def run_carma_sampler(args):
pmodel = args[0]
qmodel = args[1]
time, y, ysig = args[2]
nsamples = 75000
carma_mcmc = cm.CarmaMCMC(time,
y,
ysig,
pmodel,
nsamples,
q=qmodel,
nburnin=25000)
carma = carma_mcmc.RunMCMC()
# remove C++ wrapper since it is not 'pickleable', as required by multiprocessing.Pool.map()
del carma.sampler
return carma
def testFull(self):
path = os.path.join(os.path.dirname(os.path.realpath(__file__)),
"../examples/car4_test.dat")
xv, yv, dyv = np.loadtxt(path, unpack=True)
nSample = 100
nBurnin = 10
nThin = 1
pModel = 3
qModel = 1
carma1 = carmcmc.CarmaMCMC(xv,
yv,
dyv,
1,
nSample,
q=0,
nburnin=nBurnin,
nthin=nThin)
post1 = carma1.RunMCMC()
carmap = carmcmc.CarmaMCMC(xv,
yv,
dyv,
pModel,
nSample,
q=0,
nburnin=nBurnin,
nthin=nThin)
postp = carmap.RunMCMC()
carmapqo = carmcmc.CarmaMCMC(xv,
yv,
dyv,
pModel,
nSample,
q=qModel,
nburnin=nBurnin,
nthin=nThin)
postpqo = carmapqo.RunMCMC()
carmapqe = carmcmc.CarmaMCMC(xv,
yv,
dyv,
pModel + 1,
nSample,
q=qModel,
nburnin=nBurnin,
nthin=nThin)
postpqe = carmapqe.RunMCMC()
# tests of yamcmcpp samplers.py
post1.effective_samples("sigma")
postp.effective_samples("ar_roots")
postpqo.effective_samples("ma_coefs")
postpqe.effective_samples("ar_coefs")
post1.plot_trace("sigma")
postp.plot_trace("sigma")
postpqo.plot_trace("sigma")
postpqe.plot_trace("sigma")
post1.plot_1dpdf("mu")
postp.plot_1dpdf("mu")
postpqo.plot_1dpdf("mu")
postpqe.plot_1dpdf("mu")
post1.plot_1dpdf("psd_centroid")
postp.plot_1dpdf("psd_centroid")
postpqo.plot_1dpdf("psd_width")
postpqe.plot_1dpdf("psd_width")
post1.plot_2dpdf("sigma", "var")
postp.plot_2dpdf("sigma", "var", pindex1=0, pindex2=0)
postpqo.plot_2dpdf("sigma", "var", pindex1=1, pindex2=1)
postpqe.plot_2dpdf("sigma", "var", pindex1=2, pindex2=2)
post1.plot_2dkde("sigma", "var")
postp.plot_2dkde("sigma", "var", pindex1=0, pindex2=0)
postpqo.plot_2dkde("sigma", "var", pindex1=1, pindex2=1)
postpqe.plot_2dkde("sigma", "var", pindex1=2, pindex2=2)
post1.plot_autocorr("psd_centroid")
postp.plot_autocorr("psd_centroid")
postpqo.plot_autocorr("psd_centroid")
postpqe.plot_autocorr("psd_centroid")
post1.plot_parameter("psd_centroid")
postp.plot_parameter("psd_centroid")
postpqo.plot_parameter("psd_centroid")
postpqe.plot_parameter("psd_centroid")
post1.posterior_summaries("psd_width")
postp.posterior_summaries("psd_width")
postpqo.posterior_summaries("psd_width")
postpqe.posterior_summaries("psd_width")
post1.posterior_summaries("sigma")
postp.posterior_summaries("sigma")
postpqo.posterior_summaries("sigma")
postpqe.posterior_summaries("sigma")
# tests of carma_pack carma_pack.py
post1.plot_power_spectrum(percentile=95.0, doShow=False)
postp.plot_power_spectrum(percentile=95.0, doShow=False)
postpqo.plot_power_spectrum(percentile=95.0, doShow=False)
postpqe.plot_power_spectrum(percentile=95.0, doShow=False)
post1.plot_models(nplot=1000, doShow=False)
postp.plot_models(nplot=1000, doShow=False)
postpqo.plot_models(nplot=1000, doShow=False)
postpqe.plot_models(nplot=1000, doShow=False)
for bestfit in ["map", "median", "mean"]:
post1.assess_fit(nplot=1000, bestfit=bestfit, doShow=False)
postp.assess_fit(nplot=1000, bestfit=bestfit, doShow=False)
postpqo.assess_fit(nplot=1000, bestfit=bestfit, doShow=False)
postpqe.assess_fit(nplot=1000, bestfit=bestfit, doShow=False)
import numpy as np
import matplotlib.pyplot as plt
infile = sys.argv[1]
xv, yv, dyv = np.loadtxt(sys.argv[1], unpack=True)
nSample = 100
nBurnin = 10
nThin = 1
pModel = 3
qModel = 1
carma1 = carmcmc.CarmaMCMC(xv,
yv,
dyv,
1,
nSample,
q=0,
nburnin=nBurnin,
nthin=nThin)
post1 = carma1.RunMCMC()
carmap = carmcmc.CarmaMCMC(xv,
yv,
dyv,
pModel,
nSample,
q=0,
nburnin=nBurnin,
nthin=nThin)
postp = carmap.RunMCMC()
import sys, os
import yamcmcpp
import carmcmc
import numpy as np
import matplotlib.pyplot as plt
infile = sys.argv[1]
xv, yv, dyv = np.loadtxt(sys.argv[1], unpack=True)
nSample = 500
nBurnin = 50
nThin = 1
pModel = 4
qModel = 1
# This seems to fail when pModel is odd and qModel = 1, and works when pModel is even and qModel=1
carma_model = carmcmc.CarmaMCMC(xv,
yv,
dyv,
pModel,
nSample,
q=qModel,
nburnin=nBurnin,
nthin=nThin)
post = carma_model.RunMCMC()
# Set up a loop over all filters, and several CARMA p-orders (we set q=0 here)
posteriors = {}
for f in filters:
posteriors[f] = {}
idx = np.where(data["filter"] == f)
tdb = data["TDB"][idx]
mag = data["mag"][idx]
dmag = data["dmag"][idx]
for p in xrange(1, 10):
carma = carmcmc.CarmaMCMC(tdb,
mag,
dmag,
p,
NSAMPLE,
q=0,
nburnin=NBURNIN,
nthin=NTHIN)
post = carma.RunMCMC()
posteriors[f][p] = post
# Print out the DIC for all models
for f in filters:
print "Deviance Information Criterion:"
for p in xrange(1, 10):
print f, p, posteriors[f][p].DIC()[0]
# Print out several useful diagnostic figures
for f in filters:
for p in xrange(1, 10):