Table of Contents
Largest elements
Links with the largest flow
4069 272 15 W 21 St & 6 Ave W 22 St & 10 Ave 4111 64 264 Pershing Square North E 24 St & Park Ave S 4195 359 392 Central Park S & 6 Ave 5 Ave & E 73 St 4290 64 63 Pershing Square North W 41 St & 8 Ave 4385 358 383 E 85 St & York Ave E 85 St & 3 Ave 4396 64 143 Pershing Square North W 33 St & 7 Ave 4648 272 139 W 21 St & 6 Ave 9 Ave & W 22 St 4913 252 173 12 Ave & W 40 St West St & Chambers St 4989 137 137 Centre St & Chambers St Centre St & Chambers St 6466 359 453 Central Park S & 6 Ave 5 Ave & E 78 St 11986 359 359 Central Park S & 6 Ave Central Park S & 6 Ave
The most and the least active stations
> wdir <- 'E:/data/bike/CitiBike/csv/2015-16'
> setwd(wdir)
> T <- read.csv('SO2sp.csv',head=TRUE,sep=';',stringsAsFactors=FALSE)
> T[1,]
v Name sumO sumI o0 o1 o2 o3 o4 o5 o6 o7 o8 o9 o10 o11 o12
1 1 W 26 St & 10 Ave 60833 61442 542 388 215 130 124 94 59 56 50 68 65 195 752
o13 o14 o15 o16 o17 o18 o19 o20 o21 o22 o23 o24 o25 o26 o27 o28
1 874 988 1423 1910 2556 2257 1864 1478 1349 1276 1495 1481 1596 1515 1598 1648
o29 o30 o31 o32 o33 o34 o35 o36 o37 o38 o39 o40 o41 o42 o43
1 1555 1576 1470 1525 1594 1985 2451 3362 3167 2757 2062 1634 1543 1424 1177
o44 o45 o46 o47 i0 i1 i2 i3 i4 i5 i6 i7 i8 i9 i10 i11 i12 i13 i14
1 1153 1100 677 575 404 260 193 164 131 85 103 70 63 47 98 236 217 641 769
i15 i16 i17 i18 i19 i20 i21 i22 i23 i24 i25 i26 i27 i28 i29
1 1039 1699 2306 2157 1739 1390 1288 1307 1525 1471 1524 1591 1678 1645 1687
i30 i31 i32 i33 i34 i35 i36 i37 i38 i39 i40 i41 i42 i43 i44 i45
1 1829 1864 2027 2644 3185 3364 3410 2926 2329 2013 1605 1531 1175 1024 955 893
i46 i47
1 627 514
> total <- T$sumO+T$sumI
> tot <- order(total,decreasing=TRUE)
> tot[1:10]
[1] 64 272 88 173 48 38 40 44 328 142
> total[tot[1:10]]
[1] 281996 203855 200629 196414 188394 170593 164378 163962 162828 161931
> T[tot[1:10],1:4]
v Name sumO sumI
64 64 W 41 St & 8 Ave 143342 138654
272 272 Nassau Ave & Russell St 101487 102368
88 88 W 20 St & 8 Ave 98557 102072
173 173 W 16 St & The High Line 96876 99538
48 48 W 22 St & 8 Ave 92946 95448
38 38 W 45 St & 8 Ave 84821 85772
40 40 W 38 St & 8 Ave 81611 82767
44 44 E 14 St & Avenue B 80797 83165
328 328 E 53 St & Madison Ave 79755 83073
142 142 W 53 St & 10 Ave 81074 80857
> n <- length(tot)
> T[tot[(n-10):n],1:4]
v Name sumO sumI
508 508 Verona Pl & Fulton St 0 2
556 556 E 95 St & 3 Ave 0 2
664 664 Amsterdam Ave & W 79 St 0 2
473 473 Leonard St & Meeker Ave 0 1
502 502 City Hall 0 1
506 506 SSP Tech Workshop 0 1
517 517 3 Ave & E 71 St 0 1
520 520 Adventures NYC 0 1
523 523 9 Ave & W 28 St 0 1
533 533 W 27 St & 10 Ave 0 1
661 661 Coffey St & Conover St 0 1
[1] "v" "Name" "sumO" "sumI" "o0" "o1" "o2" "o3" "o4" "o5" "o6" "o7" "o8" "o9" [15] "o10" "o11" "o12" "o13" "o14" "o15" "o16" "o17" "o18" "o19" "o20" "o21" "o22" "o23" [29] "o24" "o25" "o26" "o27" "o28" "o29" "o30" "o31" "o32" "o33" "o34" "o35" "o36" "o37" [43] "o38" "o39" "o40" "o41" "o42" "o43" "o44" "o45" "o46" "o47" "i0" "i1" "i2" "i3" [57] "i4" "i5" "i6" "i7" "i8" "i9" "i10" "i11" "i12" "i13" "i14" "i15" "i16" "i17" [71] "i18" "i19" "i20" "i21" "i22" "i23" "i24" "i25" "i26" "i27" "i28" "i29" "i30" "i31" [85] "i32" "i33" "i34" "i35" "i36" "i37" "i38" "i39" "i40" "i41" "i42" "i43" "i44" "i45" [99] "i46" "i47"
