`altar.cuda.bayesian.cudaMetropolisVaryingSteps`

Module Contents

Classes

class altar.cuda.bayesian.cudaMetropolisVaryingSteps.cudaMetropolisVaryingSteps(name, locator, **kwds)

Bases: altar.component

The Metropolis algorithm as a sampler of the posterior distribution

scaling

doc = the parameter covariance Σ is scaled by the square of this

acceptanceWeight

doc = the weight of accepted samples during covariance rescaling

rejectionWeight

doc = the weight of rejected samples during covariance rescaling

max_mc_steps

doc = the maximum Monte-Carlo steps for one beta step

corr_check_steps

doc = the Monte-Carlo steps to compute the de

target_correlation

doc = the threshold of correlation to stop the chain

mcsteps = 1

dispatcher

ginit = False

gstep

gcandidate

gproposal

gsigma_chol

gvalid_indices

gvalid_samples

ginvalid_flags

gacceptance_flags

precision

gdice

curng

initialize(self, application): Initialize me and my parts given an {application} context

cuInitialize(self, application)

samplePosterior(self, annealer, step)

Sample the posterior distribution :param annealer - the controller: :param step - cpu CoolingStep:

Returns:: statistics (accepted/rejected/invalid) or (accepted/unlikely/rejected)

resample(self, annealer, statistics): Update my statistics based on the results of walking my Markov chains

prepareSamplingPDF(self, annealer, step): Re-scale and decompose the parameter covariance matrix, in preparation for the Metropolis update

finishSamplingPDF(self, step): procedures after sampling, e.g, copy data back to cpu

walkChains(self, annealer, step)

Run the Metropolis algorithm on the Markov chains :param annealer: cudaAnnealer :param step: cudaCoolingStep

Returns:: statistics = (accepted, rejected, unlikely)

displace(self, displacement): Construct a set of displacement vectors for the random walk from a distribution with zero mean and my covariance

adjustCovarianceScaling(self, accepted, invalid, rejected): Compute a new value for the covariance sacling factor based on the acceptance/rejection ratio

allocateGPUData(self, samples, parameters): initialize gpu work data