Overview

Summary of the API

As the API reports, B-FADE collates the following sub-modules.

• bfade.abstract contains abstract classes, which are use to opportunely define concrete classes for curves, viewers, Bayesian infrastructure.

• bfade.dataset is a collection of classes defining the structure of the datasets to be processed by B-FADE. The module include a class to generate datasets (grids for instance).

• bfade.statistics contains wrappers of scipy probability distributions and a class for Monte Carlo simulations.

• bfade.viewers includes graphical output utilities (concrete classes) allowing the user to inspect both data and results.

• bfade.elhaddad is a collection of concrete classes defining the representation of the El Haddad curve, managing the datasets, and performing MAP over El Haddad parameters.

• bfade.util is a set of functions recalled in the above-mentioned modules to support their implementation.

Notation

The documentation uses the following notation to describe the functionalities of B-FADE:

• \(P\) denotes a proability distribution.

• \(D\) is a dataset.

• \(\theta\) is the aggregated vector of the trainable parameters.