Stations with zero inflow or outflow
> m <- pmin(T$sumO,T$sumI)
> sum(m==0)
[1] 25
> T[m==0,1:4]
v Name sumO sumI
442 442 Park Ave & Marcus Garvey Blvd 0 325
448 448 48 Ave & 5 St 0 3
460 460 E 48 St & 5 Ave 0 10
467 467 Leonard St & Manhattan Ave 0 3
470 470 21 St & 41 Ave 0 3
473 473 Leonard St & Meeker Ave 0 1
474 474 Christ Hospital 0 11
480 480 Broadway & Roebling St 0 8
489 489 E 58 St & 1 Ave 0 6
492 492 University Pl & E 8 St 0 84
494 494 Rivington St & Ridge St 0 4
499 499 Montague St & Clinton St 0 2
501 501 Nassau Ave & Newell St 0 2
502 502 City Hall 0 1
506 506 SSP Tech Workshop 0 1
508 508 Verona Pl & Fulton St 0 2
515 515 Soissons Landing 0 3
517 517 3 Ave & E 71 St 0 1
520 520 Adventures NYC 0 1
523 523 9 Ave & W 28 St 0 1
533 533 W 27 St & 10 Ave 0 1
543 543 E 89 St & 3 Ave 0 3
556 556 E 95 St & 3 Ave 0 2
661 661 Coffey St & Conover St 0 1
664 664 Amsterdam Ave & W 79 St 0 2
>
Drawing distributions
>>> import sys, os, json, pickle >>> wdir = 'E:/data/bike/CitiBike/csv' >>> cdir = 'c:/users/batagelj/work/python/graph/chart' >>> os.chdir(wdir) >>> from datetime import * >>> import webbrowser >>> def TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit, fill='steelblue',xLab=70,yLab=40): TQ = [ list(q) for q in tq ] js = open(cdir+'/barData.js','w') js.write('var barData = '+str(TQ)+';\n') js.write('var TQmax = '+str(TQmax)+';\n') js.write('var Tmin = '+str(Tmin)+';\n') js.write('var Tmax = '+str(Tmax)+';\n') js.write('var Width = '+str(w)+';\n') js.write('var Height = '+str(h)+';\n') js.write('var Title = "'+tit+'";\n') js.write('var Rfill = "'+fill+'";\n') js.write('var xLab = "'+str(xLab)+'";\n') js.write('var yLab = "'+str(yLab)+'";\n') js.close() b = webbrowser.get('windows-default') b.open(cdir+'/TQchart.html') >>> ena = [ 1,"W 26 St & 10 Ave",542,388,215,130,124,94,59,56,50,68,65,195,752,874,988,1423,1910,2556,2257,1864,1478,1349,1276,1495,1481,1596,1515,1598,1648,1555,1576,1470,1525,1594,1985,2451,3362,3167,2757,2062,1634,1543,1424,1177,1153,1100,677,575,404,260,193,164,131,85,103,70,63,47,98,236,217,641,769,1039,1699,2306,2157,1739,1390,1288,1307,1525,1471,1524,1591,1678,1645,1687,1829,1864,2027,2644,3185,3364,3410,2926,2329,2013,1605,1531,1175,1024,955,893,627,514 ] >>> f = ena[2:50] >>> f[0] 542 >>> len(f) 48 >>> max(f) 3362 >>> TQmax = max(f) >>> Tmin = 0; Tmax = 48 >>> w = 600; h=180; tit = ena[1] >>> tq = [(i,i+1,v) for i,v in enumerate(f)] >>> tq [(0, 1, 542), (1, 2, 388), (2, 3, 215), (3, 4, 130), (4, 5, 124), (5, 6, 94), (6, 7, 59), (7, 8, 56), (8, 9, 50), (9, 10, 68), (10, 11, 65), (11, 12, 195), (12, 13, 752), (13, 14, 874), (14, 15, 988), (15, 16, 1423), (16, 17, 1910), (17, 18, 2556), (18, 19, 2257), (19, 20, 1864), (20, 21, 1478), (21, 22, 1349), (22, 23, 1276), (23, 24, 1495), (24, 25, 1481), (25, 26, 1596), (26, 27, 1515), (27, 28, 1598), (28, 29, 1648), (29, 30, 1555), (30, 31, 1576), (31, 32, 1470), (32, 33, 