def getMeanStationVelocities1(stationTraces=None):
# get station velocities from a set of station positions.
# the previous way we did this was probaly pretty silly - which was just to average (x_i-x_{i-1})/dt.
# in this method: get positions; fit to a linear function: lat = lat0 + v_lat*t, lon = lon0 + v_lon*t
#
#if stationPositions==None: stationPositions=getStationPositions()
times=map(operator.itemgetter(0), stationTraces[1:])
#positions=[times]
returnVelocities=[]
lflon=yp.linefit()
lflat=yp.linefit()
for i in xrange(1, len(stationTraces[1])):
thispositions=map(operator.itemgetter(i), stationTraces)
thislats=map(operator.itemgetter(0), thispositions[1:])
thislons=map(operator.itemgetter(1), thispositions[1:])
thisStation = thispositions[0]
#print "lens: %d, %d, %d" % (len(times), len(thislats), len(thislons))
#
# now, fit x,y to lines.
lflon.datas=[times, thislons, scipy.zeros(len(thislons))+1]
lflat.datas=[times, thislats, scipy.zeros(len(thislats))+1]
#print lflon.datas
lonfits=lflon.doFit()
latfits=lflat.doFit()
#
#returnVelocities+=[[thisStation, lflon.a, lflat.a, lflon.b, lflat.b, lflon.varaprime, lflat.varaprime, lflon.varbprime, lflat.varbprime]]
# note that we have returned the a parameter as well. this can be used to infer the expected starting position (aka, we do not treat
# the first measurement as being special). obviously, use times[0] to get the starting position. be careful, since the raw lat/lon outputs
# (a values) will look "normal" since movement is slow and t0 is relatively recent.
#
# but let's go ahead and return the starting position, as per the data set provided.
lon0=lflon.a + lflon.b*times[0]
lat0=lflat.a + lflat.b*times[0]
returnVelocities+=[[thisStation, lon0, lat0, lflon.b, lflat.b, lflon.varaprime, lflat.varaprime, lflon.varbprime, lflat.varbprime]]
return returnVelocities
def plotMFIdistsRho(bclust=2.2, rhorange=[0.1, 1.0, 0.2], kmaxFP=10 ** 6, nMax=200, pim=4.9, fpmodel=1, fignums=[0, 1]):
# rhorange = [minrho, maxrho, dRho]
bofpim = []
for fignum in fignums:
plt.figure(fignum)
plt.clf()
plt.xlabel("$log_{10}(k)$", size=18)
plt.ylabel("$log_{10}(N(k))$", size=18)
#
primeRhos = [0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
thisrho = rhorange[0]
icount = 0
while thisrho <= rhorange[1]:
X = getMFIsquarePropModel(bclust, thisrho, pim, kmaxFP, nMax, fpmodel)
x = []
y1 = []
y2 = []
xstart = 25
for rw in X:
# print rw
x += [math.log10(rw[2])]
# print rw[3]
y1 += [math.log10(rw[3])]
y2 += [math.log10(rw[4])]
lf1 = yp.linefit([x[xstart:], y1[xstart:]])
lf2 = yp.linefit([x[xstart:], y2[xstart:]])
lf1.doFit()
lf2.doFit()
# lf1.doFit(None, None, 3)
# lf2.doFit(None, None, 3)
#
bofpim += [[pim, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
plt.figure(fignums[0])
plt.title("$\\beta=2.2$")
plt.plot(
x,
y1,
"%s.-" % yp.pycolor(icount),
label="$\\rho = %s, \\epsilon=%s, b=%s$" % (str(thisrho), str(pim), str(-lf1.b)),
lw=2,
) # beta!=2
plt.plot(
[x[0], x[xstart], x[-1]],
[lf1.a + lf1.b * x[0], lf1.a + lf1.b * x[xstart], lf1.a + lf1.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=15,
)
plt.figure(fignums[1])
plt.title("$\\beta=2.0$")
plt.plot(
x,
y2,
"%s.-" % yp.pycolor(icount),
label="$\\rho = %s, \\epsilon=%s, b=%s$" % (str(thisrho), str(pim), str(-lf2.b)),
lw=2,
) # beta=2 approximation
plt.plot(
