Returns (marginal/conditional) Shannon entropy of the Ising model.
IsingEntropy(graph, thresholds, beta = 1, conditional = numeric(0),
marginalize = numeric(0), base = 2, responses = c(0L, 1L),
delta = 0)
Weights matrix
Thresholds vector
Inverse temperature
Indices of nodes to condition on
Indices of nodes to marginalize over
Base of the logarithm
Vector of outcome responses.
Optional per-node quadratic (Blume-Capel) term added to the Hamiltonian as delta_i * s_i^2; a single value (recycled over nodes) or one value per node. The default 0 corresponds to the ordinary Ising model. See BlumeCapelSampler.
Sacha Epskamp <mail@sachaepskamp.com>