-
Notifications
You must be signed in to change notification settings - Fork 0
/
FreeTSE.py
831 lines (737 loc) · 28.6 KB
/
FreeTSE.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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
#coding:utf-8
from util import *
import random as pyrand
from pylab import *
import pandas as pd
import sys, os
import copy
import configparser
import warnings
class FreeTSE:
def __init__(self, ini=None, name="untitled", gui_mode=False):
"""Initialization
Parameters
----------
ini: str
Path to the input ini file.
name: str
Name of the project.
"""
self.name = name
self.gui_mode = gui_mode
if ini != None:
self.read_ini(ini)
print(self.name)
def read_ini(self, ini):
"""Read .ini to define default scenario
Parameters
----------
ini: str
Path to the input ini file.
"""
self.ini = ini
try:
cfg = configparser.ConfigParser()
cfg.read(ini)
self.name = cfg["Data"]["name"]
dt = float(cfg["Resolution"]["dt"])
dx = float(cfg["Resolution"]["dx"])
mint = float(cfg["Resolution"]["time_min"])
maxt = float(cfg["Resolution"]["time_max"])
minx = float(cfg["Resolution"]["space_min"])
maxx = float(cfg["Resolution"]["space_max"])
number_of_lanes = float(cfg["Resolution"]["number_of_lanes"])
self.speed_data_name = cfg["Data"]["speed"]
self.speed_label_t = cfg["Data"]["speed_label_t"]
self.speed_label_x = cfg["Data"]["speed_label_x"]
self.speed_label_v = cfg["Data"]["speed_label_v"]
if "density" in cfg["Data"]:
self.density_data_name = cfg["Data"]["density"]
self.density_label_t = cfg["Data"]["density_label_t"]
self.density_label_x = cfg["Data"]["density_label_x"]
self.density_label_k = cfg["Data"]["density_label_k"]
else:
self.density_data_name = None
if "flow" in cfg["Data"]:
self.flow_data_name = cfg["Data"]["flow"]
self.flow_label_t = cfg["Data"]["flow_label_t"]
self.flow_label_x = cfg["Data"]["flow_label_x"]
self.flow_label_q = cfg["Data"]["flow_label_q"]
else:
self.flow_data_name = None
if "GroundTruth" in cfg and "true_density" in cfg["GroundTruth"] and cfg["GroundTruth"]["true_density"] != "None":
self.groundtruth = True
self.density_dat_true_name = cfg["GroundTruth"]["true_density"]
self.true_density_label_t = cfg["GroundTruth"]["density_label_t"]
self.true_density_label_x = cfg["GroundTruth"]["density_label_x"]
self.true_density_label_k = cfg["GroundTruth"]["density_label_k"]
self.flow_dat_true_name = None
elif "GroundTruth" in cfg and "true_flow" in cfg["GroundTruth"] and cfg["GroundTruth"]["true_flow"] != "None":
self.groundtruth = True
self.density_dat_true_name = None
self.flow_dat_true_name = cfg["GroundTruth"]["true_flow"]
self.true_flow_label_t = cfg["GroundTruth"]["flow_label_t"]
self.true_flow_label_x = cfg["GroundTruth"]["flow_label_x"]
self.true_flow_label_q = cfg["GroundTruth"]["flow_label_q"]
else:
self.groundtruth = False
except:
if self.gui_mode == True:
sg.popup(f"Error: Something is wrong with the .ini file `{ini}`")
raise Exception(f"Something is wrong with the .ini file `{ini}`")
self.set_data(mint, maxt, dt, minx, maxx, dx, number_of_lanes)
def set_scenario(self, name, dt, dx, mint, maxt, minx, maxx, number_of_lanes, speed_data_name, speed_label_t, speed_label_x, speed_label_v, density_data_name=None, density_label_t=None, density_label_x=None, density_label_k=None, flow_data_name=None, flow_label_t=None, flow_label_x=None, flow_label_q=None, density_dat_true_name=None, true_density_label_t=None, true_density_label_x=None, true_density_label_k=None, flow_dat_true_name=None, true_flow_label_t=None, true_flow_label_x=None, true_flow_label_q=None):
"""Set estimation scenario.
