def _OpenDetector(self,tteFile,eBound):
tteFile = self.directory+tteFile
lc = lightCurve('') #Give lc a fake parFile
lc.inFiles = [tteFile]
lc.ImportData()
lc.dt=self.dt
lc.tMax=self.tMax
lc.tMin=self.tMin
lc.eBins = eBound
lc.EnergyBinning()
lc.TimeBinning()
self.detList.append(lc)
def LoadData(self,lightCurveFile,sys=False):
'''
An EpEvofile is read in. It is JSON file containing the Ep and time
derived from either an SCAT or in the future, an mnSpecFit file
'''
self.lightcurve = lightCurve(lightCurveFile)
self.sys = sys
if sys:
self.stat = chi2_sys()
else:
self.stat = chi2() #Will always be using chi2
self._dataLoaded = True #Mark that data are loaded
def PlotEvo(self,fignum=1000):
fig = plt.figure(fignum)
ax = fig.add_subplot(111)
evo = lightCurve(self.lcFile)
self.linewidth = .8
contourColor = "#984ea3"
bfColor = "k"
wtColor = "#e41a1c"
pcColor = "#4daf4a"
breakColor = ["#377eb8","#ff7f00","#a65628","#377eb8","#f781bf"]
yData = []
for params in self.anal.get_equal_weighted_posterior()[::50,:-1]:
yData.append(self.model(self.dataRange, *params))
#Plot the spread in data
for y in yData:
ax.loglog(self.dataRange,y,contourColor,alpha=.07,lw=self.linewidth*.8,ls="-",zorder=-32) ## modify later
# Plot the best fit
bfModel=self.model(self.dataRange, *self.bestFit)
ax.loglog(self.dataRange,bfModel,bfColor,lw=self.linewidth,ls="-",zorder=-10) #modify later
ax.errorbar(evo.GetWTtime(),evo.GetWTflux(),yerr=evo.GetWTerr(),fmt='.',color=wtColor,capsize=self.capsize,elinewidth=self.elinewidth,markersize=3.3,alpha=.2)
ax.errorbar(evo.GetPCtime(),evo.GetPCflux(),yerr=evo.GetPCerr(),fmt='.',color=pcColor,capsize=self.capsize,elinewidth=self.elinewidth,markersize=3.3,alpha=.7)
ax.set_yscale('log',nonposy='clip')
ax.set_xlabel("Time [s]")
ax.set_ylabel(r"Flux")
#Plot the breaks if any
maxY =evo.GetFlux().max()
minY = evo.GetFlux().min()
maxY+=maxY*0.5
minY-=minY*0.5
if self.thisModel.plOrder>0:
cIdx=0
for val,err in zip(self.bestFit[-self.thisModel.plOrder:],self.anal.get_stats()["marginals"][-self.thisModel.plOrder:]):
sig = err['1sigma']
up=sig[1]
down=sig[0]
ax.vlines([10**val,10**up,10**down],1E-20,1E-1,colors=breakColor[cIdx],linestyles=['--',':',':'],zorder=-33)
ax.fill_betweenx([1E-20,1E-1],10**down,10**up,color=breakColor[cIdx],alpha=.1,zorder=-33)
cIdx+=1
ax.text(.7,.9 ,"GRB %s"%self.name ,transform=ax.transAxes)
# ax.text(.7,.8 ,"z: %.2f"%self.z ,transform=ax.transAxes)
ax.set_ylim(bottom=minY,top=maxY)
return ax
def _LoadData(self,data):
f = open(data)
fit = json.load(f)
self.modName = fit["model"]
self._composite = False
#Check to see if this is a composite model
test = self.modName.split("+")
if len(test)>1:
self._composite = True
compositeModels = test
for i in range(len(compositeModels)):
tmp = compositeModels[i].split('_')[0] #Hack off the duplicate model number tag
compositeModels[i] = tmp
self.parameters = fit["params"]
self.n_params = len(self.parameters)
self.lcFile = fit["lightcurve"]
self.basename = fit["basename"]
self.stat =fit["stat"]
self.dof = fit['dof']
self.dataRange=logspace(log10(fit["tmin"]),log10(fit["tmax"]),1000)
self.tmin = fit["tmin"]
self.tmax = fit["tmax"]
lc = lightCurve(self.lcFile)
self.name = lc.GetName()
self.z = lc.GetZ()
if self._composite:
thisModel = (models[compositeModels[0]])()
for mod in compositeModels[1:]:
tmp = (models[mod])()
thisModel = thisModel + tmp
self._componentLU = thisModel.componentLU
self._componentModel = thisModel
else:
thisModel = (models[fit["model"]])()
self.model = thisModel.model
#thisModel = models[self.modName]()
self.thisModel = thisModel
def PlotEvoFlares(self,fignum=1000):
if not self._composite:
print "This is not a composite model!"