[x[0], x[xstart], x[-1]],
[lf2.a + lf2.b * x[0], lf2.a + lf2.b * x[xstart], lf2.a + lf2.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=15,
)
icount += 1
#
thisrho += rhorange[2]
plt.show()
#
for fignum in fignums:
plt.figure(fignum)
plt.legend(loc="best")
return bofpim
def plotMFIdists2(
bclust=2.2, rho=0.59, kmaxFP=10 ** 6, nMax=100, pimrange=[0.0, 5.0], doClf=[True, True, False], fpmodel=0
):
# MFI dists with variable b(pim)
# for now, just approximate values of bclust:
bclust = [[0.0, 2.2], [1.0, 2.22], [2.0, 2.24], [3.0, 2.27], [4.0, 2.35], [4.6, 2.37]]
# and for now, we know we're going to do .2 intervals, so:
mybclusts = []
for i in xrange(len(bclust)):
# for rw in bclust:
rw = bclust[i]
mybclusts += [rw]
if mybclusts[-1] == bclust[-1]:
break
# print mybclusts[-1]
while mybclusts[-1][0] < bclust[i + 1][0]:
mybclusts += [[mybclusts[-1][0] + 0.2, mybclusts[-1][1]]]
# return mybclusts
#
pim = pimrange[0]
bofpim = []
plt.figure(0)
if doClf[0]:
plt.clf()
plt.figure(1)
if doClf[1]:
plt.clf()
i = 0
while pim <= pimrange[1]:
# X=getMFIsquarePropModel(bclust, rho, pim, kmaxFP, nMax)
X = getMFIsquarePropModel(mybclusts[i][1], rho, pim, kmaxFP, nMax, fpmodel)
x = []
y1 = []
y2 = []
for rw in X:
# print rw
x += [math.log(rw[2])]
y1 += [math.log(rw[3])]
y2 += [math.log(rw[4])]
lf1 = yp.linefit([x[1:], y1[1:]])
lf2 = yp.linefit([x[1:], y2[1:]])
lf1.doFit()
lf2.doFit()
#
bofpim += [[pim, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
plt.figure(0)
# plt.loglog(map(operator.itemgetter(2), X), map(operator.itemgetter(3), X), '.-')
plt.plot(x, y1, ".-", label="pim=%s, b=%s" % (str(pim), str(-lf1.b)))
plt.figure(1)
# plt.loglog(map(operator.itemgetter(2), X), map(operator.itemgetter(4), X), '.-')
plt.plot(x, y2, ".-", label="pim=%s, b=%s" % (str(pim), str(-lf2.b)))
#
pim += 0.2
i += 1
plt.figure(0)
plt.legend(loc="upper right")
plt.figure(1)
plt.legend(loc="upper right")
plt.figure(2)
plt.clf()
plt.plot(
map(operator.itemgetter(0), bofpim),
map(operator.itemgetter(2), bofpim),
".-",
label="$ beta =2.2, rho = %s $" % str(rho),
lw=2,
)
plt.plot(
map(operator.itemgetter(0), bofpim),
map(operator.itemgetter(5), bofpim),
".-",
label="$ beta =2.0, rho = %s $" % str(rho),
lw=2,
)
plt.xlabel("$p_{immune}$", size=16)
plt.ylabel("$b_{fires}$", size=16)
plt.legend(loc="best")
plt.show()
#
return bofpim
def plotMFIdistsBetaEps(bclust=2.2, kmaxFP=10 ** 9, nMax=200, pimrange=[0.0, 5.0, 1.0], beta=1.25, fignums=[0, 1]):
# def plotMFIdistsBetaEps(bclust=2.2, kmaxFP=10**9, nMax=200, pimlist=[0.0,1.0, 2.0, 3.0, 4.0, 4.9], beta=1.25, fignums=[0,1]):
# (this function is probably redundant. same as ...Beta(), but fix beta and range over pim.) we know we get nice PL
# when we vary the shape dimension. what about if we fix D (beta) and vary eps. with the fractal-footprint?
#
# fig 5 in PRE pub.
#
# pimrange=[0.0,4.9,1.]
# "beta" version, use footprint ~ n**beta 1<beta<2.0 (or maybe 2.25 or something).
# for now, fixed epsilon (pim), fixed rho.
# rhorange = [minrho, maxrho, dRho]
# note: bcluse is slope of b-distribution, beta is the exponent for shape area/raduis (n) scaling.
fsize = 20
thisfont = pltf.FontProperties(size=fsize)
thisfont2 = pltf.FontProperties(size=fsize - 2)
#
fpmodel = 2 # aka, MFI with fractal dimension treatment (as opposed to solid shapes, etc.)