Parameters
----------
name: str
scenario name
dt: number
temporal resolution of estimation (s)
dx: number
spatial resolution of estimation (m)
mint: number
initial time of the time-space region to be estimated
maxt: number
last time of the time-space region to be estimated
minx: number
upstream-end position of the time-space region to be estimated
maxx: number
downstream-end position of the time-space region to be estimated
number_of_lanes: number
the number of lanes of the target section
speed_data_name: str
path to speed data (probe vehicle data)
speed_label_t: str
column name of the time in the speed data
speed_label_x: str
column name of the position in the speed data
speed_label_v: str
column name of the speed in the speed data
density_data_name: str
path to density data (detector data)
if it does not exist, set `density_data_name=None`
density_label_t: str
column name of the time in the density data
density_label_x: str
column name of the position in the density data
density_label_k: str
column name of the density in the density data
flow_data_name: str
path to flow data (detector data)
if it does not exist, set `flow_data_name=None`
flow_label_t: str
flow_label_x: str
flow_label_q: str
density_dat_true_name: str
path to ground truth density data for accuracy validation
if it does not exist, set `density_dat_true_name=None`
true_density_label_t: str
true_density_label_x: str
true_density_label_k: str
flow_dat_true_name: str
path to ground truth flow data for accuracy validation
if it does not exist, set `flow_dat_true_name=None`
true_flow_label_t: str
true_flow_label_x: str
true_flow_label_q: str
"""
self.name = name
dt = dt
dx = dx
mint = mint
maxt = maxt
minx = minx
maxx = maxx
number_of_lanes = number_of_lanes
self.speed_data_name = speed_data_name
self.speed_label_t = speed_label_t
self.speed_label_x = speed_label_x
self.speed_label_v = speed_label_v
if density_data_name != None:
self.density_data_name = density_data_name
self.density_label_t = density_label_t
self.density_label_x = density_label_x
self.density_label_k = density_label_k
else:
self.density_data_name = "None"
if flow_data_name != None:
self.flow_data_name = flow_data_name
self.flow_label_t = flow_label_t
self.flow_label_x = flow_label_x
self.flow_label_q = flow_label_q
else:
self.flow_data_name = "None"
if density_dat_true_name != None:
self.groundtruth = True
self.flow_dat_true_name = None
self.density_dat_true_name = density_dat_true_name
self.true_density_label_t = true_density_label_t
self.true_density_label_x = true_density_label_x
self.true_density_label_k = true_density_label_k
elif flow_dat_true_name != None:
self.groundtruth = True
self.density_dat_true_name = None
self.flow_dat_true_name = flow_dat_true_name
self.true_flow_label_t = true_flow_label_t
self.true_flow_label_x = true_flow_label_x
self.true_flow_label_q = true_flow_label_q
else:
self.groundtruth = False
self.set_data(mint, maxt, dt, minx, maxx, dx, number_of_lanes)
def set_data(self, mint=None, maxt=None, dt=None, minx=None, maxx=None, dx=None, number_of_lanes=None):
"""Read and structure data
"""
#fundamental parameters
if mint != None:
self.mint = mint
if maxt != None:
self.maxt = maxt
if dt != None:
self.dt = dt
if minx != None:
self.minx = minx
if maxx != None:
self.MAXX = maxx
if dx != None:
self.dx = dx
if number_of_lanes != None:
self.number_of_lanes = number_of_lanes
self.tsize = int((self.maxt-self.mint)/self.dt)
self.xsize = int((self.MAXX-self.minx)/self.dx)
#precision parameters
self.detector_precision = 0.01**2
self.initial_precision = (self.number_of_lanes*0.2)**2
