-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathtestset_eval.py
72 lines (61 loc) · 3.43 KB
/
testset_eval.py
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
"""
我们收集数据的测试集评估
"""
import numpy as np
from evaluate.test_evalutator import evaluate_model
from pace_predictor.predict_pace import magic_pace_inference
from pace_predictor.acc_pace_inference import pace_inference
from pedestrian_data import PedestrianDataset, default_low_pass_filter
def test_extra_1() -> None:
""" 消融实验1 """
print("---------------------- 消融实验1：不同姿态对模型的影响 ---------------------\n")
total_dist_error = []
total_dir_error = []
dataset = PedestrianDataset(["test_extra_0"], window_size=200,
acceleration_filter=default_low_pass_filter, skip_len=5)
for name, _ in dataset.loci.items():
print("----------data: {} ----------".format(name))
(dist_error, dir_error) = evaluate_model(dataset[name], pace_inference=pace_inference, compare=True)
total_dist_error.append(dist_error)
total_dir_error.append(dir_error)
print("测试集总距离平均误差： {}".format(sum(total_dist_error) / len(total_dist_error)))
print("测试集总方位角平均误差：{}".format(sum(total_dir_error) / len(total_dir_error)))
print()
print("测试集总距离方差： {}".format(np.asarray(total_dist_error).std()))
print("测试集总方位角方差：{}".format(np.asarray(total_dir_error).std()))
def test_extra_2() -> None:
""" 消融实验2 """
print("---------------------- 消融实验2：不同设备对模型的影响 ---------------------\n")
total_dist_error = []
total_dir_error = []
dataset = PedestrianDataset(["test_extra_1"], window_size=200,
acceleration_filter=default_low_pass_filter, skip_len=5)
for name, _ in dataset.loci.items():
print("----------data: {} ----------".format(name))
(dist_error, dir_error) = evaluate_model(dataset[name], pace_inference=pace_inference, compare=True)
total_dist_error.append(dist_error)
total_dir_error.append(dir_error)
print("测试集总距离平均误差： {}".format(sum(total_dist_error) / len(total_dist_error)))
print("测试集总方位角平均误差：{}".format(sum(total_dir_error) / len(total_dir_error)))
print()
print("测试集总距离方差： {}".format(np.asarray(total_dist_error).std()))
print("测试集总方位角方差：{}".format(np.asarray(total_dir_error).std()))
def test() -> None:
""" 测试集 """
print("---------------------- 测试集：模型总体性能评估 ---------------------\n")
total_dist_error = []
total_dir_error = []
dataset = PedestrianDataset(["test_eval"], window_size=200,
acceleration_filter=default_low_pass_filter, skip_len=5)
for name, _ in dataset.loci.items():
print("----------data: {} ----------".format(name))
(dist_error, dir_error) = evaluate_model(dataset[name], pace_inference=pace_inference, compare=True)
total_dist_error.append(dist_error)
total_dir_error.append(dir_error)
print("测试集总距离平均误差： {}".format(sum(total_dist_error) / len(total_dist_error)))
print("测试集总方位角平均误差：{}".format(sum(total_dir_error) / len(total_dir_error)))
print()
if __name__ == '__main__':
test() # 测试集：模型总体性能评估
test_extra_1() # 消融实验1：不同姿态对模型的影响
test_extra_2() # 消融实验2：不同设备对模型的影响