DiagrammeR (version 1.0.0)

add_full_graph: Add a fully connected graph

Description

With a graph object of class dgr_graph, add a fully connected graph either with or without loops. If the graph object set as directed, the added graph will have edges to and from each pair of nodes. In the undirected case, a single edge will link each pair of nodes.

Usage

add_full_graph(graph, n, type = NULL, label = TRUE, rel = NULL,
  edge_wt_matrix = NULL, keep_loops = FALSE, node_aes = NULL,
  edge_aes = NULL, node_data = NULL, edge_data = NULL)

Arguments

graph

a graph object of class dgr_graph.

n

the number of nodes comprising the fully connected graph.

type

an optional string that describes the entity type for the nodes to be added.

label

either a vector object of length n that provides optional labels for the new nodes, or, a boolean value where setting to TRUE ascribes node IDs to the label and FALSE or NULL yields a blank label.

rel

an optional string for providing a relationship label to all new edges created in the connected graph.

edge_wt_matrix

an optional matrix of n by n dimensions containing values to apply as edge weights. If the matrix has row names or column names and label = TRUE, those row or column names will be used as node label values.

keep_loops

an option to simplify the fully connected graph by removing loops (edges from and to the same node). The default value is FALSE.

node_aes

an optional list of named vectors comprising node aesthetic attributes. The helper function node_aes() is strongly recommended for use here as it contains arguments for each of the accepted node aesthetic attributes (e.g., shape, style, color, fillcolor).

edge_aes

an optional list of named vectors comprising edge aesthetic attributes. The helper function edge_aes() is strongly recommended for use here as it contains arguments for each of the accepted edge aesthetic attributes (e.g., shape, style, penwidth, color).

node_data

an optional list of named vectors comprising node data attributes. The helper function node_data() is strongly recommended for use here as it helps bind data specifically to the created nodes.

edge_data

an optional list of named vectors comprising edge data attributes. The helper function edge_data() is strongly recommended for use here as it helps bind data specifically to the created edges.

Value

a graph object of class dgr_graph.

Examples

Run this code
# NOT RUN {
# Create a new graph object
# and add a directed and fully
# connected graph with 3 nodes
# and edges to and from all
# pairs of nodes; with the option
# `keep_loops = TRUE` nodes
# will also have edges from
# and to themselves
graph <-
  create_graph() %>%
  add_full_graph(
    n = 3, keep_loops = TRUE)

# Get node information
# from this graph
graph %>%
  get_node_info()

# Using `keep_loops = FALSE`
# (the default) will remove
# the loops
create_graph() %>%
  add_full_graph(n = 3) %>%
  get_node_info()

# Values can be set for
# the node `label`, node
# `type`, and edge `rel`
graph <-
  create_graph() %>%
  add_full_graph(
    n = 3,
    type = "connected",
    label = c("1st", "2nd", "3rd"),
    rel = "connected_to")

# Show the graph's node
# data frame (ndf)
graph %>%
  get_node_df()

# Show the graph's edge
# data frame (edf)
graph %>%
  get_edge_df()

# Create a fully-connected and
# directed graph with 3 nodes,
# and, where a matrix provides
# edge weights; first, create the
# matrix (with row names to be
# used as node labels)
set.seed(23)

edge_wt_matrix <-
  rnorm(100, 5, 2) %>%
  sample(9, FALSE) %>%
  round(2) %>%
  matrix(
    nc = 3,
    nr = 3,
    dimnames = list(c("a", "b", "c")))

# Create the fully-connected
# graph (without loops however)
graph <-
  create_graph() %>%
  add_full_graph(
    n = 3,
    type = "weighted",
    label = TRUE,
    rel = "related_to",
    edge_wt_matrix = edge_wt_matrix,
    keep_loops = FALSE)

# Show the graph's node
# data frame (ndf)
graph %>%
  get_node_df()

# Show the graph's edge
# data frame (edf)
graph %>%
  get_edge_df()

# An undirected graph can
# also use a matrix with
# edge weights, but only
# the lower triangle of
# that matrix will be used
create_graph(directed = FALSE) %>%
  add_full_graph(
    n = 3,
    type = "weighted",
    label = TRUE,
    rel = "related_to",
    edge_wt_matrix = edge_wt_matrix,
    keep_loops = FALSE) %>%
  get_edge_df()
# }

Run the code above in your browser using DataCamp Workspace