self.concervation_precision = 0.01**2
self.cv_speed_precision = 3**2
self.initial_density = 0.2**2
#traffic data loading
vv_raw = pd.read_csv(self.speed_data_name)
vv_mod = vstack([vv_raw[self.speed_label_t], vv_raw[self.speed_label_x], vv_raw[self.speed_label_v]]).T
self.vv = self.griddata_generation(vv_mod, min_value=0, max_value=self.dx/self.dt, interpolation="both")
self.probe_record = vv_mod
if self.density_data_name not in [None, "None"]:
kk_raw = pd.read_csv(self.density_data_name)
kk_mod = vstack([kk_raw[self.density_label_t], kk_raw[self.density_label_x], kk_raw[self.density_label_k]]).T
self.kk = self.griddata_generation(kk_raw, missing_value=-1, interpolation="time")
self.detector_record = kk_mod
elif self.flow_data_name not in [None, "None"]:
qq_raw = pd.read_csv(self.flow_data_name)
qq_mod = vstack([qq_raw[self.flow_label_t], qq_raw[self.flow_label_x], qq_raw[self.flow_label_q]]).T
self.qq = self.griddata_generation(qq_mod, missing_value=-1, interpolation="time")
self.kk = self.qq/self.vv
self.kk[self.qq == -1] = -1
self.detector_record = qq_mod
else:
if self.gui_mode == True:
sg.popup(f"Error: Either density or flow data is required")
raise Exception("Either density or flow data is required")
#ground truth data loading
if self.groundtruth:
if self.density_dat_true_name != None:
kk_true_raw = pd.read_csv(self.density_dat_true_name)
kk_true_mod = vstack([kk_true_raw[self.true_density_label_t], kk_true_raw[self.true_density_label_x], kk_true_raw[self.true_density_label_k]]).T
self.kk_true = self.griddata_generation(kk_true_mod, missing_value=-1)
else:
qq_true_raw = pd.read_csv(self.flow_dat_true_name)
qq_true_mod = vstack([qq_true_raw[self.true_flow_label_t], qq_true_raw[self.true_flow_label_x], qq_true_raw[self.true_flow_label_q]]).T
self.qq_true = self.griddata_generation(qq_true_mod, missing_value=-1, interpolation="time")
self.kk_true = self.qq_true/self.vv
self.kk_true[self.qq_true == -1] = -1
def griddata_generation(self, data_raw, min_value=None, max_value=None, missing_value=0, interpolation=0):
"""generate grid data from input table
Parameters
----------
data_raw: list
Raw input traffic data.
min_value: float or None
Lowerbound of an output value.
max_value: float or None
Upperbound of an output value.
missing_value: float
Value used to interpolate missing value.
interpolation: 0 or "both" or "time" or "space"
Direction of missing value interpolation.
Returns
-------
data_array: array
Traffic data converted to grid data.
"""
data_array = zeros([self.tsize, self.xsize]) + missing_value
data_dic = {}
for l in data_raw:
t = l[0]
x = l[1]
d = l[2]
n = int((t-self.mint)//self.dt)
i = int((x-self.minx)//self.dx)
if (n,i) in data_dic:
data_dic[n,i].append(d)
else:
data_dic[n,i] = [d]
for (n,i) in data_dic.keys():
if 0 <= n < self.tsize and 0 <= i < self.xsize:
data_array[n,i] = average(data_dic[n,i]) #harmonic mean is suitable depending on the probe vehicle data specification, but we cannot know
if min_value != None:
if data_array[n,i] < 0:
data_array[n,i] = 0
if max_value != None:
if data_array[n,i] > self.dx/self.dt:
speed_exceeds_flag = 1
print("%.1f>%.1f"%(data_array[n,i],self.dx/self.dt), end=" ")
data_array[n,i] = self.dx/self.dt
print()
#missing data interpolation
num_inter = 0
if interpolation in ("both", "space"):
for n in range(self.tsize):
for i in range(1, self.xsize):
if data_array[n,i] == missing_value and data_array[n][i-1] != missing_value:
data_array[n,i] = data_array[n][i-1]
num_inter += 1
if interpolation in ("both", "time"):
for n in range(1, self.tsize):
