PyHRF package¶

pyhrf.boldsynth.pottsfield package¶

pyhrf.boldsynth.field module¶

pyhrf.boldsynth.field.count_homo_cliques(graph, labels, weights=None)¶

pyhrf.boldsynth.field.genPepperSaltField(size, nbLabels, initProps=None)¶

pyhrf.boldsynth.field.genPotts(graph, beta, nbLabels=2, labelsIni=None, method='SW', weights=None)¶

Simulate a realisation of a Potts Field with spatial correlation amount ‘beta’. ‘graph’ is list of lists, ie a neighbors for each node index. ‘nbLabels’ is the number of labels ‘method’ can be either ‘SW’ (swensdsen-wang) or ‘gibbs’

pyhrf.boldsynth.field.genPottsMap(mask, beta, nbLabels, method='SW')¶

pyhrf.boldsynth.field.potts_generator(**args)¶

class pyhrf.boldsynth.field.random_field_generator(size, nbClasses)¶

pyhrf.boldsynth.hrf module¶

pyhrf.boldsynth.hrf.bezierCurve(p1, pc1, p2, pc2, xPrecision)¶

pyhrf.boldsynth.hrf.buildFiniteDiffMatrix(order, size)¶

pyhrf.boldsynth.hrf.genBezierHRF(timeAxis=array([ 0. , 0.6, 1.2, 1.8, 2.4, 3. , 3.6, 4.2, 4.8, 5.4, 6. , 6.6, 7.2, 7.8, 8.4, 9. , 9.6, 10.2, 10.8, 11.4, 12. , 12.6, 13.2, 13.8, 14.4, 15. , 15.6, 16.2, 16.8, 17.4, 18. , 18.6, 19.2, 19.8, 20.4, 21. , 21.6, 22.2, 22.8, 23.4, 24. , 24.6, 25.2]), pic=[6, 1], picw=2, ushoot=[15, -0.2], ushootw=3, normalize=False)¶

pyhrf.boldsynth.hrf.genCanoBezierHRF(duration=25.0, dt=0.6, normalize=False)¶

pyhrf.boldsynth.hrf.genExpHRF(timeAxis=array([ 0., 0.5, 1., 1.5, 2., 2.5, 3., 3.5, 4., 4.5, 5., 5.5, 6., 6.5, 7., 7.5, 8., 8.5, 9., 9.5, 10., 10.5, 11., 11.5, 12., 12.5, 13., 13.5, 14., 14.5, 15., 15.5, 16., 16.5, 17., 17.5, 18., 18.5, 19., 19.5, 20., 20.5, 21., 21.5, 22., 22.5, 23., 23.5, 24., 24.5]), ttp=6, pa=1, pw=0.2, ttu=11, ua=0.2, uw=0.01)¶

pyhrf.boldsynth.hrf.genGaussianSmoothHRF(zc, length, eventdt, rh, order=2)¶

pyhrf.boldsynth.hrf.genPriorCov(zc, pprcov, dt)¶

pyhrf.boldsynth.hrf.getCanoHRF(duration=25, dt=0.6, hrf_from_spm=True, delay_of_response=6.0, delay_of_undershoot=16.0, dispersion_of_response=1.0, dispersion_of_undershoot=1.0, ratio_resp_under=6.0, delay=0.0)¶

Compute the canonical HRF.

Parameters:

duration (int or float, optional) – time lenght of the HRF in seconds dt (float, optional) – time resolution of the HRF in seconds hrf_from_spm (bool, optional) – if True, use the SPM formula to compute the HRF, if False, use the hard coded values and resample if necessary. It is strongly advised to use True. delay_of_response (float, optional) – delay of the first peak response in seconds delay_of_undershoot (float, optional) – delay of the second undershoot peak in seconds dispersion_of_response (float, optional) – dispersion_of_undershoot (float, optional) – ratio_resp_under (float, optional) – ratio between the response peak and the undershoot peak delay (float, optional) – delay of the HRF

Returns:

time_axis (ndarray, shape (round(duration/dt)+1,)) – time axis of the HRF in seconds HRF (ndarray, shape (round(duration/dt)+1,))

pyhrf.boldsynth.hrf.getCanoHRF_tderivative(duration=25.0, dt=0.5)¶

pyhrf.boldsynth.scenarios module¶

pyhrf.boldsynth.scenarios.build_ctrl_tag_matrix(asl_shape)¶

pyhrf.boldsynth.scenarios.calc_asl_shape(bold_stim_induced, dsf)¶

pyhrf.boldsynth.scenarios.createBiGaussCovarNRL(condition_defs, labels, covariance)¶

pyhrf.boldsynth.scenarios.create_3Dlabels_Potts(condition_defs, beta, dims, mask)¶

pyhrf.boldsynth.scenarios.create_AR_noise(bold_shape, v_noise, order=2, v_corr=0.1)¶

pyhrf.boldsynth.scenarios.create_Xh(nrls, rastered_paradigm, hrf, condition_defs, dt, hrf_territories=None)¶

