-
Notifications
You must be signed in to change notification settings - Fork 11
/
batch_test.py
202 lines (185 loc) · 8.03 KB
/
batch_test.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
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
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
import os
import argparse
import time
import warnings
from pdr import pdr
from dataloader import TestCase
import pandas as pd
from tqdm import tqdm
import multiprocessing as mp
import sys
from itertools import product
DATA_DIR_ROOT = "../Dataset-of-Pedestrian-Dead-Reckoning/"
WALK=["Hand-Walk", "Bag-Walk", "Pocket-Walk"]
OTHER=["Bag-Ride","Hand-Ride","Pocket-Pace","Pocket-Ride"]
EXCLUDE=[
"Hand-Walk/Hand-Walk-03-015",
"Hand-Walk/Hand-Walk-05-001",
"Hand-Walk/Hand-Walk-05-002",
"Hand-Walk/Hand-Walk-15-001",
"Pocket-Walk/Pocket-Walk-03-010"
]
def collected_dataset():
warnings.filterwarnings('ignore')
parser = argparse.ArgumentParser()
parser.add_argument("--use", type=str, default="walk", choices=["walk", "other", "all"], help="datasets to use")
parser.add_argument("--method", type=str, default="SVR", choices=["Mean", "DecisionTree", "Linear", "SVR", "RandomForest", "AdaBoost", "GradientBoosting", "Bagging", "ExtraTree"], help="step predictor method")
args = parser.parse_args()
datasets = []
if args.use == "walk":
for dir in WALK:
# get names of all directories in `dir`
datasets += [os.path.join(dir, name) for name in os.listdir(os.path.join(DATA_DIR_ROOT, dir))]
elif args.use == "other":
for dir in OTHER:
datasets += [os.path.join(dir, name) for name in os.listdir(os.path.join(DATA_DIR_ROOT, dir))]
else:
for dir in WALK+OTHER:
datasets += [os.path.join(dir, name) for name in os.listdir(os.path.join(DATA_DIR_ROOT, dir))]
datasets = [dataset for dataset in datasets if dataset not in EXCLUDE]
names=[]
dist_error=[]
dir_error=[]
dir_ratio=[]
fails=0
for dataset in tqdm(datasets ,desc="Testing"):
try:
test_case = TestCase(os.path.join(DATA_DIR_ROOT, dataset))
pdr(test_case, model_name=args.method)
dist_error.append(test_case.get_dist_error())
dir_error.append(test_case.get_dir_error())
dir_ratio.append(test_case.get_dir_ratio())
names.append(dataset)
except Exception as e:
print(f"Error when processing dataset {dataset}: {e}")
# traceback.print_exc()
fails+=1
continue
df = pd.DataFrame({"name": names, "dist_error": dist_error, "dir_error": dir_error, "dir_ratio": dir_ratio})
df = pd.DataFrame({"dataset": names, "dist_error": dist_error, "dir_error": dir_error, "dir_ratio": dir_ratio})
df.to_csv(f"result_{args.use}.csv", index=False)
print("Done, results saved to result.csv")
print(f"Average distance error: {df['dist_error'].mean()}")
print(f"Average direction error: {df['dir_error'].mean()}")
print(f"Average direction ratio: {df['dir_ratio'].mean()}")
print(f"Direction error < 15: {df[df['dir_error']<15].shape[0]}/{df.shape[0]}")
print(f"Direction ratio > 0.9: {df[df['dir_ratio']>0.9].shape[0]}/{df.shape[0]}")
print(f"Failed count: {fails}")
def test_step_method():
warnings.filterwarnings('ignore')
METHODS=['Mean','DecisionTree', 'Linear', 'SVR', 'RandomForest', 'AdaBoost', 'GradientBoosting', 'Bagging', 'ExtraTree']
datasets=[]
for dir in WALK:
# get names of all directories in `dir`
datasets += [os.path.join(dir, name) for name in os.listdir(os.path.join(DATA_DIR_ROOT, dir))]
dist_error=[]
dir_error=[]
dir_ratio=[]
success=[]
for method in METHODS:
_dist_error=[]
_dir_error=[]
_dir_ratio=[]
for dataset in tqdm(datasets ,desc=method):
try:
test_case = TestCase(os.path.join(DATA_DIR_ROOT, dataset))