rho = 1.0 # basically, this puts all the empty sites into the fractal/branching structure.
thisrho = rho
bofpim = []
for fignum in fignums:
plt.figure(fignum)
plt.clf()
plt.xlabel("$log_{10}(k)$", size=fsize + 2)
plt.ylabel("$log_{10}(N(k))$", size=fsize + 2)
#
# starts=[]
#
primeRhos = [0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
# thisrho=rhorange[0]
icount = 0
# while thisrho<=rhorange[1]:
# for beta in betas:
pim = pimrange[0]
pim0 = pim
while pim0 <= pimrange[1]:
# for pim in pimlist:
# X=getMFIsquarePropModel(bclust, thisrho, pim, kmaxFP, nMax, fpmodel)
pim = pim0
if pim > 4.9:
pim = 4.9
print "prams: %s, %s, %s, %s, %s, %s, %s" % (bclust, thisrho, pim, kmaxFP, nMax, fpmodel, beta)
# pim=pimrange[0]+float(icount)*pimrange[2]
X = getMFIsquarePropModel(bclust, thisrho, pim, kmaxFP, nMax, fpmodel, beta)
print "beta=%f, len=%d" % (beta, len(X))
x = []
y1 = []
y2 = []
# xstart=25
xstart = 15
sxtop = 500
for rw in X:
# print rw
x += [math.log10(rw[2])]
# print rw[3]
y1 += [math.log10(rw[3])]
y2 += [math.log10(rw[4])]
lf1 = yp.linefit([x[xstart:], y1[xstart:]])
lf2 = yp.linefit([x[xstart:], y2[xstart:]])
lf1.doFit()
lf2.doFit()
# lf1.doFit(None, None, 3)
# lf2.doFit(None, None, 3)
#
dotsies = yp.integerSpacedPoints([x[1:], y1[1:]], 0.5)
dotsies[0] += [x[-1]]
dotsies[1] += [y1[-1]]
if fpmodel > 1:
bofpim += [[pim, beta, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
else:
bofpim += [[pim, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
plt.figure(fignums[0])
# plt.title('$\\beta=2.2$')
# plt.plot(x[1:],y1[1:],'%s.-' % yp.pycolor(icount), label='$\\rho = %s, \\epsilon=%s, D=%s, b=%s$' % (str(thisrho), str(pim), str(beta), str(-lf1.b)[0:5]), lw=2) # beta!=2
plt.plot(
dotsies[0],
dotsies[1],
"%s%s" % (yp.pycolor(icount), yp.fitMarkerShortList[icount % len(yp.fitMarkerShortList)]),
label="$\\epsilon=%s, b=%s$" % (str(pim), str(-lf1.b)[0:5]),
ms=10,
)
plt.plot(x[1:], y1[1:], "%s-" % yp.pycolor(icount), lw=2, ms=10) # beta!=2
plt.plot(
[x[0], x[xstart], x[-1]],
[lf1.a + lf1.b * x[0], lf1.a + lf1.b * x[xstart], lf1.a + lf1.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=15,
)
plt.figure(fignums[1])
# plt.title('$\\beta=2.0$')
# plt.plot(x[1:], y2[1:], '%s.-' % yp.pycolor(icount), label='$\\rho = %s, \\epsilon=%s, D=%s, b=%s$' % (str(thisrho), str(pim), str(beta), str(-lf2.b)[0:5]), lw=2) # beta=2 approximation
plt.plot(
x[1:],
y2[1:],
"%s.-" % yp.pycolor(icount),
label="$\\epsilon=%s, b=%s$" % (str(pim), str(-lf2.b)[0:5]),
lw=2,
ms=10,
) # beta=2
plt.plot(
[x[0], x[xstart], x[-1]],
[lf2.a + lf2.b * x[0], lf2.a + lf2.b * x[xstart], lf2.a + lf2.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=15,
)
icount += 1
pim0 += pimrange[2]
#
# thisrho+=rhorange[2]
#
for fignum in fignums:
plt.figure(fignum)
ax = plt.gca()
plt.subplots_adjust(bottom=0.15)
