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
|
import sys, math, subprocess
dDay = 1
obsTimes = {}
verbose = False
def readTimes(obsfile):
times = []
for line in open(obsfile):
ll = line.split(':')
if ll:
times.append(3600*int(ll[0]) + 60*int(ll[1]) + int(float(ll[2])))
return times
def parseObsTimes():
for i in range (0,9):
obsTimes[i] = []
for i in range(1,8):
if dDay==1 and i==5:
continue
if dDay==2 and i==6:
continue
obsTimes[i] = readTimes('data/obstimes_%s_%s.txt' % (dDay, i))
# convert obsTimes[][] into travel-times:
for i in range(1,8):
ni = len(obsTimes[i])
if ni == len(obsTimes[i+1]) and ni > 100:
for j in range(ni):
obsTimes[i][j] = obsTimes[i+1][j] - obsTimes[i][j]
def validate(sumoBinary):
retcode = subprocess.call([sumoBinary, "-c", "data/spd-road.sumocfg"], stdout=sys.stdout, stderr=sys.stderr)
sys.stdout.flush()
sys.stderr.flush()
# analyzing the results...
# read the empirical times
simTimes = {}
for i in range (0,9):
simTimes[i] = []
# read the simulated times
obs2Nr = {'obs1': 1, 'obs2': 2, 'obs3': 3, 'obs4': 4, 'obs5': 5, 'obs6': 6, 'obs7': 7}
for line in open('data/detector.xml'):
if line.find('<interval')!=-1:
ll = line.split('"')
iObs = obs2Nr[ll[5]]
if int(ll[7]) > 0:
simTimes[iObs].append(float(ll[1]))
# convert simTimes[][] into travel-times:
for i in range(1,8):
ni = len(simTimes[i])
if ni == len(simTimes[i+1]) and ni>100:
for j in range(ni):
simTimes[i][j] = simTimes[i+1][j] - simTimes[i][j]
# compute final statistics
err = [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
averTT = [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
errAll = 0.0
cntAll = 0
if verbose:
f = open('data/sumo-obs-error.txt','w')
for i in range(1,7):
if len(obsTimes[i])<=100 or len(obsTimes[i+1])<=100:
continue
if len(obsTimes[i]) == len(simTimes[i]):
tmp = 0.0
for o, s in zip(obsTimes[i], simTimes[i]):
d = o - s
tmp += d*d
err[i] = math.sqrt(tmp/len(obsTimes[i]))
if verbose:
print >> f, "%s %s" %(i, err[i])
errAll += err[i]
cntAll += 1
if verbose:
f.close()
# finally, write the individual travel times into a csv-file
# this is not really needed when validate is called from calibrate as an intermediate
# step, but it makes analyzing of the result more simple.
# first the header
if verbose:
c = open('data/compare-tt.csv', 'w')
c.write('# indx;')
for i in range(1,7):
if len(obsTimes[i])>100 and len(obsTimes[i+1])>100:
c.write('obs%s;sim%s;' % (i,i))
c.write('\n')
# then the data, of course on the ones which are useable
for line in range(len(simTimes[1])):
c.write(repr(line)+';')
for i in range(1,7):
if len(obsTimes[i])>100 and len(obsTimes[i+1])>100:
ttObs = int(obsTimes[i][line])
ttSim = int(simTimes[i][line])
c.write(repr(ttObs)+';'+repr(ttSim)+';')
c.write('\n')
c.close()
return errAll/cntAll
|