Retrieve the product X.h

pyhrf.boldsynth.scenarios.create_alpha_for_hrfgroup(alpha_var)¶

Create alpha from a normal distribution, for one subject

pyhrf.boldsynth.scenarios.create_asl_from_stim_induced(bold_stim_induced, perf_stim_induced, ctrl_tag_mat, dsf, perf_baseline, noise, drift=None, outliers=None)¶

Downsample stim_induced signal according to downsampling factor ‘dsf’ and add noise and drift (nuisance signals) which has to be at downsampled temporal resolution.

pyhrf.boldsynth.scenarios.create_bigaussian_nrls(labels, mean_act, var_act, var_inact)¶

Simulate bi-Gaussian NRLs (zero-centered inactive component)

pyhrf.boldsynth.scenarios.create_bold(stim_induced_signal, dsf, noise, drift=None, outliers=None)¶

a short-cut for function create_bold_from_stim_induced

pyhrf.boldsynth.scenarios.create_bold_controlled_variance(stim_induced_signal, alpha, nb_voxels, dsf, nrls, Xh, drift=None, outliers=None)¶

Create BOLD with controlled explained variance alpha: percentage of explained variance on total variance

pyhrf.boldsynth.scenarios.create_bold_from_stim_induced(stim_induced_signal, dsf, noise, drift=None, outliers=None)¶

Downsample stim_induced signal according to downsampling factor ‘dsf’ and add noise and drift (nuisance signals) which has to be at downsampled temporal resolution.

pyhrf.boldsynth.scenarios.create_bold_from_stim_induced_RealNoise(stim_induced_signal, dsf, noise, drift)¶

Downsample stim_induced signal according to downsampling factor ‘dsf’ and add noise and drift (nuisance signals) which has to be at downsampled temporal resolution.

pyhrf.boldsynth.scenarios.create_bold_stim_induced_signal(brls, rastered_paradigm, brf, condition_defs, dt, hrf_territories=None)¶

Create a stimulus induced signal for ASL from BOLD response levels, paradigm and BRF (sum_{m=1}^M a^m X^m h + sum_{m=1}^M c^m W X^m g) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given BRF and multiply by brls. Finally compute the sum over conditions.

Return a asl array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_canonical_hrf(hrf_duration=25.0, dt=0.5)¶

pyhrf.boldsynth.scenarios.create_connected_label_clusters(condition_defs, activ_label_graph)¶

pyhrf.boldsynth.scenarios.create_drift_coeffs(bold_shape, drift_order, drift_coeff_var)¶

pyhrf.boldsynth.scenarios.create_drift_coeffs_asl(asl_shape, drift_order, drift_var)¶

pyhrf.boldsynth.scenarios.create_gaussian_hrf_subject(hrf_group, var_subject_hrf, dt, alpha=0.0)¶

Creation of hrf by subject. Use group level hrf and variance for each subject (var_subjects_hrfs must be a list) Simulated hrfs must be smooth enough: correlation between temporal coeffcients

pyhrf.boldsynth.scenarios.create_gaussian_noise(bold_shape, v_noise, m_noise=0.0)¶

pyhrf.boldsynth.scenarios.create_gaussian_noise_asl(asl_shape, v_gnoise, m_noise=0.0)¶

pyhrf.boldsynth.scenarios.create_gaussian_nrls_sessions_and_mean(nrls, condition_defs, labels, var_sess)¶

Creation of nrls by session (and by voxel and cond) - for one session n° sess The nrls by session vary around an nrl mean (nrls_bar) defined by voxel and cond (var_sess corresponds to the variation of session defined nrls around the nrl_bar) Here “nrls” is nrls_bar, mean over subjects!

pyhrf.boldsynth.scenarios.create_gsmooth_hrf(hrf_duration=25.0, dt=0.5, order=2, hrf_var=1.0, zc=True, normalize_hrf=True)¶

Create a smooth HRF according to the multivariate gaussian prior used in JDE hrf_duration and dt are the HRF duration and temporal resolution, respectively (in sec.). order is derivative order constraining the covariance matrix. hrf_var is the HRF variance. zc is a flag to impose zeros at the begining and the end of the HRF

return: a np array of HRF coefficients

pyhrf.boldsynth.scenarios.create_hrf(picw, pic, under=2, hrf_duration=25.0, dt=0.5)¶

pyhrf.boldsynth.scenarios.create_hrf_from_territories(hrf_territories, primary_hrfs)¶

pyhrf.boldsynth.scenarios.create_labels_Potts(condition_defs, beta, nb_voxels)¶

pyhrf.boldsynth.scenarios.create_labels_vol(condition_defs)¶

Create a seet labels from the field “label_map” in condition_defs Available choices for the field label_map: - ‘random_small’ : binary labels are randomly generated with shape (1,5,5) - a tag (str) : corresponds to a png file in pyhrf data files - a 3D np containing the labels

pyhrf.boldsynth.scenarios.create_language_paradigm(condition_defs)¶

pyhrf.boldsynth.scenarios.create_localizer_paradigm(condition_defs, paradigm_label='av')¶

pyhrf.boldsynth.scenarios.create_localizer_paradigm_a(condition_defs)¶

pyhrf.boldsynth.scenarios.create_localizer_paradigm_avd(condition_defs)¶

pyhrf.boldsynth.scenarios.create_multisess_stim_induced_signal(nrls_session, rastered_paradigm, hrf, condition_defs, dt, hrf_territories=None)¶

Create a stimulus induced signal from neural response levels, paradigm and HRF (sum_{m=1}^M a^m X^m h) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given HRF and multiply by nrls. Finally compute the sum over conditions.

