---
title: "002 - Random Graphs"
output: html_document
---
Creating a randomized graph is actually quite useful. Seeing these graphs with specified numbers of nodes and edges will allow you to quickly get a sense of how connected graphs can be at different sizes.
## Setup
Ensure that the development version of **DiagrammeR** is installed. Load in the package with `library()`. Additionally, load in the **tidyverse** packages.
```{r load_packages, message=FALSE, warning=FALSE, include=FALSE, results=FALSE}
#devtools::install_github("rich-iannone/DiagrammeR")
library(DiagrammeR)
library(tidyverse)
```
## Part 1. Random Graphs with `create_graph()` + `add_gnm_graph()`
There are several functions that help us to create randomized graph structures. One of the easier ones to reason about is the `add_gnm_graph()` function. That function creates a graph structure using the G(n, m) Erdos-Renyi graph algorithm and inserts that into an existing graph object. Thus, for all the following example, we'll create an empty graph with `create_graph()`, then we'll use the forward pipe (`%>%`) to add the randomized graph, then finally we'll view the graph with `render_graph()`.
To start, we can create a somewhat random graph with two nodes (`n` argument) and one edge (`m` argument). The resulting graph will be a directed graph because we initialized the graph space with `create_graph()` and the default value for `directed` is `TRUE`.
```{r create_graph_gnm_graph_render}
# Create a random graph with 2 nodes, an edge,
# and display the graph
create_graph() %>%
add_gnm_graph(
n = 2, m = 1) %>%
render_graph()
```
Although there are only two nodes and a single edge, you might note that when repeatedly running the above code the edge can either have the definition `1 -> 2` or `2 -> 1`. If you'd like to make the resulting graph reproducible (and we'll do this for all subsequent examples), we can use the `set_seed` argument and supply an integer. This works for each separate call of `add_gnm_graph()` since we can add multiple random graphs to the graph space.
```{r create_graph_multi_gnm_graph_set_seed_render}
# Create a random graph with 2 nodes, an edge,
# and display the graph
create_graph() %>%
add_gnm_graph(
n = 2, m = 1,
set_seed = 23) %>%
add_gnm_graph(
n = 2, m = 1,
set_seed = 23) %>%
render_graph()
```
It might be more interesting to make a graph with a higher number of nodes and edges. Next, we will create a G(n, m) graph with 15 nodes and 30 edges:
```{r create_graph_gnm_graph_render_2}
create_graph() %>%
add_gnm_graph(
n = 15, m = 30) %>%
render_graph()
```
In all of the random graphs created with this function, there can only be up to 2 edges created for each pair of nodes (e.g, `1 -> 2` and `2 -> 1` edges for nodes `1` and `2`). In other words, there won't be what are termed *multiple edges* between nodes.
If you specify a number of edges that exceeds the number in a fully-connected graph of size `n`, you'll get an error.
```{r create_random_graph_too_many_edges, message=TRUE, error=TRUE}
try(
create_graph() %>%
add_gnm_graph(n = 5, m = 25) %>%
render_graph(layout = "kk"))
```
So, using `n = 15` and `m = 105` will yield a fully-connected graph (without loops though) with 15 nodes (this is much the same as using `create_graph() %>% add_full_graph(n = 15, keep_loops = FALSE)`). Here is the output graph:
```{r create_random_graph_fully_connected}
create_graph() %>%
add_gnm_graph(
n = 15, m = 105) %>%
render_graph(layout = "kk")
```
Alternatively, you don't need to have edges in the randomly generated graph. Simply specify `m = 0` for any number of nodes `n`:
```{r create_random_graph_512_nodes_0_edges, echo=TRUE, paged.print=TRUE}
# Create a random graph with 512 nodes and no edges;
# output a tibble with node and edge counts
create_graph() %>%
add_gnm_graph(
n = 512, m = 0) %>%
{
dplyr::tibble(
nodes = count_nodes(.),
edges = count_edges(.))
}
```
# Part 2. Random Yet Reproducible
Setting a seed is a great way to create something random yet reproduceable/re-usable. This can be done within the `add_gnm_graph()` function by specifying a seed number with the argument `set_seed`. Here's an example
```{r create_gnm_graph_w_seed}
create_graph() %>%
add_gnm_graph(
n = 4, m = 4,
set_seed = 30) %>%
render_graph()
```
Upon repeat runs, the connections in the graph will be the same each and every time (e.g., node `1` is attached to `2` and `4`, `3` is connected to `2`, etc.).
## Part 3. Directed/Undirected Randomness
By default, the graphs generated are directed. To produce undirected graphs, simply include `directed = FALSE` in the `create_graph()` statement, then, use a function that generates a random graph in the usual way.
```{r create_gnm_graph_directed}
create_graph(directed = FALSE) %>%
add_gnm_graph(
n = 15, m = 22,
set_seed = 23) %>%
render_graph()
```