plt.setp(ax.get_xticklabels(), fontsize=fsize)
plt.setp(ax.get_yticklabels(), fontsize=fsize)
# lgd=plt.legend(loc='best', numpoints=1, title='$D=1.25$', prop=thisfont2, ncol=1)
lgd = plt.legend(loc="best", numpoints=1, prop=thisfont2, ncol=1)
lgd.set_title("$D=1.25$")
plt.setp(lgd.get_title(), fontsize=fsize)
if fignum > 0:
continue
plt.savefig(
"writeup/mfi-aps/mfi-PRE-letter/figs/mfi-fractalType-varEps-bc%s-D125.png" % str(bclust).replace(".", "")
)
plt.savefig(
"writeup/mfi-aps/mfi-PRE-letter/figs/mfi-fractalType-varEps-bc%s-D125.eps" % str(bclust).replace(".", "")
)
plt.show()
return bofpim
def plotMFIdistsBeta(
bclust=2.2,
kmaxFP=10 ** 9,
nMax=200,
pim=4.9,
betas=[0.9, 1.0, 1.25, 1.5, 1.75, 2.0, 2.5],
fignums=[0, 1],
dosave=False,
):
# "Fig-4" in PRE publication.
#
# "beta" version, use footprint ~ n**beta 1<beta<2.0 (or maybe 2.25 or something).
# for now, fixed epsilon (pim), fixed rho.
# rhorange = [minrho, maxrho, dRho]
# note: bcluse is slope of b-distribution, beta is the exponent for shape area/raduis (n) scaling.
fpmodel = 2
rho = 1.0 # basically, this puts all the empty sites into the fractal/branching structure.
thisrho = rho
#
fsize = 20
thisfont = pltf.FontProperties(size=fsize)
thisfont2 = pltf.FontProperties(size=fsize - 2)
#
bofpim = []
for fignum in fignums:
plt.figure(fignum)
plt.clf()
plt.xlabel("$log_{10}(k)$", size=fsize + 2)
plt.ylabel("$log_{10}(N(k))$", size=fsize + 2)
#
# starts=[]
#
primeRhos = [0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
# thisrho=rhorange[0]
icount = 0
# while thisrho<=rhorange[1]:
fitMarkerShortList = ["o", "^", "s", "p", "*", "h", "+", "H", "D", "x"]
for beta in betas:
# X=getMFIsquarePropModel(bclust, thisrho, pim, kmaxFP, nMax, fpmodel)
print "prams: %s, %s, %s, %s, %s, %s, %s" % (bclust, thisrho, pim, kmaxFP, nMax, fpmodel, beta)
X = getMFIsquarePropModel(bclust, thisrho, pim, kmaxFP, nMax, fpmodel, beta)
print "beta=%f, len=%d" % (beta, len(X))
x = []
y1 = []
y2 = []
# xstart=25
xstart = 2
sxtop = 500
for rw in X:
# print rw
x += [math.log10(rw[2])]
# print rw[3]
y1 += [math.log10(rw[3])]
y2 += [math.log10(rw[4])]
lf1 = yp.linefit([x[xstart:], y1[xstart:]])
lf2 = yp.linefit([x[xstart:], y2[xstart:]])
lf1.doFit()
lf2.doFit()
# lf1.doFit(None, None, 3)
# lf2.doFit(None, None, 3)
#
if fpmodel > 1:
bofpim += [[pim, beta, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
else:
bofpim += [[pim, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
plt.figure(fignums[0])
# plt.title('$\\beta=2.2$')
# plt.plot(x[1:],y1[1:],'%s.-' % yp.pycolor(icount), label='$\\rho = %s, \\epsilon=%s, \\beta_s=%s, b=%s$' % (str(thisrho), str(pim), str(beta), str(-lf1.b)[0:5]), lw=2) # beta!=2
dotsies = yp.integerSpacedPoints([x[1:], y1[1:]], 0.5)
dotsies[0] += [x[-1]]
dotsies[1] += [y1[-1]] # add end-bits for completeness.