Return a bold array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_multisess_stim_induced_signal_asl(prls_session, rastered_paradigm, prf, condition_defs, dt, hrf_territories=None)¶

Create a stimulus induced signal from neural response levels, paradigm and HRF (sum_{m=1}^M a^m X^m h) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given HRF and multiply by nrls. Finally compute the sum over conditions.

Return a bold array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_null_drift(bold_shape)¶

pyhrf.boldsynth.scenarios.create_outliers(bold_shape, stim_induced_signal, nb_outliers, outlier_scale=5.0)¶

pyhrf.boldsynth.scenarios.create_paradigm_un_evnt(condition_defs)¶

pyhrf.boldsynth.scenarios.create_perf_baseline(asl_shape, perf_baseline_var, perf_baseline_mean=0.0)¶

pyhrf.boldsynth.scenarios.create_perf_stim_induced_signal(prls, rastered_paradigm, prf, condition_defs, dt, hrf_territories=None)¶

Create a stimulus induced signal for ASL from perfusion response levels, paradigm and PRF (sum_{m=1}^M c^m X^m g) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given PRF and multiply by prls. Finally compute the sum over conditions.

Return a asl array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_polynomial_drift(bold_shape, tr, drift_order, drift_var)¶

pyhrf.boldsynth.scenarios.create_polynomial_drift_from_coeffs(bold_shape, tr, drift_order, drift_coeffs, drift_mean=0.0, drift_amplitude=1.0)¶

pyhrf.boldsynth.scenarios.create_polynomial_drift_from_coeffs_asl(asl_shape, tr, drift_order, drift_coeffs)¶

pyhrf.boldsynth.scenarios.create_prf(prf_duration=25.0, dt=0.5)¶

pyhrf.boldsynth.scenarios.create_small_bold_simulation(snr='high', output_dir=None, simu_items=None)¶

pyhrf.boldsynth.scenarios.create_stim_induced_signal(nrls, rastered_paradigm, hrf, dt)¶

Create a stimulus induced signal from neural response levels, paradigm and HRF (sum_{m=1}^M a^m X^m h) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given HRF and multiply by nrls. Finally compute the sum over conditions.

Return a bold array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_stim_induced_signal_Parsi(nrls, rastered_paradigm, hrf, condition_defs, dt, w)¶

Create a stimulus induced signal from neural response levels, paradigm and HRF (sum_{m=1}^M a^m w^m X^m h) For each condition, compute the convolution of the paradigm binary sequence ‘rastered_paradigm’ with the given HRF and multiply by nrls and W. Finally compute the sum over conditions.

Return a bold array of shape (nb scans, nb voxels)

pyhrf.boldsynth.scenarios.create_time_invariant_gaussian_brls(condition_defs, labels)¶

BOLD response levels for ASL

pyhrf.boldsynth.scenarios.create_time_invariant_gaussian_nrls(condition_defs, labels)¶

pyhrf.boldsynth.scenarios.create_time_invariant_gaussian_prls(condition_defs, labels)¶

Perfusion response levels for ASL

pyhrf.boldsynth.scenarios.create_varying_hrf(hrf_duration=25.0, dt=0.5)¶

pyhrf.boldsynth.scenarios.duplicate_brf(nb_voxels, primary_brf)¶

Duplicate brf over all voxels. Return an array of shape (nb_voxels, len(brf))

pyhrf.boldsynth.scenarios.duplicate_hrf(nb_voxels, primary_hrf)¶

Duplicate hrf over all voxels. Return an array of shape (nb_voxels, len(hrf))

pyhrf.boldsynth.scenarios.duplicate_noise_var(nb_voxels, v_gnoise)¶

Duplicate variance of noise over all voxels. Return an array of shape (nb_voxels, var noise)

pyhrf.boldsynth.scenarios.duplicate_prf(nb_voxels, primary_prf)¶

Duplicate prf over all voxels. Return an array of shape (nb_voxels, len(prf))

pyhrf.boldsynth.scenarios.flatten_labels_vol(labels_vol)¶

pyhrf.boldsynth.scenarios.get_bold_shape(stim_induced_signal, dsf)¶

pyhrf.boldsynth.scenarios.load_drawn_labels(name)¶

pyhrf.boldsynth.scenarios.load_hrf_territories(nb_hrf_territories=0, hrf_territories_name=None)¶

pyhrf.boldsynth.scenarios.load_many_hrf_territories(nb_hrf_territories)¶

pyhrf.boldsynth.scenarios.randn(d0, d1, ..., dn)¶

Return a sample (or samples) from the “standard normal” distribution.

If positive, int_like or int-convertible arguments are provided, randn generates an array of shape (d0, d1, ..., dn), filled with random floats sampled from a univariate “normal” (Gaussian) distribution of mean 0 and variance 1 (if any of the are floats, they are first converted to integers by truncation). A single float randomly sampled from the distribution is returned if no argument is provided.

