pyhrf.jde.wsampler module

class pyhrf.jde.wsampler.WSampler(do_sampling=True, use_true_value=False, val_ini=None, pr_sigmoid_slope=1.0, pr_sigmoid_thresh=0.0)

Bases: pyhrf.xmlio.Initable, pyhrf.jde.samplerbase.GibbsSamplerVariable

CLASSES = array([0, 1])

CLASS_NAMES = ['inactiv', 'activ']

L_CA = 1

L_CI = 0

checkAndSetInitValue(variables)

computeProbW1(Qgj, gTQgj, rb, moyqj, t1, t2, mCAj, vCIj, vCAj, j, cardClassCAj)

ProbW1 is the probability that condition is relevant It is a vecteur on length nbcond

computeVarXhtQ(h, matXQ)

computemoyq(cardClassCA, nbVoxels)

Compute mean of labels in ROI

finalizeSampling()

getOutputs()

initObservables()

linkToData(dataInput)

sampleNextInternal(variables)

saveCurrentValue(it)

saveObservables(it)

threshold_W(meanW, thresh)

updateObsersables()

class pyhrf.jde.wsampler.W_Drift_Sampler(do_sampling=True, use_true_value=False, val_ini=None, pr_sigmoid_slope=1.0, pr_sigmoid_thresh=0.0)

Bases: pyhrf.xmlio.Initable, pyhrf.jde.samplerbase.GibbsSamplerVariable

CLASSES = array([0, 1])

CLASS_NAMES = ['inactiv', 'activ']

L_CA = 1

L_CI = 0

checkAndSetInitValue(variables)

computeProbW1(gj, gTgj, rb, t1, t2, mCAj, vCIj, vCAj, j, cardClassCAj)

ProbW1 is the probability that condition is relevant It is a vecteur on length nbcond

computemoyq(cardClassCA, nbVoxels)

Compute mean of labels in ROI

finalizeSampling()

getOutputs()

initObservables()

linkToData(dataInput)

sampleNextInternal(variables)

saveCurrentValue(it)

saveObservables(it)

threshold_W(meanW, thresh)

updateObsersables()