Pnode.Properties: Properties of class '"NeticaNode"' as a '"Pnode"'
In ralmond/PNetica: Parameterized Bayesian Networks Netica Interface

Pnode.Properties

R Documentation

Properties of class `"NeticaNode"` as a `"Pnode"`

Description

The PNetica package supplies the needed methods so that the RNetica::NeticaNode object is an instance of the Peanut::Pnode object. As a Pnode is nominally parameterized, the are given the special label “pnode” to indicate that this note has parametric information. This document describes the extra properties of Pnodes that are added by PNetica.

Extends

See NeticaNode for a description of the Netica class.

With these methods, NeticaNode now extends Pnode.

All reference classes extend and inherit methods from "envRefClass".

Methods

All methods are implementations of generic functions in the Peanut package. These methods are related to the parameteric information which makes a node a Pnode. To inidcate that a node has this extra information, it should have the “"pnode"” label. The functions Pnode and as.Pnode will do this.

Pnode: signature(node = "NeticaNode", lnAlphas, betas, rules = "Compensatory", link = "partialCredit", Q = TRUE, linkScale = NULL, priorWeight = NULL): This function forces a NeticaNode into a Pnode by initializing the Pnode-specific fields.
PnodeLnAlphas: signature(node = NeticaNode): Returns the log of discrimination parameters associated with the node.
PnodeLnAlphas<-: signature(node = NeticaNode): Sets the log of discrimination parameters associated with the node.
PnodeBetas: signature(node = NeticaNode): Returns the difficulty parameters associated with the node.
PnodeBetas<-: signature(node = NeticaNode): Sets the difficulty parameters associated with the node.
PnodeQ: signature(node = NeticaNode): Returns the local Q matrix associated with the node.
PnodeQ<-: signature(node = NeticaNode): Sets the local Q matrix associated with the node.
PnodeRules: signature(node = NeticaNode): Returns the names of the combination rules associated with the node.
PnodeRules<-: signature(node = NeticaNode): Sets the names of the combination rules associated with the node.
PnodeLink: signature(node = NeticaNode): Returns the link function associated with the node.
PnodeLink<-: signature(node = NeticaNode): Sets the link function associated with the node.
PnodeLinkScale: signature(node = NeticaNode): Returns the link function scale parameter associated with the node.
PnodeLinkScale<-: signature(node = NeticaNode): Sets the link function scale parameter associated with the node.
PnodePriorWeight: signature(node = NeticaNode): Returns the weight or weights assigned to prior information associated with the node.
PnodePriorWeight<-: signature(node = NeticaNode): Sets the weight or weights assigned to prior information associated with the node.
PnodePostWeight: signature(node = NeticaNode): Returns the combined prior and data weights associated with the node.
as.Pnode: signature(x = NeticaNode): Forces x to be a Pnode; in particular, it adds the lable "pnode".
is.Pnode: signature(x = NeticaNode): Returns true if the node has the special label "pnode".

Documentation for other methods of the Pnode generic functions for NeticaNode objects can be found in the documentation objects Pnode.NeticaNode and Pnode.States.

Note

The “Pnode properies”, lnAlphas, betas, Q, rules, link, linkScale, and priorWeight are stored in user fields (NodeUserObj) of the Netica node. A NeticaNode object which has those fields behaves as a Pnode and is suitable for the use with Peanut. The function Pnode will add default values for these fields if they are not set.

To mark a node as a Pnode, it is added to the node set “pnode”. The is.Pnode function checks for this method.

Author(s)

Russell Almond

Examples

sess <- NeticaSession()
startSession(sess)
curd <- setwd(system.file("testnets",package="PNetica"))

tNet <- CreateNetwork("TestNet",sess)

## Alphas
theta1 <- NewDiscreteNode(tNet,"theta1",
                         c("VH","High","Mid","Low","VL"))
PnodeStateValues(theta1) <- effectiveThetas(PnodeNumStates(theta1))
PnodeProbs(theta1) <- rep(1/PnodeNumStates(theta1),PnodeNumStates(theta1))
theta2 <- NewDiscreteNode(tNet,"theta2",
                         c("VH","High","Mid","Low","VL"))
PnodeStateValues(theta2) <- effectiveThetas(PnodeNumStates(theta2))
PnodeProbs(theta2) <- rep(1/PnodeNumStates(theta1),PnodeNumStates(theta2))

partial3 <- NewDiscreteNode(tNet,"partial3",
                            c("FullCredit","PartialCredit","NoCredit"))
PnodeParents(partial3) <- list(theta1,theta2)