This is a convenience function. If you want an interface that takes a tuple as the first argument, use numpy.random.standard_normal instead.

Parameters:	d1, .., dn (d0,) – The dimensions of the returned array, should be all positive. If no argument is given a single Python float is returned.
Returns:	Z – A (d0, d1, ..., dn)-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied.
Return type:	ndarray or float

See also

random.standard_normal()

Similar, but takes a tuple as its argument.

Notes

For random samples from , use:

sigma * np.random.randn(...) + mu

Examples

>>> np.random.randn()
2.1923875335537315 #random

Two-by-four array of samples from N(3, 6.25):

>>> 2.5 * np.random.randn(2, 4) + 3
array([[-4.49401501,  4.00950034, -1.81814867,  7.29718677],  #random
       [ 0.39924804,  4.68456316,  4.99394529,  4.84057254]]) #random

pyhrf.boldsynth.scenarios.rasterize_paradigm(paradigm, dt, condition_defs)¶

Return binary sequences of onsets approximated on temporal grid of temporal resolution dt, for all conditions. ‘paradigm’ is expected to be an instance of ‘pyhrf.paradigm.mpar.Paradigm’

pyhrf.boldsynth.scenarios.save_simulation(simulation, output_dir)¶

short-hand for simulation_save_vol_outputs

pyhrf.boldsynth.scenarios.simulation_save_vol_outputs(simulation, output_dir, bold_3D_vols_dir=None, simulation_graph_output=None, prefix=None, vol_meta=None)¶

simulation_graph_output : None, ‘simple’, ‘thumbnails’ #TODO

pyhrf.boldsynth.spatialconfig module¶

class pyhrf.boldsynth.spatialconfig.Mapper1D(mapping, expandedShape)¶

Handles a mapping between a nD coordinate space (expanded) and a 1D coordinate space (flatten). Can be applied to numpy.ndarray objects

createExpandedArray(flatShape, type, mappedAxis=0, fillValue=0)¶

expandArray(a, mappedAxis=0, dest=None, fillValue=0)¶

Expand dimensions of ‘a’ following predefined mapping. ‘mappedAxis’ is the axis index to be expanded in ‘a’. If dest is not None the map values from ‘a’ to dest. If dest is None then return a new array and fill positions not involved in mapping with ‘fillValue’.

flattenArray(array, firstMappedAxis=0)¶

Reduce dimensions of ‘array’. ‘firstMappedAxis’ is index of the axis to be reduced (other mapped axes are assumed to follow this one).

class pyhrf.boldsynth.spatialconfig.NeighbourhoodSystem(neighboursSets)¶

static fromLattice(kerMask=None, depth=1, torusFlag=False)¶

Creates a NeighbourhoodSystem instance from a n-dimensional lattice

static fromMesh()¶

getMaxIndex()¶

getMaxNeighbours()¶

getNeighbours(nodeId)¶

getNeighboursArrays()¶

getNeighboursLists()¶

getNeighboursSets()¶

kerMask2D_4n = array([[-1, 0], [ 1, 0], [ 0, 1], [ 0, -1]])¶

kerMask3D_6n = array([[ 1, 0, 0], [ 0, 1, 0], [ 0, 0, 1], [-1, 0, 0], [ 0, -1, 0], [ 0, 0, -1]])¶

sub(nodeIds)¶

class pyhrf.boldsynth.spatialconfig.PottsField(classNames, spConf, initProps=None)¶

Bases: pyhrf.boldsynth.spatialconfig.StateField

class pyhrf.boldsynth.spatialconfig.RegularLatticeMapping(shape=None, mapping=None, order=1, depth=1)¶

Bases: pyhrf.boldsynth.spatialconfig.SpatialMapping

Define a SpatialMapping on a 3D regular lattice.

buildNeighboursCoordLists()¶

buildNeighboursIndexLists()¶

c = [1, 1, 1]¶

static createFromGUI()¶

Creates the actual object based on the parameters

getClosestNeighboursIndexes(idVoxel)¶

getCoord(index)¶

getIndex(coord)¶

getMapping()¶

getNbCliques()¶

getNbVoxels()¶

getNdArrayMask()¶

getNeighboursCoordLists()¶

getNeighboursCoords(idvoxel)¶

getNeighboursIndexLists()¶

getNeighboursIndexes(idVoxel)¶

getRoiMask()¶

Return a binary or n-ary 3D mask which has the shape of the target data

getTargetAxesNames()¶

mapVoxData(data, fillValue=0)¶

nbNeighboursOrder1 = 6¶

nbNeighboursOrder2 = 26¶

order1Mask = array([[ 1, 0, 0], [ 0, 1, 0], [ 0, 0, 1], [-1, 0, 0], [ 0, -1, 0], [ 0, 0, -1]])¶

order2Mask = array([[ 0, 0, -1], [ 0, 0, 1], [ 0, -1, 0], [ 0, -1, -1], [ 0, -1, 1], [ 0, 1, 0], [ 0, 1, -1], [ 0, 1, 1], [-1, 0, 0], [-1, 0, -1], [-1, 0, 1], [-1, -1, 0], [-1, -1, -1], [-1, -1, 1], [-1, 1, 0], [-1, 1, -1], [-1, 1, 1], [ 1, 0, 0], [ 1, 0, -1], [ 1, 0, 1], [ 1, -1, 0], [ 1, -1, -1], [ 1, -1, 1], [ 1, 1, 0], [ 1, 1, -1], [ 1, 1, 1]])¶

class pyhrf.boldsynth.spatialconfig.RegularLatticeMapping2(maskLattice, kerMask=None, nsDepth=1, parentMapping=None, torusFlag=False)¶