for i in range(self.xsize):
if data_array[n,i] == missing_value and data_array[n-1][i] != missing_value:
data_array[n,i] = data_array[n-1][i]
num_inter += 1
for n in range(self.tsize-2, -1, -1):
for i in range(self.xsize):
if data_array[n,i] == missing_value and data_array[n+1][i] != missing_value:
data_array[n,i] = data_array[n+1][i]
num_inter += 1
return data_array
def smooth_speeddata(self, tagg, xagg):
vv_new = zeros([self.tsize, self.xsize])
for t in range(self.tsize):
for x in range(self.xsize):
vlist = []
for tt in range(tagg):
for xx in range(xagg):
if t+tt < self.tsize and x+xx < self.xsize:
vlist.append(self.vv[t+tt,x+xx])
vv_new[t,x] = average(vlist)
self.vv = vv_new
def filtering(self):
"""traffic state estimation by Kalman filtering
"""
self.yy = self.kk
self.k_prio = zeros([self.tsize, self.xsize])
self.k_post = zeros([self.tsize, self.xsize])
self.k_smoo = zeros([self.tsize, self.xsize])
xx_prio = ones(self.xsize)*self.initial_density
xx_post = ones(self.xsize)*self.initial_density
xx_smoo = ones(self.xsize)*self.initial_density
V_prio = zeros([self.xsize, self.xsize])
V_post = ones([self.xsize, self.xsize])*self.initial_precision
V_smoo = zeros([self.xsize, self.xsize])
#Q = identiry(self.xsize])*self.concervation_precision
Q = zeros([self.xsize, self.xsize])
self.k_prio[0] = xx_prio
self.k_post[0] = xx_post
self.V_prio_dic = {}
self.V_post_dic = {}
self.V_smoo_dic = {}
self.xx_prio_dic = {}
self.xx_post_dic = {}
self.xx_smoo_dic = {}
self.F_dic = {}
for n in range(self.tsize):
F = zeros([self.xsize, self.xsize])
for i in range(self.xsize):
if i > 0:
F[i,i] = (1-self.dt/self.dx*self.vv[n,i])
F[i,i-1] = self.dt/self.dx*self.vv[n,i-1]
Q[i,i] = self.cv_speed_precision
Q[i,i-1] = self.cv_speed_precision
else:
F[0,0] = 1
Q[0,0] = self.cv_speed_precision
H = zeros([self.xsize, self.xsize])
R = zeros([self.xsize, self.xsize])
for i in range(self.xsize):
if self.kk[n,i] != -1:
H[i,i] = 1
R[i,i] = self.detector_precision
xx_prio = F @ xx_post
V_prio = F @ V_post @ F.T + Q
K = V_prio @ H.T @ pinv(H @ V_prio @ H.T + R)
xx_post = xx_prio + K @ (self.yy[n] - H @ xx_prio)
V_post = V_prio - K @ H @ V_prio
#ad-hoc cleansing
for i in lange(xx_prio):
if xx_prio[i] < 0:
xx_prio[i] = 0
#if xx_prio[i] > 0.2*self.number_of_lanes:
# xx_prio[i] = 0.2*self.number_of_lanes
if xx_post[i] < 0:
xx_post[i] = 0
#if xx_post[i] > 0.2*self.number_of_lanes:
# xx_post[i] = 0.2*self.number_of_lanes
self.k_prio[n] = xx_prio
self.k_post[n] = xx_post
self.V_prio_dic[n] = copy.copy(V_prio)
self.V_post_dic[n] = copy.copy(V_post)
self.xx_prio_dic[n] = copy.copy(xx_prio)
self.xx_post_dic[n] = copy.copy(xx_post)
self.F_dic[n] = copy.copy(F)
self.V_smoo_dic[self.tsize-1] = copy.copy(V_post)
self.xx_smoo_dic[self.tsize-1] = copy.copy(xx_post)
self.k_smoo[self.tsize-1] = xx_post
def smoothing(self):
"""traffic state estimation by RST smoothing
"""
for n in range(self.tsize-2, -1, -1):
A = self.V_post_dic[n] @ self.F_dic[n].T @ pinv(self.V_prio_dic[n+1])
xx_smoo = self.xx_post_dic[n] + A @ (self.xx_smoo_dic[n+1] - self.xx_prio_dic[n+1])
V_smoo = self.V_post_dic[n] + A @ (self.V_smoo_dic[n+1] - self.V_prio_dic[n+1]) @ A.T
#without smoothing
#xx_smoo = self.xx_post_dic[n]
#V_smoo = self.V_post_dic[n]
#ad-hoc cleansing
for i in lange(xx_smoo):
if xx_smoo[i] < 0:
xx_smoo[i] = 0
#if xx_smoo[i] > 0.2*self.number_of_lanes:
# xx_smoo[i] = 0.2*self.number_of_lanes
self.k_smoo[n] = xx_smoo
self.xx_smoo_dic[n] = copy.copy(xx_smoo)
self.V_smoo_dic[n] = copy.copy(V_smoo)
if sum(xx_smoo.flatten()) == sum(self.xx_post_dic[n].flatten()):
print("no smoothing!")
