View source: R/simulation_util.R
sim.random.field | R Documentation |
Simulate an random MRF with random unary (node) and pair (edge) potentials.
sim.random.field(
adj.mat = NULL,
num.nodes,
num.edges = NULL,
prob.edge = 1,
node.pot.nu = 30,
node.pot.mean,
node.pot.sd,
edge.pot.nu = 30,
edge.pot.mean,
edge.pot.sd,
num.sims,
seed.node.pots = NULL,
seed.edge.pots = NULL,
seed.sample = NULL,
plotQ = F
)
XX |
The XX |
Uses standard parameterization w11=w22, w12=w21. Graph is generated with erdos.renyi.game in igraph. If an edge appears in the graph, a relatively high to high potential (i.e. non-zero) should be generated to ensure the edge is potentially strong enough to be found from a random sample from the graph. However, the user can choose the the general strength and variability of the edge and node potentials though the edge.pot.mean/sd and node.pot.mean/sd arguments. They are generated with a t-distribution. The default nu is 30 making the random selections approximately normal. The number of edges in a randomly generated graph can be chosen by specification or by probability.
The function will XX
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.