Bases: pyhrf.boldsynth.spatialconfig.SpatialMapping2

flattenData(data, firstMappedAxis=0)¶

mapData(data, mappedAxis=0, fillValue=0)¶

class pyhrf.boldsynth.spatialconfig.SpatialMapping¶

Interface specification for the handling of a mapping between integer indexes and positions in a 3D space.

getCoord()¶

Return coord mapped with ‘index’

getIndex()¶

Return index mapped with ‘coord’

getMapping()¶

Return a mapping object (list or dict) which maps an integer index to its 3D coordinates.

getNbVoxels()¶

Return the total number of mapped position

getNdArrayMask()¶

Return the set of mapped 3D coordinates in a tuple usable as a mask for numpy.ndarray

getNeighboursCoordLists()¶

Get lists of neighbours for all positions @param idVoxel: index of the voxel @return: a mapping object (list or dict) which maps each integer index to a list of 3D coordinates (the neighbours).

getNeighboursCoords(idVoxel)¶

@param idVoxel: index of the voxel @return: the list of 3D coordinates corresponding to the neighbours of the specified voxel.

getNeighboursIndexLists()¶

Get lists of neighbours for all positions @param idVoxel: index of the voxel @return: a mapping object (list or dict) which maps each integer index to a list of integer indexes (the neighbours).

getNeighboursIndexes(idVoxel)¶

@param idVoxel: index of the voxel @return: the list of integer indexes corresponding to the neighbours of the specified voxel.

getRoiMask()¶

Return a binary or n-ary mask which has the shape of the target data

class pyhrf.boldsynth.spatialconfig.SpatialMapping2(positions, ns, parentMapping=None, parentIndex=None)¶

static fromLattice(kerMask=None, nsDepth=1, torusFlag=False)¶

static fromMesh(positions)¶

getPositions()¶

sub(nodeIds)¶

class pyhrf.boldsynth.spatialconfig.StateField(classNames, spConf, initProps=None)¶

Class handling a field of states : a set of integers (ie labels) whose ranks can be spatially mapped to 3D coordinates. Each label refers to a class wich is identified by an ID and a name.

generate()¶

Generate values for every states. By default : if initProportions is set, generate values according to it. State values will be ordered by class ID

getClassId(className)¶

Return the class id corresponding to the string ‘className’

getClassName(classId)¶

Return the class name corresponding to the integer ‘classId’

getClassNames()¶

Return all the class names

getFieldValues()¶

Return all field values.

getMappedFieldValues()¶

getNbClasses()¶

getSize()¶

Return to size of the field.

randomize()¶

Randomize state values with a ramdom permutation.

setFieldValues(values, mask=None)¶

Copy the content of ‘values’ to state values masked by ‘mask’.

setFieldValues0(values, mask=None)¶

Copy the content of ‘values’ to state values masked by ‘mask’.

updateClassCounts()¶

Compute the size of every classes.

class pyhrf.boldsynth.spatialconfig.UnboundSpatialMapping(nbVoxels=100)¶

Bases: pyhrf.boldsynth.spatialconfig.SpatialMapping

Convinient class to provide an implementation of SpatialMapping when there is no mapping.

static createFromGUI()¶

Creates the actual object based on the parameters

getCoord(index)¶

getIndex(coord)¶

getMapping()¶

getNbVoxels()¶

getNdArrayMask()¶

getNeighboursCoordLists()¶

getNeighboursCoords(idVoxel)¶

getNeighboursIndexLists()¶

getNeighboursIndexes(idVoxel)¶

getRoiMask()¶

Return a binary or n-ary 3D mask which has the shape of the target data

pyhrf.boldsynth.spatialconfig.flattenElements(l)¶

pyhrf.boldsynth.spatialconfig.getRotationMatrix(axis, angle)¶

Compute the 3x3 matrix for the 3D rotation defined by ‘angle’ and the direction ‘axis’.

pyhrf.boldsynth.spatialconfig.hashMask(m)¶

pyhrf.boldsynth.spatialconfig.lattice_indexes(mask)¶

pyhrf.boldsynth.spatialconfig.maskToMapping(m)¶

pyhrf.boldsynth.spatialconfig.mask_to_coords(m)¶

class pyhrf.boldsynth.spatialconfig.xndarrayMapper1D(mapping, expandedShape, expandedAxesNames, flatAxisName)¶

