Infer.NET

Factors and Constraints

Infer.NET has a number of built-in factors and constraints which you can use to build your models. These are normally exposed as static methods on the Variable class, such as Variable.Discrete(). You can also add your own factors or constraints and include them in your model by using Variable<T>.Factor or Variable.Constrain.

The built-in factors fall into the following categories:

Distribution factors - factors for creating random variables with various prior distributions.
Boolean and comparison operations - factors for Boolean operations (e.g. AND/OR) and comparison operators (e.g equals).
Arithmetic and mathematical operations - factors for arithmetic operations (e.g. plus/times) and mathematical operations (e.g. exp/logistic).
Linear algebra factors - factors for linear algebra manipulations (e.g. matrix multiplication).
Constraints - the list of built-in constraints, which affect the distribution of their arguments but do not return a value.
Miscellaneous factors - other useful factors.
List of factors and constraints - shows which algorithms support which factors.

Note: some factors are marked as experimental. These are newly introduced factors whose implementation only supports a limited class of models. Also, they may not be as thoroughly tested as other factors. If you have an issue or question concerning an experimental factor, please raise it in the forums.

Last modified at 8/25/2009 2:21 PM by Tom Minka