def estimation(self):
"""call filtering and smoothing
"""
self.filtering()
self.smoothing()
self.compute_cum_curves()
def compute_cum_curves(self):
"""compute cumulative arrival/depature curves
"""
self.N = zeros([self.tsize, self.xsize])
for t in range(self.tsize):
if t > 0:
self.N[t,0] = self.N[t-1,0] + self.k_smoo[t,0]*self.vv[t,0]*self.dt
for x in range(1, self.xsize):
self.N[t,x] = self.N[t,x-1] - self.k_smoo[t,x]*self.dx
def accuracy_evaluation(self, print_mode=1):
"""evaluate accuracy against ground truth data
"""
if self.groundtruth:
if self.density_dat_true_name != None:
if print_mode:
print("density error")
self.rmse_prio = rmse(self.kk_true[self.kk_true>0].flatten(), self.k_prio[self.kk_true>0].flatten())
self.rmse_post = rmse(self.kk_true[self.kk_true>0].flatten(), self.k_post[self.kk_true>0].flatten())
self.rmse_smoo = rmse(self.kk_true[self.kk_true>0].flatten(), self.k_smoo[self.kk_true>0].flatten())
self.mape_prio = mae(self.kk_true[self.kk_true>0].flatten(), self.k_prio[self.kk_true>0].flatten(), 1)*100
self.mape_post = mae(self.kk_true[self.kk_true>0].flatten(), self.k_post[self.kk_true>0].flatten(), 1)*100
self.mape_smoo = mae(self.kk_true[self.kk_true>0].flatten(), self.k_smoo[self.kk_true>0].flatten(), 1)*100
else:
if print_mode:
print("flow error")
self.rmse_prio = rmse(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_prio[self.qq_true>0]).flatten())
self.rmse_post = rmse(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_post[self.qq_true>0]).flatten())
self.rmse_smoo = rmse(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_smoo[self.qq_true>0]).flatten())
self.mape_prio = mae(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_prio[self.qq_true>0]).flatten(), 1)*100
self.mape_post = mae(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_post[self.qq_true>0]).flatten(), 1)*100
self.mape_smoo = mae(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_smoo[self.qq_true>0]).flatten(), 1)*100
if print_mode:
#print("RMSE (prio): %.4f"%self.rmse_prio)
#print("RMSE (post): %.4f"%self.rmse_post)
print("RMSE: %.4f"%self.rmse_smoo)
#print("MAPE (prio): %.1f%%"%self.mape_prio)
#print("MAPE (post): %.1f%%"%self.mape_post)
print("MAPE: %.1f%%"%self.mape_smoo)
def save_results(self, name):
"""output estimation results as csv
"""
out = [["t (s)", "x (m)", "estimated k (veh/m)", "v (m/s)", "estimated q (veh/s)", "true k (veh/m)"]]
if self.groundtruth and self.density_dat_true_name == None:
out[0][-1] = "true q (veh/s)"
if not self.groundtruth:
out[0] = out[0][:-1]
for n in range(self.tsize):
for i in range(self.xsize):
l = [
n*self.dt,
i*self.dx,
self.k_smoo[n,i],
self.vv[n,i],
self.k_smoo[n,i]*self.vv[n,i],
]
if self.groundtruth:
l.append(self.kk_true[n,i] if self.density_dat_true_name != None else self.qq_true[n,i])
else:
pass
out.append(l)
writecsv(name, out)
def get_results(self):
return self.k_smoo*self.vv, self.k_smoo, self.vv
def visualize(self, speed=0, prior=0, posterior=0, smooth=0, posterior_stddev=0, smooth_stddev=0, true=0, observation=0, qk=0, qmax=0.2, cumcurves=0, qmax_cum=0, scatter=0, cum_true=0, timeseries=0, save=0, inputdata=0, fname="img"):
"""visualize the results
"""
if prior:
figure(figsize=(12,3))
title("a prior")
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
imshow(self.k_prio.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.1*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("density (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_prior.png"%fname)
if posterior:
figure(figsize=(12,3))
title("estimated density")
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
imshow(self.k_post.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.1*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("estimated density (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_posterior.png"%fname)
if smooth:
figure(figsize=(12,3))
title("estimated density")
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