Bases: pyhrf.boldsynth.spatialconfig.Mapper1D

expandxndarray(c, dest=None, fillValue=0)¶

flattenxndarray(c)¶

isExpendable(c)¶

isFlattenable(c)¶

pyhrf.jde.nrl package¶

pyhrf.jde.asl module¶

class pyhrf.jde.asl.ASLSampler(nb_iterations=3000, obs_hist_pace=-1.0, glob_obs_hist_pace=-1, smpl_hist_pace=-1.0, burnin=0.3, callback=<pyhrf.jde.samplerbase.GSDefaultCallbackHandler object>, bold_response_levels=<pyhrf.jde.asl.BOLDResponseLevelSampler object>, perf_response_levels=<pyhrf.jde.asl.PerfResponseLevelSampler object>, labels=<pyhrf.jde.asl.LabelSampler object>, noise_var=<pyhrf.jde.asl.NoiseVarianceSampler object>, brf=<pyhrf.jde.asl.BOLDResponseSampler object>, brf_var=<pyhrf.jde.asl.BOLDResponseVarianceSampler object>, prf=<pyhrf.jde.asl.PerfResponseSampler object>, prf_var=<pyhrf.jde.asl.PerfResponseVarianceSampler object>, bold_mixt_params=<pyhrf.jde.asl.BOLDMixtureSampler object>, perf_mixt_params=<pyhrf.jde.asl.PerfMixtureSampler object>, drift=<pyhrf.jde.asl.DriftCoeffSampler object>, drift_var=<pyhrf.jde.asl.DriftVarianceSampler object>, perf_baseline=<pyhrf.jde.asl.PerfBaselineSampler object>, perf_baseline_var=<pyhrf.jde.asl.PerfBaselineVarianceSampler object>, check_final_value=None, output_fit=False)¶

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

computeFit()¶

default_nb_its = 3000¶

finalizeSampling()¶

getGlobalOutputs()¶

inputClass¶

alias of WN_BiG_ASLSamplerInput

parametersToShow = ['nb_its', 'bold_response_levels', 'brf', 'brf_var', 'prf', 'prf_var']¶

class pyhrf.jde.asl.BOLDMixtureSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.MixtureParamsSampler, pyhrf.xmlio.Initable

get_true_values_from_simulation_cdefs(cdefs)¶

class pyhrf.jde.asl.BOLDResponseLevelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.ResponseLevelSampler, pyhrf.xmlio.Initable

computeVarYTildeOpt(update_perf=False)¶

if update_perf is True then also update sumcXg and prl.ytilde update_perf should only be used at init of variable values.

getOutputs()¶

samplingWarmUp(v)¶

class pyhrf.jde.asl.BOLDResponseSampler(smooth_order=2, zero_constraint=True, duration=25.0, normalise=1.0, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.ResponseSampler, pyhrf.xmlio.Initable

computeYTilde()¶

y - sum cWXg - Pl - wa

get_mat_X()¶

get_mat_XtX()¶

get_stackX()¶

class pyhrf.jde.asl.BOLDResponseVarianceSampler(val_ini=array([ 0.001]), do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.ResponseVarianceSampler, pyhrf.xmlio.Initable

class pyhrf.jde.asl.DriftCoeffSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_y_tilde()¶

get_accuracy(abs_error, rel_error, fv, tv, atol, rtol)¶

Return the accuray of the estimate fv, compared to the true value tv

Output:

axes_names (list of str), accuracy (numpy array of booleans)

get_final_value()¶

get_true_value()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

updateNorm()¶

class pyhrf.jde.asl.DriftVarianceSampler(val_ini=array([ 1.]), do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl.LabelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

CLASSES = array([0, 1])¶

CLASS_NAMES = ['inactiv', 'activ']¶

L_CA = 1¶

L_CI = 0¶

checkAndSetInitValue(variables)¶

compute_ext_field()¶

countLabels()¶

get_MAP_labels()¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

samplingWarmUp(v)¶

Called before the launch of the main sampling loop by the sampler engine. Should be overriden and perform precalculations.

class pyhrf.jde.asl.MixtureParamsSampler(name, response_level_name, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.samplerbase.GibbsSamplerVariable

I_MEAN_CA = 0¶

I_VAR_CA = 1¶

I_VAR_CI = 2¶

L_CA = 1¶

L_CI = 0¶

NB_PARAMS = 3¶

PARAMS_NAMES = ['Mean_Activ', 'Var_Activ', 'Var_Inactiv']¶

checkAndSetInitValue(variables)¶

computeWithJeffreyPriors(j, cardCIj, cardCAj)¶

get_current_means()¶

get_current_vars()¶

get_true_values_from_simulation_dict()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

class pyhrf.jde.asl.NoiseVarianceSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_y_tilde()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl.PerfBaselineSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_residuals()¶

compute_wa(a=None)¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl.PerfBaselineVarianceSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl.PerfMixtureSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.MixtureParamsSampler, pyhrf.xmlio.Initable

checkAndSetInitValue(variables)¶

get_true_values_from_simulation_cdefs(cdefs)¶

class pyhrf.jde.asl.PerfResponseLevelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.ResponseLevelSampler, pyhrf.xmlio.Initable

checkAndSetInitValue(variables)¶

computeVarYTildeOpt()¶

class pyhrf.jde.asl.PerfResponseSampler(smooth_order=2, zero_constraint=True, duration=25.0, normalise=1.0, val_ini=None, do_sampling=True, use_true_value=False, diff_res=True)¶

