Extract conditional probabilities and clique potentials from data.

Extract list of conditional probability tables and list of clique potentials from data.

extractCPT(data_, graph, smooth = 0)

extractPOT(data_, graph, smooth = 0)

extractMARG(data_, graph, smooth = 0)

Arguments

data_

A named array or a dataframe.

graph

An igraph object or a list or formula which can be turned into a igraph object by calling ug or dag. For extract_cpt, graph must be/define a DAG while for extract_pot, graph must be/define undirected triangulated graph.

smooth

See 'details' below.

Value

• extract_cpt: A list of conditional probability tables.

• extract_pot: A list of clique potentials.

• extract_marg: A list of clique marginals.

Details

If smooth is non-zero then smooth is added to all cell counts before normalization takes place.

References

Søren Højsgaard (2012). Graphical Independence Networks with the gRain Package for R. Journal of Statistical Software, 46(10), 1-26. https://www.jstatsoft.org/v46/i10/.

See also

compileCPT, compilePOT, grain

Author

Søren Højsgaard, sorenh@math.aau.dk

Examples

## Extract cpts / clique potentials from data and graph
# specification and create network. There are different ways:

data(lizard, package="gRbase")

# DAG: height <- species -> diam
daG <- dag(~species + height:species + diam:species, result="igraph")

# UG : [height:species][diam:species]
uG  <- ug(~height:species + diam:species, result="igraph")

pt <- extract_pot(lizard, ~height:species + diam:species) 
cp <- extract_cpt(lizard, ~species + height:species + diam:species)

pt
#> [[1]]
#>         species
#> height       anoli      dist
#>   >4.75  0.1051345 0.2493888
#>   <=4.75 0.2958435 0.3496333
#> 
#> [[2]]
#>      species
#> diam      anoli      dist
#>   <=4 0.7195122 0.5469388
#>   >4  0.2804878 0.4530612
#> 
#> attr(,"rip")
#> cliques
#>   1 : height species 
#>   2 : species diam 
#> separators
#>   1 :  
#>   2 : species 
#> parents
#>   1 : 0 
#>   2 : 1 
#> attr(,"graph")
#> IGRAPH f63b773 UN-- 3 2 -- 
#> + attr: name (v/c)
#> + edges from f63b773 (vertex names):
#> [1] height --species species--diam   
#> attr(,"class")
#> [1] "pot_representation"
cp
#> [[1]]
#> species
#>    anoli     dist 
#> 0.400978 0.599022 
#> 
#> [[2]]
#>         species
#> height       anoli      dist
#>   >4.75  0.2621951 0.4163265
#>   <=4.75 0.7378049 0.5836735
#> 
#> [[3]]
#>      species
#> diam      anoli      dist
#>   <=4 0.7195122 0.5469388
#>   >4  0.2804878 0.4530612
#> 
#> attr(,"graph")
#> IGRAPH b2e0ec3 DN-- 3 2 -- 
#> + attr: name (v/c)
#> + edges from b2e0ec3 (vertex names):
#> [1] species->height species->diam  
#> attr(,"class")
#> [1] "cpt_representation"

# Both specify the same probability distribution
tabListMult(pt) |> as.data.frame.table()
#>   height diam species       Freq
#> 1  >4.75  <=4   anoli 0.07564554
#> 2 <=4.75  <=4   anoli 0.21286302
#> 3  >4.75   >4   anoli 0.02948894
#> 4 <=4.75   >4   anoli 0.08298050
#> 5  >4.75  <=4    dist 0.13640038
#> 6 <=4.75  <=4    dist 0.19122798
#> 7  >4.75   >4    dist 0.11298837
#> 8 <=4.75   >4    dist 0.15840527
tabListMult(cp) |> as.data.frame.table()
#>   height diam species       Freq
#> 1  >4.75  <=4   anoli 0.07564554
#> 2 <=4.75  <=4   anoli 0.21286302
#> 3  >4.75   >4   anoli 0.02948894
#> 4 <=4.75   >4   anoli 0.08298050
#> 5  >4.75  <=4    dist 0.13640038
#> 6 <=4.75  <=4    dist 0.19122798
#> 7  >4.75   >4    dist 0.11298837
#> 8 <=4.75   >4    dist 0.15840527

if (FALSE) { # \dontrun{
# Bayesian networks can be created as
bn.uG   <- grain(pt)
bn.daG  <- grain(cp)

# The steps above are wrapped into a convenience method which
# builds a network from at graph and data.
bn.uG   <- grain(uG, data=lizard)
bn.daG  <- grain(daG, data=lizard)
} # }