mlindsk/jti

Juntion Tree Inference

jti's Issues

Error in example

When running the example on the Github page while the bnlearn package is loaded, I get the following error:

Error in which.undirected(arcs, nodes = nodes) : 
  'duplicated' applies only to vectors

If bnlearn is not loaded, the example works fine. However, I intend to use the package with networks estimated using bnlearn.

Full reprex below:

library(jti)
#> 
#> Attaching package: 'jti'
#> The following object is masked from 'package:methods':
#> 
#>     initialize
library(igraph)
#> 
#> Attaching package: 'igraph'
#> The following objects are masked from 'package:stats':
#> 
#>     decompose, spectrum
#> The following object is masked from 'package:base':
#> 
#>     union

el <- matrix(c(
  "A", "T",
  "T", "E",
  "S", "L",
  "S", "B",
  "L", "E",
  "E", "X",
  "E", "D",
  "B", "D"),
  nc = 2,
  byrow = TRUE
)

g <- igraph::graph_from_edgelist(el)
plot(g)

cl <- cpt_list(asia, g)
cl
#>  List of CPTs 
#>  -------------------------
#>   P( A )
#>   P( T | A )
#>   P( E | T, L )
#>   P( S )
#>   P( L | S )
#>   P( B | S )
#>   P( X | E )
#>   P( D | E, B )
#> 
#>   <bn, cpt_list, list> 
#>  -------------------------

library(bnlearn)
#> 
#> Attaching package: 'bnlearn'
#> The following objects are masked from 'package:igraph':
#> 
#>     as.igraph, compare, degree, subgraph
#> The following objects are masked from 'package:jti':
#> 
#>     asia, parents

el <- matrix(c(
  "A", "T",
  "T", "E",
  "S", "L",
  "S", "B",
  "L", "E",
  "E", "X",
  "E", "D",
  "B", "D"),
  nc = 2,
  byrow = TRUE
)

g <- igraph::graph_from_edgelist(el)
plot(g)

cl <- cpt_list(asia, g)
cl
#> Error in which.undirected(arcs, nodes = nodes): 'duplicated' applies only to vectors

^{Created on 2022-06-02 by the reprex package (v2.0.1)}