FitMLVSBM function

An R6 Class object, a fitted multilevel network once $dovem() is done

An R6 Class object, a fitted multilevel network once $dovem() is done

An R6 Class object, a fitted multilevel network once $dovem() is done

Public fields

• vbound: The vector of variational bound for monitoring convergence

Active bindings

• affiliation_matrix: Get the affiliation matrix

• adjacency_matrix: Get the list of adjacency matrices

• nb_nodes: Get the list of the number of nodes

• nb_clusters: Get the list of the number of blocks

• parameters: Get the list of the model parameters

• membership: Get the list of the variational parameters

• independent: Are the levels independent?

• distribution: Emission distribution of each level

• directed: Are the levels directed?

• entropy: Get the entropy of the model

• bound: Get the variational bound of the model

• df_mixture: Get the degrees of freedom of the mixture parameters

• df_connect: Get the degrees of freedom of the connection parameters

• connect: Get the number of possible observed connections

• ICL: Get the ICL model selection criterion of the model

• full_penalty: Get the penalty used to compute the ICL

• Z: Get the list of block memberships (vector form)

• X_hat: Get the list of the matrices of probability connection predictions

• map: Get the list of block memberships (matrix form)

• penalty: Get the ICL penalty

• likelihood: Compute the likelihood of both levels

• complete_likelihood: Get the complete likelihood of the model

Methods

Public methods

• list("FitMLVSBM$new()")

• list("FitMLVSBM$update_alpha()")

• list("FitMLVSBM$update_pi()")

• list("FitMLVSBM$update_gamma()")

• list("FitMLVSBM$init_clustering()")

• list("FitMLVSBM$clear()")

• list("FitMLVSBM$m_step()")

• list("FitMLVSBM$ve_step()")

• list("FitMLVSBM$do_vem()")

• list("FitMLVSBM$permute_empty_class()")

• list("FitMLVSBM$plot()")

• list("FitMLVSBM$show()")

• list("FitMLVSBM$print()")

• list("FitMLVSBM$clone()")

Method new()

Constructor for the FitMLVSBM class

Usage

FitMLVSBM$new(
  Q = list(I = 1, O = 1),
  A = NA,
  X = NA,
  M = list(I = NA, O = NA),
  directed = NA,
  distribution = list("bernoulli", "bernoulli"),
  independent = FALSE
)

Arguments

• Q: List of number of blocks

• A: Affiliation matrix

• X: List of adjacency matrices

• M: List of Mask matrices

• directed: List of boolean

• distribution: List of string

• independent: Boolean

Returns

A FitMLVSBM object

Method update_alpha()

Update the connection parameters for the M step

Usage

FitMLVSBM$update_alpha(safeguard = 2 * .Machine$double.eps)

Arguments

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

Method update_pi()

Update the upper level mixture parameter for the M step

Usage

FitMLVSBM$update_pi(safeguard = 0.001)

Arguments

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

Method update_gamma()

Update the lower level mixture parameter for the M step

Usage

FitMLVSBM$update_gamma(safeguard = 1e-06)

Arguments

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

Method init_clustering()

init_clustering Initial clustering for VEM algorithm

Usage

FitMLVSBM$init_clustering(
  safeguard = 2 * .Machine$double.eps,
  method = "hierarchical",
  Z = NULL
)

Arguments

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

• method: Algorithm used to initiate the clustering, either "spectral", "hierarchical" or "merge_split" (if Z is provided)

• Z: Initial clustering if provided

Method clear()

Reset all parameters

Usage

FitMLVSBM$clear()

Method m_step()

m_step Compute the M step of the VEM algorithm

Usage

FitMLVSBM$m_step(safeguard = 1e-06)

Arguments

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

Method ve_step()

Compute the VE step of the VEM algorithm

Usage

FitMLVSBM$ve_step(threshold = 1e-06, fixPointIter = 10, safeguard = 1e-06)

Arguments

• threshold: The convergence threshold

• fixPointIter: The maximum number of fixed point iterations

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

Method do_vem()

Launch a Variational EM algorithm

Usage

FitMLVSBM$do_vem(
  init = "hierarchical",
  threshold = 1e-06,
  maxIter = 1000,
  fixPointIter = 100,
  safeguard = 1e-06,
  Z = NULL
)

Arguments

• init: The method for self$init_clustering

• threshold: The convergence threshold

• maxIter: The max number of VEM iterations

• fixPointIter: The max number of fixed point iterations for VE step

• safeguard: Parameter live in a compact [safeguard, 1-safeguard]

• Z: Initial clustering if provided

Method permute_empty_class()

permute_empty_class Put empty blocks numbers at the end

Usage

FitMLVSBM$permute_empty_class()

Method plot()

Plot of FitMLVSBM objects

Usage

FitMLVSBM$plot(type = c("matrix"), ...)

Arguments

• type: A string for the type of plot, just "matrix" for now

Returns

a ggplot2 object

Method show()

print method

Usage

FitMLVSBM$show(type = "Multilevel Stochastic Block Model")

Arguments

• type: character to tune the displayed name

Method print()

print method

Usage

FitMLVSBM$print()

Method clone()

The objects of this class are cloneable with this method.

Usage

FitMLVSBM$clone(deep = FALSE)

Arguments

• deep: Whether to make a deep clone.