SPFlow API

Structure

spn.structure.Base.Context([meta_types, ...])

spn.structure.Base.Sum([weights, children])

spn.structure.Base.Product([children])

spn.structure.Base.assign_ids(node[, ids])

spn.structure.Base.rebuild_scopes_bottom_up(node)

spn.structure.StatisticalTypes.MetaType(value)

An enumeration.

spn.structure.leaves.cltree.CLTree.create_cltree_leaf(...)

spn.structure.leaves.parametric.Parametric.Bernoulli([...])

Implements a univariate Bernoulli distribution with parameter p (probability of a success)

spn.structure.leaves.parametric.Parametric.Categorical([...])

Implements a univariate categorical distribution with $k$ parameters {pi_{k}}

spn.structure.leaves.parametric.Parametric.Gaussian([...])

Implements a univariate gaussian distribution with parameters mu(mean) sigma ^ 2 (variance) (alternatively sigma is the standard deviation(stdev) and sigma ^ {-2} the precision)

Learning

spn.algorithms.LearningWrappers.learn_classifier(...)

spn.algorithms.LearningWrappers.learn_cnet(...)

spn.algorithms.LearningWrappers.learn_mspn(...)

spn.algorithms.LearningWrappers.learn_parametric(...)

Inference

spn.algorithms.Inference.log_likelihood(...)

spn.algorithms.Marginalization.marginalize(...)

spn.algorithms.MPE.mpe(node, input_data[, ...])

Utility Methods

These generally relate to visualization, statistics, and whether the structure is valid.

spn.io.Graphics.draw_spn(spn)

spn.io.Graphics.plot_spn(spn[, fname])

spn.algorithms.Statistics.get_structure_stats(node)

spn.algorithms.Validity.is_valid(node[, ...])

Datasets

The following data sets are included here:

spn.data.datasets

Created on March 30, 2018