1525), (33, 34, 1594), (34, 35, 1985), (35, 36, 2451), (36, 37, 3362), (37, 38, 3167), (38, 39, 2757), (39, 40, 2062), (40, 41, 1634), (41, 42, 1543), (42, 43, 1424), (43, 44, 1177), (44, 45, 1153), (45, 46, 1100), (46, 47, 677), (47, 48, 575)] >>> TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='red') >>> f = ena[50:] >>> f[0] 404 >>> len(f) 48 >>> max(f) 3410 >>> tq = [(i,i+1,v) for i,v in enumerate(f)] >>> TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit)
Reading CSV SO file
import csv with open('./2015-16/SOsp.csv') as csvf: readCSV = csv.reader(csvf, delimiter=';') names = []; freqs = []; nodes = [] head = next(readCSV) for row in readCSV: node = row[1:3]; nodes.append(node) name = row[3:5]; names.append(name) freq = [ int(f) for f in row[5:53]] freqs.append(freq) longn = [ nu+' -> '+nv for nu,nv in names ] short = [ u+'->'+v for u,v in nodes ]
Saving clustering in R and reading it in Python
> clu <- file("cls7.clu","w")
> getwd()
[1] "E:/data/bike/CitiBike/csv/2015-16"
> cat(cls7,file=clu)
> close(clu)
>>> with open('./2015-16/cls7.clu', 'r') as clu: data = clu.read() >>> C = [ int(c) for c in data.split() ]
Computing distributions for clusters
nC = max(C) # number of clusters Cnt = [0]*nC for c in range(nC): Cnt[c] = [0]*48 for i in range(len(C)): c = C[i]-1 for j in range(48): Cnt[c][j] = Cnt[c][j]+freqs[i][j] Cs = [sum(v) for v in Cnt]
Drawing distributions for clusterings
W3 schools: Colors names.
# clusters of clustering C Tmin = 0; Tmax = 48; w = 900; h=250 color = ['darkorange', 'darkgreen', 'darkmagenta', 'darkslateblue', 'firebrick', 'darkcyan', 'darkorchid' ] for c in range(nC): tit = "Leader "+str(c+1); TQmax = max(Cnt[c]) tq = [(i,i+1,v) for i,v in enumerate(Cnt[c])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill=color[c]) nap = input("Naprej ?") # Top 6 flows Fs = [ (i,sum(v)) for i,v in enumerate(freqs) ] FS = sorted(Fs, key=lambda t: t[1]) top6 = FS[-6:]; c = 0 for i,s in top6: tit = "Top "+str(6-c)+': '+long[i]; TQmax = max(freqs[i]) tq = [(j,j+1,v) for j,v in enumerate(freqs[i])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill=color[c],xLab=20,yLab=10) nap = input("Naprej ?"); c = c+1
The most active stations / Top 10
import csv os.chdir('E:/data/bike/CitiBike/csv/2015-16') with open('SO2sp.csv') as csvf: readCSV = csv.reader(csvf, delimiter=';') names = []; freqs = []; nodes = [] head = next(readCSV) for row in readCSV: node = row[0]; nodes.append(node) name = row[1]; names.append(name) freq = [ int(f) for f in row[4:]] freqs.append(freq) short = [ 'S'+v for v in nodes ] >>> sio = [ (i,sum(v)) for i,v in enumerate(freqs) ] >>> SIO = sorted(sio, key=lambda t: t[1]) >>> top10 = SIO[-10:] >>> top10 [(141, 161931), (327, 162828), (43, 163962), (39, 164378), (37, 170593), (47, 188394), (172, 196414), (87, 200629), (271, 203855), (63, 281996)] >>> top10.reverse() >>> for i,s in top10: print(names[i],s) W 41 St & 8 Ave 281996 Nassau Ave & Russell St 203855 W 20 St & 8 Ave 200629 W 16 St & The High Line 196414 W 22 St & 8 Ave 188394 W 45 St & 8 Ave 170593 W 38 St & 8 Ave 164378 E 14 St & Avenue B 163962 E 