Bases: pyhrf.jde.asl.ResponseSampler, pyhrf.xmlio.Initable

computeYTilde()¶

y - sum aXh - Pl - wa

get_mat_X()¶

get_mat_XtX()¶

get_stackX()¶

class pyhrf.jde.asl.PerfResponseVarianceSampler(val_ini=array([ 0.001]), do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl.ResponseVarianceSampler, pyhrf.xmlio.Initable

class pyhrf.jde.asl.ResponseLevelSampler(name, response_name, mixture_name, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.samplerbase.GibbsSamplerVariable

checkAndSetInitValue(variables)¶

computeRR()¶

computeVarYTildeOpt()¶

getOutputs()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

samplingWarmUp(variables)¶

updateObsersables()¶

class pyhrf.jde.asl.ResponseSampler(name, response_level_name, variance_name, smooth_order=2, zero_constraint=True, duration=25.0, normalise=1.0, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.samplerbase.GibbsSamplerVariable

Generic parent class to perfusion response & BOLD response samplers

calcXResp(resp, stackX=None)¶

checkAndSetInitValue(variables)¶

computeYTilde()¶

get_mat_X()¶

get_mat_XtX()¶

get_rlrl()¶

get_stackX()¶

get_ybar()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

setFinalValue()¶

updateNorm()¶

updateXResp()¶

class pyhrf.jde.asl.ResponseVarianceSampler(name, response_name, val_ini=array([ 0.001]), do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.samplerbase.GibbsSamplerVariable

checkAndSetInitValue(v)¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

class pyhrf.jde.asl.WN_BiG_ASLSamplerInput(data, dt, typeLFD, paramLFD, hrfZc, hrfDuration)¶

Bases: pyhrf.jde.models.WN_BiG_Drift_BOLDSamplerInput

cleanPrecalculations()¶

makePrecalculations()¶

pyhrf.jde.asl.b()¶

pyhrf.jde.asl.compute_StS_StY(rls, v_b, mx, mxtx, ybar, rlrl, yaj, ajak_vb)¶

yaj and ajak_vb are only used to store intermediate quantities, they’re not inputs.

pyhrf.jde.asl.randn(d0, d1, ..., dn)¶

Return a sample (or samples) from the “standard normal” distribution.

If positive, int_like or int-convertible arguments are provided, randn generates an array of shape (d0, d1, ..., dn), filled with random floats sampled from a univariate “normal” (Gaussian) distribution of mean 0 and variance 1 (if any of the are floats, they are first converted to integers by truncation). A single float randomly sampled from the distribution is returned if no argument is provided.

This is a convenience function. If you want an interface that takes a tuple as the first argument, use numpy.random.standard_normal instead.

Parameters:	d1, .., dn (d0,) – The dimensions of the returned array, should be all positive. If no argument is given a single Python float is returned.
Returns:	Z – A (d0, d1, ..., dn)-shaped array of floating-point samples from the standard normal distribution, or a single such float if no parameters were supplied.
Return type:	ndarray or float

See also

random.standard_normal()

Similar, but takes a tuple as its argument.

Notes

For random samples from , use:

sigma * np.random.randn(...) + mu

Examples

>>> np.random.randn()
2.1923875335537315 #random

Two-by-four array of samples from N(3, 6.25):

>>> 2.5 * np.random.randn(2, 4) + 3
array([[-4.49401501,  4.00950034, -1.81814867,  7.29718677],  #random
       [ 0.39924804,  4.68456316,  4.99394529,  4.84057254]]) #random

pyhrf.jde.asl.simulate_asl(output_dir=None, noise_scenario='high_snr', spatial_size='tiny', v_noise=None, dt=0.5, tr=2.5)¶

pyhrf.jde.asl_2steps module¶

pyhrf.jde.asl_2steps.dummy_jde(fmri_data, dt)¶

pyhrf.jde.asl_2steps.jde_analyse_2steps_v1(output_dir, fmri_data, dt, nb_iterations, brf_var=None, do_sampling_brf_var=False, prf_var=None, do_sampling_prf_var=False)¶

#Return: # dict of outputs

pyhrf.jde.asl_2steps.physio_build_jde_mcmc_sampler(nb_iterations, rf_prior, flag_zc=False, brf_var_ini=None, prf_var_ini=None, do_sampling_brf_var=False, do_sampling_prf_var=False, prf_ini=None, do_sampling_prf=True, prls_ini=None, do_sampling_prls=True, brf_ini=None, do_sampling_brf=True, brls_ini=None, do_sampling_brls=True, perf_bl_ini=None, do_sampling_perf_bl=True, do_sampling_perf_var=True, drift_ini=None, do_sampling_drift=True, drift_var_ini=None, do_sampling_drift_var=True, noise_var_ini=None, labels_ini=None, do_sampling_labels=True)¶

