-
Notifications
You must be signed in to change notification settings - Fork 0
/
SRM_main.py
166 lines (162 loc) · 6.82 KB
/
SRM_main.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
# -*- coding: utf-8 -*-
"""
Created on Thu Jan 4 15:58:32 2018
@author: monakhov
"""
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.patches as patches
from scipy import stats
from matplotlib.collections import PatchCollection
def SRM_general(data,axis,clip_name,ring_radius,draw_plot=False,users=None,ring_width=10):
#ring_width = 50
if users == None:
users = list(data.keys())
users_scores = {}
timestamps = np.arange(0,160000,len(data[users[0]][clip_name][axis]))
for user in users:
users_scores[user] = 0
clip_length = len(data[users[0]][clip_name][axis])
num_of_tested_users = len(users)
result_arr = []
mode_arr = []
for i in range(0,clip_length-ring_width,ring_width):
tmp_arr = []
for user in users:
tmp_arr.append(data[user][clip_name][axis][i:i+ring_width+1])
tmp_arr = np.array(tmp_arr)
tmp_arr = tmp_arr.flatten()
#ring_mode = stats.mode(tmp_arr,axis=None).mode[0]
ring_mode = np.median(tmp_arr)
mode_arr.append(ring_mode)
for j in range(i,i+ring_width):
check_if_in_ring_arr = []
for user in users:
if (data[user][clip_name][axis][j]>=ring_mode-ring_radius and
data[user][clip_name][axis][j]<=ring_mode+ring_radius):
check_if_in_ring_arr.append(0)
#users_scores[user] += 1
elif (ring_mode+ring_radius > 180):
diff = ring_mode+ring_radius - 180
if (data[user][clip_name][axis][j]<=-180+diff and
data[user][clip_name][axis][j]>=-180):
check_if_in_ring_arr.append(0)
#users_scores[user] += 1
else:
users_scores[user] += 1
check_if_in_ring_arr.append(1)
elif (ring_mode-ring_radius < -180):
diff = 180 - (ring_mode-ring_radius)
if (data[user][clip_name][axis][j]<=180 and
data[user][clip_name][axis][j]>= 180-diff):
check_if_in_ring_arr.append(0)
#users_scores[user] += 1
else:
users_scores[user] += 1
check_if_in_ring_arr.append(1)
else:
users_scores[user] += 1
check_if_in_ring_arr.append(1)
result_arr.append(np.sum(check_if_in_ring_arr)/num_of_tested_users)
res = (1-np.sum(result_arr)/len(result_arr))*100
if (draw_plot == True):
ringboxes = []
for i in range(len(mode_arr)):
rect =patches.Rectangle((ring_width*i,mode_arr[i]-ring_radius),ring_width,2*ring_radius,hatch='/',fill=False)
ringboxes.append(rect)
pc = PatchCollection(ringboxes,match_original=True,hatch='//')
fig, ax = plt.subplots(1)
for user in users:
ax.plot(data[user][clip_name][axis])
#ax.set_xticklabels([0,0,10000,20000,30000,400000,50000,60000,70000])
ax.add_collection(pc)
# plt.title(clip_name)
plt.ylim( (-180, 180) )
plt.show()
plt.title('Similarity Ring Metric')
plt.ylabel('yaw values, degrees')
plt.xlabel('timestamp, ms')
#plt.xticks(np.arange(0,160000,len(data[user][clip_name][axis])))
#plt.savefig('byClip/'+clip_name[:-3]+'1.png')
#plt.xticks(plt.xticks()[0], np.linspace(0,160000,9))
return res,mode_arr,users_scores
def SRM_pair(clip_ref,clip_test,ring_radius,draw_plot=False,title=None):
clip_length = len(clip_ref)
res_arr = []
for i in range(clip_length):
if (clip_test[i]>=clip_ref[i]-ring_radius and
clip_test[i]<=clip_ref[i]+ring_radius):
res_arr.append(1)
elif (clip_ref[i]+ring_radius > 180):
diff = clip_ref[i]+ring_radius - 180
if (clip_test[i]<=-180+diff and
clip_test[i]>=-180):
res_arr.append(1)
else:
res_arr.append(0)
elif (clip_ref[i]-ring_radius < -180):
diff = 180 - (clip_ref[i]-ring_radius)
if (clip_test[i]<=180 and
clip_test[i]>= 180-diff):
res_arr.append(1)
else:
res_arr.append(0)
else:
res_arr.append(0)
res = np.sum(res_arr)/len(res_arr)*100
if (draw_plot == True):
fig, ax = plt.subplots(1)
ax.plot(clip_ref)
ax.plot(clip_test)
ringboxes = []
for i in range(clip_length):
rect =patches.Rectangle((i,clip_ref[i]-ring_radius),2,2*ring_radius)
ringboxes.append(rect)
pc = PatchCollection(ringboxes,match_original=True,alpha=0.1,edgecolor="none")
ax.add_collection(pc)
plt.ylim( (-180, 180) )
if title != None:
plt.title(title)
return res
def SRM_merge(datasets,axis,clip_names,ring_radius,draw_plot=False):
ring_width = 50
clip_length = len(datasets[0][list(datasets[0].keys())[0]][clip_names[0]][axis])
num_of_tested_users = len(datasets[0].keys())
full_data_arr = []
for user in datasets[0].keys():
for i in range(len(datasets)):
full_data_arr.append(datasets[i][user][clip_names[i]][axis])
result_arr = []
mode_arr = []
for i in range(0,clip_length-ring_width,ring_width):
tmp_arr = []
for traj_num in range(len(full_data_arr)):
tmp_arr.append(full_data_arr[traj_num][i:i+ring_width+1])
tmp_arr = np.array(tmp_arr)
tmp_arr = tmp_arr.flatten()
#ring_mode = stats.mode(tmp_arr,axis=None).mode[0]
ring_mode = np.median(tmp_arr)
mode_arr.append(ring_mode)
for j in range(i,i+ring_width):
check_if_in_ring_arr = []
for traj_num in range(len(full_data_arr)):
if (full_data_arr[traj_num][j]>=ring_mode-ring_radius and
full_data_arr[traj_num][j]<=ring_mode+ring_radius):
check_if_in_ring_arr.append(0)
else:
check_if_in_ring_arr.append(1)
result_arr.append(np.sum(check_if_in_ring_arr)/num_of_tested_users)
res = (1-np.sum(result_arr)/len(result_arr))*100
if (draw_plot == True):
ringboxes = []
for i in range(len(mode_arr)):
rect =patches.Rectangle((ring_width*i,mode_arr[i]-ring_radius),ring_width,2*ring_radius,hatch='/',fill=False)
ringboxes.append(rect)
pc = PatchCollection(ringboxes,match_original=True,hatch='//')
fig, ax = plt.subplots(1)
for traj_num in range(len(full_data_arr)):
ax.plot(full_data_arr[traj_num])
ax.add_collection(pc)
plt.title(clip_names[0][:-3])
plt.show()
return res,mode_arr