return
fig = plt.figure(fignum)
ax = fig.add_subplot(111)
evo = lightCurve(self.lcFile)
self.linewidth = .8
contourColor = "#984ea3"
bfColor = "k"
flareColor = "grey"
wtColor = "#e41a1c"
pcColor = "#4daf4a"
#breakColor = ["#377eb8","#ff7f00","#a65628","#377eb8","#f781bf"]
components = self._componentLU.keys()
yData = []
leg = []
for comp in components:
thisComp = self._componentModel.SelectComponent(comp)
tt = self.GetParamIndex(thisComp["params"])
bfParams = self.bestFit[tt]
# Plot the best fit
yData=thisComp["model"](self.dataRange, *bfParams)
if thisComp["plOrder"] == -1:
color=flareColor
else:
color = bfColor
ax.loglog(self.dataRange, yData, color=color, lw=self.linewidth,ls="-",zorder=-10)
# Now plot the contours
yData = []
for params in self.anal.get_equal_weighted_posterior()[::50,:-1]:
params = params[tt]
yData.append( thisComp["model"](self.dataRange, *params) )
#Plot the spread in data
for y in yData:
ax.loglog(self.dataRange,y,contourColor,alpha=.07,lw=self.linewidth*.8,ls="-",zorder=-32) ## modify later
ax.errorbar(evo.GetWTtime(),evo.GetWTflux(),yerr=evo.GetWTerr(),fmt='.',color=wtColor,capsize=self.capsize,elinewidth=self.elinewidth,markersize=3.3,alpha=.2)
ax.errorbar(evo.GetPCtime(),evo.GetPCflux(),yerr=evo.GetPCerr(),fmt='.',color=pcColor,capsize=self.capsize,elinewidth=self.elinewidth,markersize=3.3,alpha=.7)
ax.set_yscale('log',nonposy='clip')
ax.set_xlabel("Time [s]")
ax.set_ylabel(r"Flux")
# #Plot the breaks if any
maxY =evo.GetFlux().max()
minY = evo.GetFlux().min()
maxY+=maxY*0.5
minY-=minY*0.5
# if self.thisModel.plOrder>0:
# cIdx=0
# for val,err in zip(self.bestFit[-self.thisModel.plOrder:],self.anal.get_stats()["marginals"][-self.thisModel.plOrder:]):
# sig = err['1sigma']
# up=sig[1]
# down=sig[0]
# ax.vlines([10**val,10**up,10**down],1E-20,1E-1,colors=breakColor[cIdx],linestyles=['--',':',':'],zorder=-33)
# ax.fill_betweenx([1E-20,1E-1],10**down,10**up,color=breakColor[cIdx],alpha=.1,zorder=-33)
# cIdx+=1
# ax.text(.7,.9 ,"GRB %s"%self.name ,transform=ax.transAxes)
# ax.text(.7,.8 ,"z: %.2f"%self.z ,transform=ax.transAxes)
ax.set_ylim(bottom=minY,top=maxY)
return ax
