initialize-simulatedDAG-method: creation of a "simulatedDAG" containing a list of DAGs and...

In D2C: Predicting Causal Direction from Dependency Features

Description

creation of a "simulatedDAG" containing a list of DAGs and associated observations

Usage

## S4 method for signature 'simulatedDAG'
initialize(.Object, NDAG = 1,
  noNodes = sample(10:20, size = 1), functionType = "linear",
  quantize = FALSE, verbose = TRUE, N = sample(100:500, size = 1),
  seed = 1234, sdn = 0.5, goParallel = FALSE)

Arguments

.Object	: simulatedDAG object
NDAG	: number of DAGs to be created and simulated
noNodes	: number of Nodes of the DAGs. If it is a two-valued vector , the value of Nodes is randomly sampled in the interval
functionType	: type of the dependency. It is of class "character" and is one of ("linear", "quadratic","sigmoid")
quantize	: if TRUE it discretize the observations into two bins. If it is a two-valued vector [a,b], the value of quantize is randomly sampled in the interval [a,b]
verbose	: if TRUE it prints out the state of progress
N	: number of sampled observations for each DAG. If it is a two-valued vector [a,b], the value of N is randomly sampled in the interval [a,b]
seed	: random seed
sdn	: standard deviation of aditive noise. If it is a two-valued vector, the value of N is randomly sampled in the interval
goParallel	: if TRUE it uses parallelism

References

Gianluca Bontempi, Maxime Flauder (2014) From dependency to causality: a machine learning approach. Under submission

Examples

require(RBGL)
require(gRbase)
require(foreach)
descr=new("D2C.descriptor")
descr.example<-new("D2C.descriptor",bivariate=FALSE,ns=3,acc=TRUE)
trainDAG<-new("simulatedDAG",NDAG=10, N=c(50,100),noNodes=c(15,40),
             functionType = "linear", seed=0,sdn=c(0.45,0.75))

D2C documentation built on May 29, 2017, 10:44 a.m.