Quantum Fog  0.9.3
Public Member Functions | Public Attributes | List of all members
prefabricated_nodes.Marginalizer.Marginalizer Class Reference
Inheritance diagram for prefabricated_nodes.Marginalizer.Marginalizer:
Inheritance graph
[legend]
Collaboration diagram for prefabricated_nodes.Marginalizer.Marginalizer:
Collaboration graph
[legend]

Public Member Functions

def __init__ (self, id_num, name, is_quantum, pa_nd, projected_axis, has_commas=True)
 
- Public Member Functions inherited from nodes.BayesNode.BayesNode
def __init__ (self, id_num, name="blank", size=2)
 
def set_potential (self, pot)
 
def resize (self, size)
 
def set_state_name (self, position, name)
 
def st_name_index (self, st_name)
 
def forget_all_evidence (self)
 
def get_active_states (self)
 
def set_active_states (self, states)
 
def set_state_names_to_product (self, list_of_iters, repeat=1, trim=False)
 
- Public Member Functions inherited from nodes.DirectedNode.DirectedNode
def __init__ (self, id_num, name="blank")
 
def add_parent (self, node)
 
def add_parents (self, node_list)
 
def remove_parent (self, node)
 
def has_parent (self, node)
 
def add_child (self, node)
 
def add_children (self, node_list)
 
def remove_child (self, node)
 
def has_child (self, node)
 
def undirect (self)
 
def get_markov_blanket (self)
 
- Public Member Functions inherited from nodes.Node.Node
def __init__ (self, id_num, name="blank")
 
def __lt__ (self, other)
 
def add_neighbor (self, node)
 
def remove_neighbor (self, node)
 
def has_neighbor (self, node)
 

Public Attributes

 projected_axis
 
 has_commas
 
 state_names
 
 potential
 
- Public Attributes inherited from nodes.BayesNode.BayesNode
 size
 
 state_names
 
 clique
 
 potential
 
- Public Attributes inherited from nodes.DirectedNode.DirectedNode
 parents
 
 children
 
 neighbors
 
- Public Attributes inherited from nodes.Node.Node
 id_num
 
 topo_index
 
 name
 
 neighbors
 
 visited
 

Additional Inherited Members

- Properties inherited from nodes.BayesNode.BayesNode
 active_states = property(get_active_states, set_active_states)
 

Detailed Description

The Constructor of this class builds a BayesNode that has a transition
matrix appropriate for a marginalizer node. Given a quantum state |a0,
a1, a2,....> and asked to project along axis or component 2,
a marginalizer node will give back |a2>. That is the gist of it.

The following is expected:

* the focus node has exactly one parent node,

* the states of the parent node are of one of two forms: has_commas=True
or False

When 'has_commas' is True, the states of the parent node must be of the
form (a_0, a_1, a_2,... a_{n-1}) where a_0 \in S_0, a_1 \in S_1,
etc. The sets S_0, S_1, ... S_{n-1} need not be the same. If
'projected_axis' equals k, then QFog will use S_k as the states of the
focus node. The transition matrix will be assigned the value \delta(b_k,
a_k), where \delta() is the Kronecker delta function, b_k is the state
of the focus node and a_k is the kth component of the state of the
parent node.

The case 'has_commas' is False is virtually identical to the True case,
except that now the states of the parent node are not tuples, and don't
have commas. Instead they are all strings of the same length. In this
case, projected axis k refers to the position k (0 based) on the string.

More information about marginalizer nodes can be found in the documents
entitled "Quantum Fog Manual", and "Quantum Fog Library Of Essays" that
are included with the legacy QFog.

Attributes
----------
has_commas : bool
projected_axis : int

potential : Potential
active_states : list[int]
clique : Clique
size : int
state_names : list[str]
children : set[BayesNode]
neighbors : set[BayesNode]
parents : set[BayesNode]
id_num : int
topo_index : int
name : str
visited : bool

Constructor & Destructor Documentation

def prefabricated_nodes.Marginalizer.Marginalizer.__init__ (   self,
  id_num,
  name,
  is_quantum,
  pa_nd,
  projected_axis,
  has_commas = True 
)
Constructor

Parameters
----------
id_num : int
    id number of self (focus node)
name : str
    name of self (focus node)
is_quantum : bool
pa_nd : BayesNode
    parent node
projected_axis : int
has_commas : bool

Returns
-------

The documentation for this class was generated from the following file: