|
1
|
|
|
# Author: Simon Blanke |
|
2
|
|
|
# Email: [email protected] |
|
3
|
|
|
# License: MIT License |
|
4
|
|
|
|
|
5
|
|
|
|
|
6
|
|
|
import numpy as np |
|
7
|
|
|
import pandas as pd |
|
8
|
|
|
|
|
9
|
|
|
from multiprocessing import Pool |
|
10
|
|
|
from importlib import import_module |
|
11
|
|
|
|
|
12
|
|
|
|
|
13
|
|
|
class Search: |
|
14
|
|
|
def __init__(self, function_parameter, search_processes, verb): |
|
15
|
|
|
self.function_parameter = function_parameter |
|
16
|
|
|
self.search_processes = search_processes |
|
17
|
|
|
self.verb = verb |
|
18
|
|
|
|
|
19
|
|
|
self.n_processes = len(search_processes) |
|
20
|
|
|
self._n_process_range = range(0, self.n_processes) |
|
21
|
|
|
|
|
22
|
|
|
self.results = {} |
|
23
|
|
|
self.eval_times = {} |
|
24
|
|
|
self.iter_times = {} |
|
25
|
|
|
self.best_scores = {} |
|
26
|
|
|
self.pos_list = {} |
|
27
|
|
|
self.score_list = {} |
|
28
|
|
|
self.position_results = {} |
|
29
|
|
|
|
|
30
|
|
|
def _get_results(self, results_list): |
|
31
|
|
|
position_results_dict = {} |
|
32
|
|
|
|
|
33
|
|
|
self.eval_times_dict = {} |
|
34
|
|
|
self.iter_times_dict = {} |
|
35
|
|
|
self.para_best_dict = {} |
|
36
|
|
|
self.score_best_dict = {} |
|
37
|
|
|
self.memory_dict_new = {} |
|
38
|
|
|
|
|
39
|
|
|
for results in results_list: |
|
40
|
|
|
search_name = results.search_name |
|
41
|
|
|
|
|
42
|
|
|
self.eval_times_dict[search_name] = results.eval_times |
|
43
|
|
|
self.iter_times_dict[search_name] = results.iter_times |
|
44
|
|
|
self.para_best_dict[search_name] = results.para_best |
|
45
|
|
|
self.score_best_dict[search_name] = results.score_best |
|
46
|
|
|
self.memory_dict_new[search_name] = results.memory_dict_new |
|
47
|
|
|
self.position_results[search_name] = self._memory_dict2dataframe( |
|
48
|
|
|
results.memory_dict_new, results.search_space |
|
49
|
|
|
) |
|
50
|
|
|
|
|
51
|
|
|
print( |
|
52
|
|
|
"Process", |
|
53
|
|
|
results.nth_process, |
|
54
|
|
|
"->", |
|
55
|
|
|
results.model.__name__, |
|
56
|
|
|
"search results:", |
|
57
|
|
|
) |
|
58
|
|
|
print("best parameter =", results.para_best) |
|
59
|
|
|
print("best score =", results.score_best, "\n") |
|
60
|
|
|
|
|
61
|
|
|
def _run_job(self, nth_process): |
|
62
|
|
|
self.process = self.search_processes[nth_process] |
|
63
|
|
|
return self.process.search(self.start_time, self.max_time, nth_process) |
|
64
|
|
|
|
|
65
|
|
|
def _run_multiple_jobs(self): |
|
66
|
|
|
"""Wrapper for the parallel search. Passes integer that corresponds to process number""" |
|
67
|
|
|
pool = Pool(self.n_processes) |
|
68
|
|
|
results_list = pool.map(self._run_job, self._n_process_range) |
|
69
|
|
|
|
|
70
|
|
|
for _ in range(int(self.n_processes / 2) + 2): |
|
71
|
|
|
print("\n") # make room in cmd for prints |
|
72
|
|
|
|
|
73
|
|
|
return results_list |
|
74
|
|
|
|
|
75
|
|
|
def _memory_dict2dataframe(self, memory_dict, search_space): |
|
76
|
|
|
columns = list(search_space.keys()) |
|
77
|
|
|
|
|
78
|
|
|
if not bool(memory_dict): |
|
79
|
|
|
return pd.DataFrame([], columns=columns) |
|
80
|
|
|
|
|
81
|
|
|
pos_tuple_list = list(memory_dict.keys()) |
|
82
|
|
|
result_list = list(memory_dict.values()) |
|
83
|
|
|
|
|
84
|
|
|
results_df = pd.DataFrame(result_list) |
|
85
|
|
|
np_pos = np.array(pos_tuple_list) |
|
86
|
|
|
|
|
87
|
|
|
pd_pos = pd.DataFrame(np_pos, columns=columns) |
|
88
|
|
|
dataframe = pd.concat([pd_pos, results_df], axis=1) |
|
89
|
|
|
|
|
90
|
|
|
return dataframe |
|
91
|
|
|
|
|
92
|
|
|
def _run(self, start_time, max_time): |
|
93
|
|
|
self.start_time = start_time |
|
94
|
|
|
self.max_time = max_time |
|
95
|
|
|
|
|
96
|
|
|
if len(self.search_processes) == 1: |
|
97
|
|
|
results_list = [self._run_job(0)] |
|
98
|
|
|
else: |
|
99
|
|
|
results_list = self._run_multiple_jobs() |
|
100
|
|
|
|
|
101
|
|
|
self._get_results(results_list) |
|
102
|
|
|
self._save_memory(results_list) |
|
103
|
|
|
|
|
104
|
|
|
def run(self, start_time, max_time): |
|
105
|
|
|
self._run(start_time, max_time) |
|
106
|
|
|
|
|
107
|
|
|
def _save_memory(self, results): |
|
108
|
|
|
for result in results: |
|
109
|
|
|
if result.memory == "long": |
|
110
|
|
|
result.save_long_term_memory() |
|
111
|
|
|
|
|
112
|
|
|
|
SimonBlanke /
Hyperactive
