|
1
|
|
|
import numpy as np |
|
2
|
|
|
import pandas as pd |
|
3
|
|
|
|
|
4
|
|
|
from hyperactive import Hyperactive |
|
5
|
|
|
|
|
6
|
|
|
|
|
7
|
|
|
def test_issue_25(): |
|
8
|
|
|
# set a path to save the dataframe |
|
9
|
|
|
path = "./search_data.csv" |
|
10
|
|
|
search_space = { |
|
11
|
|
|
"n_neighbors": list(range(1, 50)), |
|
12
|
|
|
} |
|
13
|
|
|
|
|
14
|
|
|
# get para names from search space + the score |
|
15
|
|
|
para_names = list(search_space.keys()) + ["score"] |
|
16
|
|
|
|
|
17
|
|
|
# init empty pandas dataframe |
|
18
|
|
|
search_data = pd.DataFrame(columns=para_names) |
|
19
|
|
|
search_data.to_csv(path, index=False) |
|
20
|
|
|
|
|
21
|
|
|
def objective_function(para): |
|
22
|
|
|
# score = random.choice([1.2, 2.3, np.nan]) |
|
23
|
|
|
score = np.nan |
|
24
|
|
|
|
|
25
|
|
|
# you can access the entire dictionary from "para" |
|
26
|
|
|
parameter_dict = para.para_dict |
|
27
|
|
|
|
|
28
|
|
|
# save the score in the copy of the dictionary |
|
29
|
|
|
parameter_dict["score"] = score |
|
30
|
|
|
|
|
31
|
|
|
# append parameter dictionary to pandas dataframe |
|
32
|
|
|
search_data = pd.read_csv(path, na_values="nan") |
|
33
|
|
|
search_data_new = pd.DataFrame(parameter_dict, columns=para_names, index=[0]) |
|
34
|
|
|
search_data = search_data.append(search_data_new) |
|
35
|
|
|
search_data.to_csv(path, index=False, na_rep="nan") |
|
36
|
|
|
|
|
37
|
|
|
return score |
|
38
|
|
|
|
|
39
|
|
|
hyper0 = Hyperactive() |
|
40
|
|
|
hyper0.add_search(objective_function, search_space, n_iter=50) |
|
41
|
|
|
hyper0.run() |
|
42
|
|
|
|
|
43
|
|
|
search_data_0 = pd.read_csv(path, na_values="nan") |
|
44
|
|
|
""" |
|
45
|
|
|
the second run should be much faster than before, |
|
46
|
|
|
because Hyperactive already knows most parameters/scores |
|
47
|
|
|
""" |
|
48
|
|
|
hyper1 = Hyperactive() |
|
49
|
|
|
hyper1.add_search( |
|
50
|
|
|
objective_function, search_space, n_iter=50, memory_warm_start=search_data_0 |
|
51
|
|
|
) |
|
52
|
|
|
hyper1.run() |
|
53
|
|
|
|
SimonBlanke /
Hyperactive