53 St & Madison Ave 162828 W 53 St & 10 Ave 161931 >>> >>> top3 = SIO[-3:] c = 0 for i,s in top3: tit = "Top "+str(3-c)+': '+names[i]; TQmax = max(freqs[i]) tq = [(j,j+1,v) for j,v in enumerate(freqs[i][:48])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='red',xLab=20,yLab=10) nap = input(tit+" > Naprej ?") tq = [(j,j+1,v) for j,v in enumerate(freqs[i][48:])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='blue',xLab=20,yLab=10) nap = input("Naprej ?"); c = c+1
Sizes
>>> SO = [ sum(f[:48]) for f in freqs ] >>> SI = [ sum(f[48:]) for f in freqs ] >>> sum(SO) 13266296 >>> sum(SI) 13266296 >>> len(SI) 678
Imbalance out(v)-int(v)
>>> aBa = [ abs(SO[i]-SI[i]) for i in range(len(SI)) ] >>> aBa = [ (i,abs(SO[i]-SI[i])) for i in range(len(SI)) ] >>> ABA = sorted(aBa, key=lambda t: t[1]) >>> top10 = ABA[-10:] >>> top10 [(179, 4978), (354, 5726), (362, 6515), (325, 7403), (42, 7411), (62, 8132), (199, 8327), (142, 10465), (494, 15042), (392, 25965)] >>> top10.reverse() >>> for i,s in top10: print(names[i],SO[i],SI[i],SO[i]-SI[i]) 5 Ave & E 73 St 60524 34559 25965 Van Vorst Park 29962 14920 15042 8 Ave & W 33 St 57127 67592 -10465 W Broadway & Spring St 15217 23544 -8327 E 51 St & 1 Ave 72651 80783 -8132 E 75 St & 3 Ave 56302 48891 7411 Catherine St & Monroe St 36858 29455 7403 E 45 St & 3 Ave 48116 41601 6515 Water - Whitehall Plaza 71364 65638 5726 6 Ave & Canal St 23473 28451 -4978 >>> top4 = ABA[-4:] ---------------------------------------------------------------------- c = 0 for i,s in top4: tit = "Top "+str(4-c)+': '+names[i]; TQmax = max(freqs[i]) tq = [(j,j+1,v) for j,v in enumerate(freqs[i][:48])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='red',xLab=20,yLab=10) nap = input(tit+" > Naprej ?") tq = [(j,j+1,v) for j,v in enumerate(freqs[i][48:])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='blue',xLab=20,yLab=10) nap = input("Naprej ?"); c = c+1
Imbalance fDist
>>> nS = len(SI) >>> dBa = [ (k,sum(abs(freqs[k][:48][i]-freqs[k][48:][i]) for i in range(48))) for k in range(nS) ] >>> dBa[392] (392, 84703) >>> DBA = sorted(dBa, key=lambda t: t[1]) >>> top10 = DBA[-10:] >>> top10.reverse() >>> top10 [(392, 84703), (211, 66453), (42, 51297), (47, 50530), (117, 47893), (354, 45554), (62, 34086), (273, 33865), (261, 32562), (158, 30293)] >>> for i,s in top10: print(i,names[i],s) 392 5 Ave & E 73 St 84703 211 Fulton St & William St 66453 42 E 75 St & 3 Ave 51297 47 W 22 St & 8 Ave 50530 117 E 33 St & 2 Ave 47893 354 Water - Whitehall Plaza 45554 62 E 51 St & 1 Ave 34086 273 W 37 St & 10 Ave 33865 261 Cambridge Pl & Gates Ave 32562 158 E 16 St & Irving Pl 30293 >>> top4 = DBA[-4:] c = 0 for i,s in top4: tit = "Top "+str(4-c)+': '+names[i]; TQmax = max(freqs[i]) tq = [(j,j+1,v) for j,v in enumerate(freqs[i][:48])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='red',xLab=20,yLab=10) nap = input(tit+" > Naprej ?") tq = [(j,j+1,v) for j,v in enumerate(freqs[i][48:])] TQshow(tq,cdir,TQmax,Tmin,Tmax,w,h,tit,fill='blue',xLab=20,yLab=10) nap = input("Naprej ?"); c = c+1
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 3.0 Unported