imshow(self.k_smoo.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.1*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("estimated density (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_estimated_density.png"%fname)
if posterior_stddev:
vars = []
for n in range(1,self.tsize):
vars.append([self.V_post_dic[n][i,i] for i in range(self.xsize)])
figure(figsize=(12,3))
title("stddev of estimated density")
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
imshow(sqrt(array(vars)).T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.05*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("posterior density stddev (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_posterior_stddev.png"%fname)
if smooth_stddev:
vars = []
for n in range(1,self.tsize):
vars.append([self.V_smoo_dic[n][i,i] for i in range(self.xsize)])
figure(figsize=(12,3))
title("stddev of estimated density")
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
imshow(sqrt(array(vars)).T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.05*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("estimated density stddev (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_smooth_stddev.png"%fname)
if true:
if self.groundtruth:
figure(figsize=(12,3))
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
title("true density")
imshow(self.kk_true.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.1*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("density (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_true_density.png"%fname)
if observation:
figure(figsize=(12,3))
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
title("observed density")
imshow(self.kk.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=0.1*self.number_of_lanes, cmap=cm_kusakabe_pb2)
colorbar().set_label("density (veh/m)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_observation.png"%fname)
if speed:
figure(figsize=(12,3))
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
title("observed speed")
imshow(self.vv.T, origin="lower", interpolation="nearest", aspect="auto", extent=(self.mint, self.maxt, self.minx, self.MAXX), vmin=0, vmax=self.dx/self.dt, cmap=cm_kusakabe_pb2)
colorbar().set_label("speed (m/s)")
xlabel("t (s)")
ylabel("x (m)")
grid()
if save:
savefig("%s_speed.png"%fname)
if inputdata:
figure(figsize=(12,3))
subplots_adjust(top=.9, bottom=.2, left=.125, right=.9, wspace=.2, hspace=.2)
title("input data")
plot(self.probe_record[:,0], self.probe_record[:,1], "b.", label="speed")
plot(self.detector_record[:,0], self.detector_record[:,1], "rx", label="detector")
xlim([self.mint, self.maxt])
ylim([self.minx, self.MAXX])
xlabel("t (s)")
ylabel("x (m)")
legend()
grid()
if save:
savefig("%s_inputdata.png"%fname)
if qk:
figure()
k_max = 0.1*self.number_of_lanes
q_max = qmax
q_max = percentile(self.qq_true.flatten(), 95)*1.5
k_max = percentile(self.kk_true.flatten(), 95)*1.5
hist2d(self.k_smoo[self.k_smoo>0].flatten(), (self.k_smoo[self.k_smoo>0]*self.vv[self.k_smoo>0]).flatten(), bins=20, range=[[0,k_max],[0,q_max]], cmap=cm_kusakabe_pb2)
colorbar().set_label("flequency")
xlabel("estimated k (veh/m)")
ylabel("estimated q (veh/s)")
grid()
if save:
savefig("%s_qk.png"%fname)
if cumcurves:
tt = arange(self.mint, self.maxt, self.dt)
figure(figsize=(12,3))
plot(tt, self.N[:,0], "r", label="arrival")
plot(tt, self.N[:,-1], "b", label="departure")
grid()
xlabel("t (s)")
ylabel("cumulative N (veh)")
legend()
figure(figsize=(12,3))
plot(tt, self.N[:,0]-self.N[:,-1], "r", label="# of vehicles")
ylim(ymin=0)
grid()
xlabel("t (s)")
ylabel("existing N (veh)")
if qmax_cum == 0:
qmax = (self.N[-1,-1] - self.N[0,-1])/(self.maxt-self.mint)
figure(figsize=(12,3))
plot(tt, self.N[:,0]-(tt-self.mint)*qmax, "r", label="arrival")
plot(tt, self.N[:,-1]-(tt-self.mint)*qmax, "b", label="departure")
plot([self.mint, self.mint+self.dt*self.xsize], [0, -self.dt*self.xsize*qmax], "--", c="gray", label="ref. slope %.3f"%(-qmax))