plt.plot(x[1:], y1[1:], "%s-" % yp.pycolor(icount), lw=2, ms=10) # beta!=2
plt.plot(
dotsies[0],
dotsies[1],
"%s%s" % (yp.pycolor(icount), fitMarkerShortList[icount % len(fitMarkerShortList)]),
label="$D=%s, b=%s$" % (str(beta), str(-lf1.b)[0:5]),
ms=10,
)
plt.plot(
[x[0], x[xstart], x[-1]],
[lf1.a + lf1.b * x[0], lf1.a + lf1.b * x[xstart], lf1.a + lf1.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=20,
lw=2,
)
#
plt.figure(fignums[1])
# plt.title('$\\beta=2.0$')
# plt.plot(x[1:], y2[1:], '%s.-' % yp.pycolor(icount), label='$\\rho = %s, \\epsilon=%s, \\beta_s=%s, b=%s$' % (str(thisrho), str(pim), str(beta), str(-lf2.b)[0:5]), lw=2) # beta=2 approximation
plt.plot(
x[1:],
y2[1:],
"%s.-" % yp.pycolor(icount),
label="$D=%s, b=%s$" % (str(beta), str(-lf2.b)[0:5]),
lw=2,
ms=10,
) # beta=2 approximation
plt.plot(
[x[0], x[xstart], x[-1]],
[lf2.a + lf2.b * x[0], lf2.a + lf2.b * x[xstart], lf2.a + lf2.b * x[-1]],
"%s|-" % yp.pycolor(icount),
ms=15,
lw=2,
)
icount += 1
#
# thisrho+=rhorange[2]
#
for fignum in fignums:
plt.figure(fignum)
ax = plt.gca()
plt.subplots_adjust(bottom=0.15)
plt.setp(ax.get_xticklabels(), fontsize=fsize)
plt.setp(ax.get_yticklabels(), fontsize=fsize)
# plt.legend(loc='best', numpoints=1, prop=thisfont2, ncol=1, title='$\\epsilon=4.9$')
lgd = plt.legend(loc="best", numpoints=1, prop=thisfont2, ncol=1)
lgd.set_title("$\\epsilon=4.9$")
plt.setp(lgd.get_title(), fontsize=fsize)
# mfi-fractalType-varEps-bc22-D125
if fignum > 0:
continue
if dosave:
plt.savefig(
"writeup/mfi-aps/mfi-PRE-letter/figs/mfi-fractalType-eps49-bc%s-varD.png" % str(bclust).replace(".", "")
)
plt.savefig(
"writeup/mfi-aps/mfi-PRE-letter/figs/mfi-fractalType-eps49-bc%s-varD.eps" % str(bclust).replace(".", "")
)
plt.show()
return bofpim
def plotMFIdists(
bclust=2.2, rho=0.59, kmaxFP=10 ** 6, nMax=200, pimrange=[0.0, 4.9], doClf=[True, True, False], fpmodel=1, lineNum=0
):