pdr(test_case, model_name=method)
_dist_error.append(test_case.get_dist_error())
_dir_error.append(test_case.get_dir_error())
_dir_ratio.append(test_case.get_dir_ratio())
except:
continue
dist_error.append(sum(_dist_error)/len(_dist_error))
dir_error.append(sum(_dir_error)/len(_dir_error))
dir_ratio.append(sum(_dir_ratio)/len(_dir_ratio))
success.append(len(_dist_error))
df = pd.DataFrame({"method": METHODS, "dist_error": dist_error, "dir_error": dir_error, "dir_ratio": dir_ratio, "success": success})
df.to_csv("result_step_method.csv", index=False)
def worker(_args):
try:
test_case = TestCase(os.path.join(DATA_DIR_ROOT, _args["dataset"]))
pdr(
test_case,
model_name=_args["model_name"],
distance_frac_step=_args["distance_frac_step"],
clean_data=_args["clean_data"],
optimized_mode_ratio=_args["optimized_mode_ratio"],
butter_Wn=_args["butter_Wn"])
dist_error = test_case.get_dist_error()
dir_error = test_case.get_dir_error()
dir_ratio = test_case.get_dir_ratio()
return _args["dataset"], _args["model_name"], _args["distance_frac_step"], _args["clean_data"], _args["optimized_mode_ratio"], _args["butter_Wn"], dist_error, dir_error, dir_ratio
except Exception as e:
return _args["dataset"], _args["model_name"], _args["distance_frac_step"], _args["clean_data"], _args["optimized_mode_ratio"], _args["butter_Wn"], 0, 0, 0
def mute():
sys.stdout = open(os.devnull, 'w')
sys.stderr = open(os.devnull, 'w')
def mp_test():
datasets = []
for dir in WALK:
datasets += [os.path.join(dir, name) for name in os.listdir(os.path.join(DATA_DIR_ROOT, dir))]
datasets = [dataset for dataset in datasets if dataset not in EXCLUDE]
args = {
"model_name": {
"default":"RandomForest",
"options": ["SVR", "RandomForest", "Mean", "AdaBoost", "ExtraTree"]
},
"distance_frac_step": {
"default": 5,
"options": [3,4,5,6,7]
},
"clean_data": {
"default": 5,
"options": [3,4,5,6,7,8,9,10]
},
"optimized_mode_ratio": {
"default": 0.9,
"options": [0, 0.1, 0.2, 0.4, 0.5, 0.6, 0.8, 0.9, 1.0]
},
"butter_Wn": {
"default": 0.005,
"options": [0.001, 0.002, 0.005, 0.01, 0.02, 0.03, 0.05, 0.08, 0.1]
},
}
datasets.append("../test_case0")
todos=[]
# for dataset in datasets:
# for key, value in args.items():
# for option in value["options"]:
# _args = {
# "dataset": dataset,
# "model_name": args["model_name"]["default"],
# "distance_frac_step": args["distance_frac_step"]["default"],
# "clean_data": args["clean_data"]["default"],
# "optimized_mode_ratio": args["optimized_mode_ratio"]["default"],
# "butter_Wn": args["butter_Wn"]["default"]
# }
# _args[key] = option
# todos.append(_args)
for todo in product(datasets,
args["model_name"]["options"],
args["distance_frac_step"]["options"],
args["clean_data"]["options"],
args["optimized_mode_ratio"]["options"],
args["butter_Wn"]["options"]):
todos.append({
"dataset": todo[0],
"model_name": todo[1],
"distance_frac_step": todo[2],
"clean_data": todo[3],
"optimized_mode_ratio": todo[4],
"butter_Wn": todo[5]
})
with mp.Pool(processes=72, initializer=mute) as pool:
res = list(tqdm(pool.imap(worker, todos), total=len(todos)))
df = pd.DataFrame(res, columns=["dataset", "model_name", "distance_frac_step", "clean_data", "optimized_mode_ratio", "butter_Wn", "dist_error", "dir_error", "dir_ratio"])
df.to_csv("result_mp_adjust_case0.csv", index=False)
def test_on_TestSet():
start=time.time()
datasets = [f"../TestSet/test{i}" for i in (1,2,3,5,6,7,8,9,10,11)]
for dataset in datasets:
test_case = TestCase(dataset)
pdr(test_case)
print(f"Finished {len(datasets)} test cases in {time.time()-start} seconds")
if __name__ == "__main__":
collected_dataset()