1
|
|
|
# Author: Simon Blanke |
2
|
|
|
# Email: [email protected] |
3
|
|
|
# License: MIT License |
4
|
|
|
|
5
|
|
|
import numpy as np |
6
|
|
|
|
7
|
|
|
from sklearn.datasets import load_iris |
8
|
|
|
from sklearn.model_selection import cross_val_score |
9
|
|
|
from sklearn.tree import DecisionTreeClassifier |
10
|
|
|
|
11
|
|
|
from hyperactive import Optimizer |
12
|
|
|
|
13
|
|
|
data = load_iris() |
14
|
|
|
X, y = data.data, data.target |
15
|
|
|
|
16
|
|
|
|
17
|
|
|
def objective_function(para): |
18
|
|
|
dtc = DecisionTreeClassifier( |
19
|
|
|
max_depth=para["max_depth"], min_samples_split=para["min_samples_split"], |
20
|
|
|
) |
21
|
|
|
scores = cross_val_score(dtc, para["features"], para["target"], cv=2) |
22
|
|
|
|
23
|
|
|
return scores.mean() |
24
|
|
|
|
25
|
|
|
|
26
|
|
|
search_space = { |
27
|
|
|
"max_depth": range(1, 21), |
28
|
|
|
"min_samples_split": range(2, 21), |
29
|
|
|
} |
30
|
|
|
|
31
|
|
|
|
32
|
|
|
def _base_test(search, opt_args={}, time=None): |
33
|
|
|
opt = Optimizer(**opt_args) |
34
|
|
|
opt.add_search(**search) |
35
|
|
|
opt.run(time) |
36
|
|
|
|
37
|
|
|
|
38
|
|
|
def test_max_time(): |
39
|
|
|
search = { |
40
|
|
|
"objective_function": objective_function, |
41
|
|
|
"function_parameter": {"features": X, "target": y}, |
42
|
|
|
"search_space": search_space, |
43
|
|
|
} |
44
|
|
|
_base_test(search, time=0.01) |
45
|
|
|
|
46
|
|
|
|
47
|
|
|
def test_init_para(): |
48
|
|
|
search = { |
49
|
|
|
"objective_function": objective_function, |
50
|
|
|
"function_parameter": {"features": X, "target": y}, |
51
|
|
|
"search_space": search_space, |
52
|
|
|
} |
53
|
|
|
|
54
|
|
|
init_para1 = { |
55
|
|
|
"max_depth": 3, |
56
|
|
|
"min_samples_split": 3, |
57
|
|
|
} |
58
|
|
|
init_para_list = [[init_para1]] |
59
|
|
|
for init_para in init_para_list: |
60
|
|
|
search["init_para"] = init_para |
61
|
|
|
_base_test(search) |
62
|
|
|
|
63
|
|
|
|
64
|
|
|
def test_verbosity(): |
65
|
|
|
search = { |
66
|
|
|
"objective_function": objective_function, |
67
|
|
|
"function_parameter": {"features": X, "target": y}, |
68
|
|
|
"search_space": search_space, |
69
|
|
|
} |
70
|
|
|
|
71
|
|
|
verbosity_list = [0, 1, 2, 3] |
72
|
|
|
for verbosity in verbosity_list: |
73
|
|
|
_base_test(search, opt_args={"verbosity": verbosity}) |
74
|
|
|
|
75
|
|
|
|
76
|
|
|
def test_n_jobs(): |
77
|
|
|
search = { |
78
|
|
|
"objective_function": objective_function, |
79
|
|
|
"function_parameter": {"features": X, "target": y}, |
80
|
|
|
"search_space": search_space, |
81
|
|
|
} |
82
|
|
|
|
83
|
|
|
n_jobs_list = [1, 2, 4, 10, -1] |
84
|
|
|
for n_jobs in n_jobs_list: |
85
|
|
|
search["n_jobs"] = n_jobs |
86
|
|
|
_base_test(search) |
87
|
|
|
|
88
|
|
|
|
89
|
|
|
def test_positional_args(): |
90
|
|
|
search = { |
91
|
|
|
"objective_function": objective_function, |
92
|
|
|
"function_parameter": {"features": X, "target": y}, |
93
|
|
|
"search_space": search_space, |
94
|
|
|
} |
95
|
|
|
_base_test(search) |
96
|
|
|
|
97
|
|
|
|
98
|
|
|
def test_n_iter(): |
99
|
|
|
search = { |
100
|
|
|
"objective_function": objective_function, |
101
|
|
|
"function_parameter": {"features": X, "target": y}, |
102
|
|
|
"search_space": search_space, |
103
|
|
|
} |
104
|
|
|
|
105
|
|
|
n_iter_list = [0, 1, 2, 4, 10, 100] |
106
|
|
|
for n_iter in n_iter_list: |
107
|
|
|
search["n_iter"] = n_iter |
108
|
|
|
_base_test(search) |
109
|
|
|
|
110
|
|
|
|
111
|
|
|
def test_memory(): |
112
|
|
|
search = { |
113
|
|
|
"objective_function": objective_function, |
114
|
|
|
"function_parameter": {"features": X, "target": y}, |
115
|
|
|
"search_space": search_space, |
116
|
|
|
} |
117
|
|
|
|
118
|
|
|
memory_list = [False, "short"] |
119
|
|
|
for memory in memory_list: |
120
|
|
|
search["memory"] = memory |
121
|
|
|
_base_test(search) |
122
|
|
|
|
123
|
|
|
|