for t in linspace(self.mint, self.maxt, 10):
plot([t, t+self.dt*self.xsize], [0, -self.dt*self.xsize*qmax], "--", c="gray")
grid()
xlabel("t (s)")
ylabel("oblique N (veh)")
legend()
#if cum_true and self.groundtruth:
if scatter and self.groundtruth:
k_max = 0.1*self.number_of_lanes
# figure(figsize=(4,4))
# subplot(111, aspect="equal")
# title("a prior")
# hist2d(self.kk_true[self.kk_true>0].flatten(), self.k_prio[self.kk_true>0].flatten(), bins=20, range=[[0,k_max],[0,k_max]], cmap=cm_kusakabe_pb2)
# colorbar().set_label("flequency")
# plot([0,k_max],[0,k_max],"r--")
# xlabel("true density")
# ylabel("smoothed density")
# grid()
#
# figure(figsize=(4,4))
# subplot(111, aspect="equal")
# title("a posterior")
# hist2d(self.kk_true[self.kk_true>0].flatten(), self.k_post[self.kk_true>0].flatten(), bins=20, range=[[0,k_max],[0,k_max]], cmap=cm_kusakabe_pb2)
# colorbar().set_label("flequency")
# plot([0,k_max],[0,k_max],"r--")
# xlabel("true density")
# ylabel("smoothed density")
# grid()
k_max = percentile(self.kk_true.flatten(), 95)*1.5
figure(figsize=(4,4))
subplot(111, aspect="equal")
subplots_adjust(top=.9, bottom=.1, left=.2, right=.9, wspace=.2, hspace=.2)
#title("estimates (MAP/smoothed)")
hist2d(self.kk_true[self.kk_true>0].flatten(), self.k_smoo[self.kk_true>0].flatten(), bins=20, range=[[0,k_max],[0,k_max]], cmap=cm_kusakabe_pb2)
colorbar().set_label("flequency")
plot([0,k_max],[0,k_max],"r--")
xlabel("true density (veh/m)")
ylabel("estimated density (veh/m)")
grid()
if save:
savefig("%s_k_scatter.png"%fname)
if self.density_dat_true_name == None:
q_max = qmax
q_max = percentile(self.qq_true.flatten(), 95)*1.5
figure(figsize=(4,4))
subplot(111, aspect="equal")
subplots_adjust(top=.9, bottom=.1, left=.2, right=.9, wspace=.2, hspace=.2)
#title("estimates (MAP/smoothed)")
hist2d(self.qq_true[self.qq_true>0].flatten(), (self.vv[self.qq_true>0]*self.k_smoo[self.qq_true>0]).flatten(), bins=20, range=[[0,q_max],[0,q_max]], cmap=cm_kusakabe_pb2)
colorbar().set_label("flequency")
plot([0,q_max],[0,q_max],"r--")
xlabel("true flow (veh/s)")
ylabel("estimated flow (veh/s)")
grid()
if save:
savefig("%s_q_scatter.png"%fname)
if timeseries and self.groundtruth:
for x in lange(self.kk_true[0]):
if self.kk_true[0,x] != -1:
if self.density_dat_true_name != None:
figure(figsize=(12,4))
title("x=%.0f"%(self.dx*x))
plot(arange(self.mint, self.maxt, self.dt), self.kk_true[:,x], "r", label="ground truth")
plot(arange(self.mint, self.maxt, self.dt), self.k_smoo[:,x], "b--", label="estimated")
legend(loc="best")
xlabel("t")
ylabel("flow")
ylim(ymin=0)
else:
figure(figsize=(12,4))
title("x=%.0f"%(self.dx*x))
plot(arange(self.mint, self.maxt, self.dt), self.qq_true[:,x], "r", label="ground truth")
plot(arange(self.mint, self.maxt, self.dt), self.vv[:,x]*self.k_smoo[:,x], "b--", label="estimated")
legend(loc="best")
xlabel("t")
ylabel("flow")
ylim(ymin=0)
if save:
savefig("%s_timeseries_%d.png"%(fname, x))
show()
if __name__ == "__main__":
tse = FreeTSE()
tse.set_scenario(
name = "ngsim_trajectory",
dt = 4,
dx = 100,
mint = 0,
maxt = 800,
minx = 0,
maxx = 500,
number_of_lanes = 5,
speed_data_name = "./dat/ngsim_sampled_trajectories.csv",
speed_label_t = "t",
speed_label_x = "x",
speed_label_v = "v",
density_data_name = None,
density_label_t = "t",
density_label_x = "x",
density_label_k = "k",
flow_data_name = "./dat/ngsim_grid_flow_200m.csv",
flow_label_t = "t",
flow_label_x = "x",
flow_label_q = "q",
density_dat_true_name = None,
true_density_label_t = "t",
true_density_label_x = "x",
true_density_label_k = "k",
flow_dat_true_name = "./dat/ngsim_grid_flow_400m.csv",
true_flow_label_t = "t",
true_flow_label_x = "x",
true_flow_label_q = "q"
)
tse.estimation()
tse.accuracy_evaluation()
fname = "res_test"
tse.save_results(fname+".csv")
tse.visualize(smooth=1, true=1, speed=1, timeseries=1, inputdata=1, save=1, fname=fname)
q, k, v = tse.get_results()
print("flow", q)
print("density", k)
print("speed", v)