mxfusion.components.variables.variable¶
Members¶
-
class
mxfusion.components.variables.variable.
VariableType
¶ Bases:
enum.Enum
Variables types:
- CONSTANT: a variable with a constant value
- PARAMETER: a variable with no prior distribution
- RANDVAR: a variable that follows a distribution or module
- FUNCVAR: a variable that is an output of a function
-
CONSTANT
= 0¶
-
PARAMETER
= 1¶
-
RANDVAR
= 2¶
-
FUNCVAR
= 3¶
-
class
mxfusion.components.variables.variable.
Variable
(value=None, shape=None, transformation=None, isInherited=False, initial_value=None)¶ Bases:
mxfusion.components.model_component.ModelComponent
Define a variable in MXFusion. The variable can be either a constant, a parameter, an outcome of a function evaluation or a random variable following a probabilistic distribution.
Parameters: - value ((optional) None or numpy array or MXNet array or float or int or FunctionEvaluation or Distribution.) – The value of variable. If it is a numpy or MXNet array, the variable is considered as a constant. If it is a function evaluation, the variable is considered as the outcome of a function evaluation. If it is a probabilistic distribution, the variable is considered as a random variable. If it is None, the variable is considered as a parameter.
- shape ((optional) tuple or None) – The expected shape of the Variable.
- transformation ((optional) transformation or None) – The constraint on the domain of the variable, e.g., a positive value.
- isInherited (bool) – True if the Variable is constructed from a GluonBlock parameter.
-
type
¶
-
as_json
()¶
-
replicate_self
(attribute_map=None)¶ This functions as a copy constructor for the object. In order to do a copy constructor we first call
__new__
on the class which creates a blank object. We then initialize that object using the methods standard init procedures, and do any extra copying of attributes.Replicates this Factor, using new inputs, outputs, and a new uuid. Used during model replication to functionally replicate a factor into a new graph.
Parameters: attribute_map ({Variable: replicated Variable}) – A mapping from attributes of this object that were Variables to their replicants.
-
shape_str
¶
-
display_str
(temp_name=None)¶
-
set_prior
(distribution)¶ Set the prior over this variable.
Param: distribution The distribution the variable is drawn from.
-
assign_factor
(factor)¶ Assign (or reassign) the distribution this variable is drawn from.
Param: distribution The distribution this variable is drawn from.
-
factor
¶
-
constant
¶ Return the stored constant.
Returns: the constant value
-
transformation
¶
-
initial_value
¶
-
initial_value_before_transformation
¶ The initial value of a variable before applying the chosen transformation (if exists)