## Usual way to set rules is in constructor
partial3 <- Pnode(partial3,rules="Compensatory", link="partialCredit")
PnodePriorWeight(partial3) <- 10
BuildTable(partial3)

## slopes of 1 for both transitions
PnodeLnAlphas(partial3) <- c(0,0)
BuildTable(partial3)

## log slope 0 = slope 1
stopifnot(
   all(abs(PnodeAlphas(partial3) -1) <.0001)
)

## Make Skill 1 more important than Skill 2
PnodeLnAlphas(partial3) <- c(.25,-.25)
BuildTable(partial3)

## increasing intercepts for both transitions
PnodeLink(partial3) <- "gradedResponse"
PnodeBetas(partial3) <- list(FullCredit=1,PartialCredit=0)
BuildTable(partial3)
stopifnot(
   all(abs(do.call("c",PnodeBetas(partial3)) -c(1,0) ) <.0001)
)


## increasing intercepts for both transitions
PnodeLink(partial3) <- "partialCredit"
## Full Credit is still rarer than partial credit under the partial
## credit model
PnodeBetas(partial3) <- list(FullCredit=0,PartialCredit=0)
BuildTable(partial3)
stopifnot(
   all(abs(do.call("c",PnodeBetas(partial3)) -c(0,0) ) <.0001)
)


## Make Skill 1 more important for the transition to ParitalCredit
## And Skill 2 more important for the transition to FullCredit
PnodeLnAlphas(partial3) <- list(FullCredit=c(-.25,.25),
                                PartialCredit=c(.25,-.25))
BuildTable(partial3)

## Set up so that first skill only needed for first transition, second
## skill for second transition; Adjust alphas to match
PnodeQ(partial3) <- matrix(c(TRUE,TRUE,
                             TRUE,FALSE), 2,2, byrow=TRUE)
PnodeLnAlphas(partial3) <- list(FullCredit=c(-.25,.25),
                                PartialCredit=0)
BuildTable(partial3)

## Using OffsetConjunctive rule requires single slope
PnodeRules(partial3) <- "OffsetConjunctive"
## Single slope parameter for each transition
PnodeLnAlphas(partial3) <- 0
PnodeQ(partial3) <- TRUE
PnodeBetas(partial3) <- c(0,1)
BuildTable(partial3)

## Make Skill 1 more important for the transition to ParitalCredit
## And Skill 2 more important for the transition to FullCredit
PnodeLnAlphas(partial3) <- 0
PnodeBetas(partial3) <- list(FullCredit=c(-.25,.25),
                                PartialCredit=c(.25,-.25))
BuildTable(partial3)


## Separate slope parameter for each transition;  
## Note this will only different from the previous transition when
## mapDPC is called.  In the former case, it will learn a single slope
## parameter, in the latter, it will learn a different slope for each
## transition. 
PnodeLnAlphas(partial3) <- list(0,0)
BuildTable(partial3)

## Set up so that first skill only needed for first transition, second
## skill for second transition; Adjust betas to match
PnodeQ(partial3) <- matrix(c(TRUE,TRUE,
                             TRUE,FALSE), 2,2, byrow=TRUE)
PnodeBetas(partial3) <- list(FullCredit=c(-.25,.25),
                                PartialCredit=0)
BuildTable(partial3)


## Can also do this with special parameter values
PnodeQ(partial3) <- TRUE
PnodeBetas(partial3) <- list(FullCredit=c(-.25,.25),
                                PartialCredit=c(0,Inf))
BuildTable(partial3)

## The normal link function is the only one which takes a scale parameter
PnodeLink(partial3) <- "normalLink"
PnodeLinkScale(partial3) <- 1.0
PnodeLnAlphas(partial3) <- 0
PnodeBetas(partial3) <- c(0,1)
BuildTable(partial3)
stopifnot(
  all(abs(PnodeLinkScale(partial3)-1)<.0001)
)

DeleteNetwork(tNet)

stopSession(sess)
setwd(curd)

ralmond/PNetica documentation built on Sept. 19, 2023, 8:27 a.m.