# pim=0.0
# lineNum is the index of the line on the plot when we plot multiple lines on a single canvas.
beta = 1.5
print "lineNum: %d" % lineNum
xstart = 15
pim = pimrange[0]
bofpim = []
plt.figure(0)
if doClf[0]:
plt.clf()
plt.figure(1)
if doClf[1]:
plt.clf()
plt.figure(3)
plt.clf()
plt.figure(4)
plt.clf()
# primePims=[0.0, .4, 1.0, 1.4, 2.0, 2.4, 3.0, 3.4, 4.0, 4.6, 4.8]
# primePims=[0, 4, 10, 14, 20, 24, 30, 34, 40, 46, 49]
primePims = [0, 10, 20, 30, 40, 48]
icount = -1
icount2 = 0
while pim <= (pimrange[1] + 0.0001):
print pim
icount += 1
X = getMFIsquarePropModel(bclust, rho, pim, kmaxFP, nMax, fpmodel, beta)
x = []
y1 = []
y2 = []
for rw in X:
# print rw
x += [math.log10(rw[2])]
# print rw[3]
y1 += [math.log10(rw[3])]
y2 += [math.log10(rw[4])]
lf1 = yp.linefit([x[xstart:], y1[xstart:]])
lf2 = yp.linefit([x[xstart:], y2[xstart:]])
lf1.doFit()
lf2.doFit()
# lf1.doFit(None, None, 3)
# lf2.doFit(None, None, 3)
#
bofpim += [[pim, lf1.a, -lf1.b, lf1.meanVar(), lf2.a, -lf2.b, lf2.meanVar()]]
a1 = bofpim[-1][1]
b1 = bofpim[-1][2]
a2 = bofpim[-1][4]
b2 = bofpim[-1][5]
plt.figure(0)
# plt.loglog(map(operator.itemgetter(2), X), map(operator.itemgetter(3), X), '.-')
plt.plot(x, y1, "%s.-" % yp.pycolor(icount), label="$\\epsilon=%s, b=%s$" % (str(pim), str(-lf1.b))) # beta!=2
# plt.plot([x[0], x[-1]], [a1-b1*x[0], a1-b1*x[-1]], '%s*-' % yp.pycolor(icount))
#
plt.figure(1)
# plt.loglog(map(operator.itemgetter(2), X), map(operator.itemgetter(4), X), '.-')
plt.plot(
x, y2, "%s.-" % yp.pycolor(icount), label="$\\epsilon=%s, b=%s$" % (str(pim), str(-lf2.b))
) # beta=2 approximation
# plt.plot([x[0], x[-1]], [a2-b2*x[0], a2-b2*x[-1]], '%s*-' % yp.pycolor(icount))
#
# plt.loglog(map(operator.itemgetter(2), X), map(operator.itemgetter(3), X), '.-')
if int(10.0 * pim) in primePims:
plt.figure(3)
plt.plot(
x,
y1,
"%s.-" % yp.pycolor(icount2),
label="$\\epsilon=%s, b=%s$" % (str(pim), str(-round(lf1.b, 2))),
lw=2,
)
plt.plot(
[x[0], x[xstart], x[-1]],
[a1 - b1 * x[0], a1 - b1 * x[xstart], a1 - b1 * x[-1]],
"%s*-" % yp.pycolor(icount2),
)
plt.figure(4)
plt.plot(
x,
y2,
"%s.-" % yp.pycolor(icount2),
label="$\\epsilon=%s, b=%s$" % (str(pim), str(-round(lf2.b, 2))),
lw=2,
)
plt.plot(
[x[0], x[xstart], x[-1]],
[a2 - b2 * x[0], a2 - b2 * x[xstart], a2 - b2 * x[-1]],
"%s*-" % yp.pycolor(icount2),
)
icount2 += 1
pim += 0.2
plt.figure(0)
plt.legend(loc="upper right")
plt.figure(1)
plt.legend(loc="upper right")
plt.figure(3)
plt.legend(loc="lower left")
plt.xlabel("$log(k)$", size=18)
plt.ylabel("$log(N(k))$", size=18)
plt.figure(4)
plt.legend(loc="lower left")
plt.xlabel("$log(k)$", size=18)
plt.ylabel("$log(N(k))$", size=18)
for i in xrange(5):
plt.figure(i)
plt.title("$\\rho=%f$" % rho)
plt.figure(2)
if doClf[2]:
plt.clf()
# if int(10.0*pim) in primePims:
plt.plot(
map(operator.itemgetter(0), bofpim),
map(operator.itemgetter(2), bofpim),
"%s.-" % yp.pycolor(lineNum),
label="$ \\beta = 2.2, \\rho = %s $" % str(rho),
lw=2,
)
plt.plot(
map(operator.itemgetter(0), bofpim),
map(operator.itemgetter(5), bofpim),
"%s>-" % yp.pycolor(lineNum),
label="$ \\beta = 2.0, \\rho = %s $" % str(rho),
lw=2,
)
plt.xlabel("$\\epsilon$", size=18)
plt.ylabel("$b_{fires}$", size=18)
plt.show()
#
return bofpim