element] for element in data2])

    # 스피어맨 순위 상관 계수 계산
    spearman_corr, _ = scipy.stats.spearmanr(covert_rank_arr_1, covert_rank_arr_2)

    return spearman_corr

combination_name_list=[]
retrun_spearman_corr_list=[]

Z_0_max_x_point_list=[]
Z_0_max_y_point_list=[]
Z_0_min_x_point_list=[]
Z_0_min_y_point_list=[]
Z_1_max_x_point_list=[]
Z_1_max_y_point_list=[]
Z_1_min_x_point_list=[]
Z_1_min_y_point_list=[]

max_z_0_list=[]
min_z_0_list=[]
max_z_1_list=[]
min_z_1_list=[]

for combination in selected_surfaces_list:

    surface_0 = surface_loader.import_surface(combination[0])()
    surface_1 = surface_loader.import_surface(combination[1])()

    domain = np.linspace(0, 1, 100)
    X, Y = np.meshgrid(domain, domain)
    Z_0 = np.zeros((len(domain), len(domain)))
    Z_1 = np.zeros((len(domain), len(domain)))
    for x_index, x in enumerate(domain):
        for y_index, y in enumerate(domain):
            value_0 = surface_0.run(params=[x, y])
            value_1 = surface_1.run(params=[x, y])
            Z_0[x_index, y_index] = value_0[0][0]
            Z_1[x_index, y_index] = value_1[0][0]

    combination_name="{}/{}".format(surface_0.kind, surface_1.kind)
    retrun_spearman_corr = calculate_spearman_rank_correlation(Z_0, Z_1)

    max_z_0=Z_0.max()
    min_z_0=Z_0.min()
    max_z_1=Z_1.max()
    min_z_1=Z_1.min()

    # max나 min이 중복되면 여러개 뱉어냄
    Z_0_max_row_index, Z_0_max_col_index = np.where(Z_0 == max_z_0)
    Z_0_min_row_index, Z_0_min_col_index = np.where(Z_0 == min_z_0)
    Z_1_max_row_index, Z_1_max_col_index = np.where(Z_1 == max_z_1)
    Z_1_min_row_index, Z_1_min_col_index = np.where(Z_1 == min_z_1)

    Z_0_max_point=[X[Z_0_max_row_index, Z_0_max_col_index], Y[Z_0_max_row_index, Z_0_max_col_index]]
    Z_0_min_point=[X[Z_0_min_row_index, Z_0_min_col_index], Y[Z_0_min_row_index, Z_0_min_col_index]]
    Z_1_max_point=[X[Z_1_max_row_index, Z_1_max_col_index], Y[Z_1_max_row_index, Z_1_max_col_index]]
    Z_1_min_point=[X[Z_1_min_row_index, Z_1_min_col_index], Y[Z_1_min_row_index, Z_1_min_col_index]]

    combination_name_list.append(combination_name)
    retrun_spearman_corr_list.append(retrun_spearman_corr)
    Z_0_max_x_point_list.append(X[Z_0_max_row_index, Z_0_max_col_index])
    Z_0_max_y_point_list.append(Y[Z_0_max_row_index, Z_0_max_col_index])
    Z_0_min_x_point_list.append(X[Z_0_min_row_index, Z_0_min_col_index])
    Z_0_min_y_point_list.append(Y[Z_0_min_row_index, Z_0_min_col_index])
    Z_1_max_x_point_list.append(X[Z_1_max_row_index, Z_1_max_col_index])
    Z_1_max_y_point_list.append(Y[Z_1_max_row_index, Z_1_max_col_index])
    Z_1_min_x_point_list.append(X[Z_1_min_row_index, Z_1_min_col_index])
    Z_1_min_y_point_list.append(Y[Z_1_min_row_index, Z_1_min_col_index])
    max_z_0_list.append(max_z_0)
    min_z_0_list.append(min_z_0)
    max_z_1_list.append(max_z_1)
    min_z_1_list.append(min_z_1)

# 예제 DataFrame 생성
data = {
    'Combination': combination_name_list,
    'spearman_corr': retrun_spearman_corr_list,
    "Z_0_max_x_point": Z_0_max_x_point_list,
    "Z_0_max_y_point": Z_0_max_y_point_list,
    "Z_0_min_x_point": Z_0_min_x_point_list,
    "Z_0_min_y_point": Z_0_min_y_point_list,
    "Z_1_max_x_point": Z_1_max_x_point_list,
    "Z_1_max_y_point": Z_1_max_y_point_list,
    "Z_1_min_x_point": Z_1_min_x_point_list,
    "Z_1_min_y_point": Z_1_min_y_point_list,
    "max_z_0": max_z_0_list,
    "min_z_0": min_z_0_list,
    "max_z_1": max_z_1_list,
    "min_z_1": min_z_1_list
}

df = pd.DataFrame(data)

# Excel 파일로 저장
df.to_csv('example_mountain.csv', index=False)