Invokes Factor constructors for a standard set of Elements.
Invokes Factor constructors for a standard set of Elements. This method uses various secondary factories.
Create a DenseFactor from the supplied parent and children variables
Get the variable associated with the given element in the given component collection.
Get the variable associated with the given element in the given component collection. If the element does not exist in the component collection, an intermediate variable is created whose range is { Star }.
Make a conditional selector factor used in the decomposition of chain and other elements.
Make a conditional selector factor used in the decomposition of chain and other elements. A chain defines a factor over the parent element, each of the possible result elements of the chain, and the overall chain element. This can produce a very large factor when there are many result elements. This is solved by decomposing the chain factor into a product of factors, each of which contains the parent element, one of the result elements, and the overall chain element.
Create the probabilistic factor encoding the probability of evidence in the dependent universe as a function of the values of variables in the parent universe.
Create the probabilistic factor encoding the probability of evidence in the dependent universe as a function of the values of variables in the parent universe. The third argument is the the function to use for computing probability of evidence in the dependent universe. It is assumed that the definition of this function will already contain the right evidence.
Make the non-constraint factors corresponding to the given element in the component collection.
Make the non-constraint factors corresponding to the given element in the component collection.
If true, parameterized elements are assumed to be equal to their previously computed MAP value. If false, they are treated like any other element.
Make factors for a particular element.
Make factors for a particular element. This function wraps the SFI method of creating factors using component collections
Given a sequence of variables, create a new variable representing the tuple of the inputs and create the factor mapping the inputs to their tuple.
Given a sequence of variables, create a new variable representing the tuple of the inputs and create the factor mapping the inputs to their tuple.
the variables to be formed into a tuple
Create an intermediate variable in the given component collection with the given value set.
Create a new factor in which one variable is replaced with another.
The mutliplicative identity factor.
Methods for creating probabilistic factors associated with elements.