pyhrf.jde.asl_physio module¶

class pyhrf.jde.asl_physio.ASLPhysioSampler(nb_iterations=3000, obs_hist_pace=-1.0, glob_obs_hist_pace=-1, smpl_hist_pace=-1.0, burnin=0.3, callback=<pyhrf.jde.samplerbase.GSDefaultCallbackHandler object>, bold_response_levels=<pyhrf.jde.asl_physio.BOLDResponseLevelSampler object>, perf_response_levels=<pyhrf.jde.asl_physio.PerfResponseLevelSampler object>, labels=<pyhrf.jde.asl_physio.LabelSampler object>, noise_var=<pyhrf.jde.asl_physio.NoiseVarianceSampler object>, brf=<pyhrf.jde.asl_physio.PhysioBOLDResponseSampler object>, brf_var=<pyhrf.jde.asl_physio.PhysioBOLDResponseVarianceSampler object>, prf=<pyhrf.jde.asl_physio.PhysioPerfResponseSampler object>, prf_var=<pyhrf.jde.asl_physio.PhysioPerfResponseVarianceSampler object>, bold_mixt_params=<pyhrf.jde.asl_physio.BOLDMixtureSampler object>, perf_mixt_params=<pyhrf.jde.asl_physio.PerfMixtureSampler object>, drift=<pyhrf.jde.asl_physio.DriftCoeffSampler object>, drift_var=<pyhrf.jde.asl_physio.DriftVarianceSampler object>, perf_baseline=<pyhrf.jde.asl_physio.PerfBaselineSampler object>, perf_baseline_var=<pyhrf.jde.asl_physio.PerfBaselineVarianceSampler object>, check_final_value=None, output_fit=False)¶

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

computeFit()¶

default_nb_its = 3000¶

finalizeSampling()¶

getGlobalOutputs()¶

inputClass¶

alias of WN_BiG_ASLSamplerInput

parametersToShow = ['nb_its', 'response_levels', 'hrf', 'hrf_var']¶

class pyhrf.jde.asl_physio.BOLDMixtureSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl_physio.MixtureParamsSampler, pyhrf.xmlio.Initable

get_true_values_from_simulation_cdefs(cdefs)¶

class pyhrf.jde.asl_physio.BOLDResponseLevelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl_physio.ResponseLevelSampler, pyhrf.xmlio.Initable

computeVarYTildeOpt(update_perf=False)¶

if update_perf is True then also update sumcXg and prl.ytilde update_perf should only be used at init of variable values.

getOutputs()¶

samplingWarmUp(v)¶

class pyhrf.jde.asl_physio.DriftCoeffSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_y_tilde()¶

getOutputs()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

samplingWarmUp(v)¶

Called before the launch of the main sampling loop by the sampler engine. Should be overriden and perform precalculations.

updateNorm()¶

class pyhrf.jde.asl_physio.DriftVarianceSampler(val_ini=array([ 1.]), do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl_physio.LabelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

CLASSES = array([0, 1])¶

CLASS_NAMES = ['inactiv', 'activ']¶

L_CA = 1¶

L_CI = 0¶

checkAndSetInitValue(variables)¶

compute_ext_field()¶

countLabels()¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

samplingWarmUp(v)¶

Called before the launch of the main sampling loop by the sampler engine. Should be overriden and perform precalculations.

class pyhrf.jde.asl_physio.MixtureParamsSampler(name, response_level_name, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.samplerbase.GibbsSamplerVariable

I_MEAN_CA = 0¶

I_VAR_CA = 1¶

I_VAR_CI = 2¶

L_CA = 1¶

L_CI = 0¶

NB_PARAMS = 3¶

PARAMS_NAMES = ['Mean_Activ', 'Var_Activ', 'Var_Inactiv']¶

checkAndSetInitValue(variables)¶

computeWithJeffreyPriors(j, cardCIj, cardCAj)¶

get_current_means()¶

get_current_vars()¶

get_true_values_from_simulation_dict()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

class pyhrf.jde.asl_physio.NoiseVarianceSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_y_tilde()¶

linkToData(dataInput)¶

sampleNextInternal(variables)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl_physio.PerfBaselineSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

compute_residuals()¶

compute_wa(a=None)¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl_physio.PerfBaselineVarianceSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

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

checkAndSetInitValue(variables)¶

linkToData(dataInput)¶

sampleNextInternal(v)¶

Define the behaviour of the variable at each sampling step when its sampling is not activated. Must be overriden in child classes.

class pyhrf.jde.asl_physio.PerfMixtureSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl_physio.MixtureParamsSampler, pyhrf.xmlio.Initable

checkAndSetInitValue(variables)¶

get_true_values_from_simulation_cdefs(cdefs)¶

class pyhrf.jde.asl_physio.PerfResponseLevelSampler(val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl_physio.ResponseLevelSampler, pyhrf.xmlio.Initable

checkAndSetInitValue(variables)¶

computeVarYTildeOpt()¶

class pyhrf.jde.asl_physio.PhysioBOLDResponseSampler(smooth_order=2, zero_constraint=True, duration=25.0, normalise=0.0, val_ini=None, do_sampling=True, use_true_value=False)¶

Bases: pyhrf.jde.asl_physio.ResponseSampler, pyhrf.xmlio.Initable

computeYTilde()¶

y - sum cWXg - Pl - wa

get_mat_X()¶

get_mat_XtX()¶

get_stackX()¶

sampleNextInternal(variables)¶

Sample BRF

changes to mean: changes to var:

samplingWarmUp(v)¶

Called before the launch of the main sampling loop by the sampler engine. Should be overriden and perform precalculations.