pkg/R/main.R

   1 #' runValse
   2 #'
   3 #' Main function
   4 #'
   5 #' @param X matrix of covariates (of size n*p)
   6 #' @param Y matrix of responses (of size n*m)
   7 #' @param procedure among 'LassoMLE' or 'LassoRank'
   8 #' @param selecMod method to select a model among 'DDSE', 'DJump', 'BIC' or 'AIC'
   9 #' @param gamma integer for the power in the penaly, by default = 1
  10 #' @param mini integer, minimum number of iterations in the EM algorithm, by default = 10
  11 #' @param maxi integer, maximum number of iterations in the EM algorithm, by default = 100
  12 #' @param eps real, threshold to say the EM algorithm converges, by default = 1e-4
  13 #' @param kmin integer, minimum number of clusters, by default = 2
  14 #' @param kmax integer, maximum number of clusters, by default = 10
  15 #' @param rank.min integer, minimum rank in the low rank procedure, by default = 1
  16 #' @param rank.max integer, maximum rank in the low rank procedure, by default = 5
  17 #' @param ncores_outer Number of cores for the outer loop on k
  18 #' @param ncores_inner Number of cores for the inner loop on lambda
  19 #' @param thresh real, threshold to say a variable is relevant, by default = 1e-8
  20 #' @param grid_lambda, a vector with regularization parameters if known, by default numeric(0)
  21 #' @param size_coll_mod (Maximum) size of a collection of models
  22 #' @param fast TRUE to use compiled C code, FALSE for R code only
  23 #' @param verbose TRUE to show some execution traces
  24 #' @param plot TRUE to plot the selected models after run
  25 #'
  26 #' @return
  27 #' The selected model if enough data are available to estimate it,
  28 #' or a list of models otherwise.
  29 #'
  30 #' @examples
  31 #' n = 50; m = 10; p = 5
  32 #' beta = array(0, dim=c(p,m,2))
  33 #' beta[,,1] = 1
  34 #' beta[,,2] = 2
  35 #' data = generateXY(n, c(0.4,0.6), rep(0,p), beta, diag(0.5, p), diag(0.5, m))
  36 #' X = data$X
  37 #' Y = data$Y
  38 #' res = runValse(X, Y)
  39 #' X <- matrix(runif(100), nrow=50)
  40 #' Y <- matrix(runif(100), nrow=50)
  41 #' res = runValse(X, Y)
  42 #'
  43 #' @export
  44 runValse <- function(X, Y, procedure = "LassoMLE", selecMod = "DDSE", gamma = 1, mini = 10,
  45   maxi = 50, eps = 1e-04, kmin = 2, kmax = 3, rank.min = 1, rank.max = 5, ncores_outer = 1,
  46   ncores_inner = 1, thresh = 1e-08, grid_lambda = numeric(0), size_coll_mod = 10,
  47   fast = TRUE, verbose = FALSE, plot = TRUE)
  48 {
  49   n <- nrow(X)
  50   p <- ncol(X)
  51   m <- ncol(Y)
  52
  53   if (verbose)
  54     print("main loop: over all k and all lambda")
  55
  56   if (ncores_outer > 1) {
  57     cl <- parallel::makeCluster(ncores_outer, outfile = "")
  58     parallel::clusterExport(cl = cl, envir = environment(), varlist = c("X",
  59       "Y", "procedure", "selecMod", "gamma", "mini", "maxi", "eps", "kmin",
  60       "kmax", "rank.min", "rank.max", "ncores_outer", "ncores_inner", "thresh",
  61       "size_coll_mod", "verbose", "p", "m"))
  62   }
  63
  64   # Compute models with k components
  65   computeModels <- function(k)
  66   {
  67     if (ncores_outer > 1)
  68       require("valse") #nodes start with an empty environment
  69
  70     if (verbose)
  71       print(paste("Parameters initialization for k =", k))
  72     # smallEM initializes parameters by k-means and regression model in each
  73     # component, doing this 20 times, and keeping the values maximizing the
  74     # likelihood after 10 iterations of the EM algorithm.
  75     P <- initSmallEM(k, X, Y, fast)
  76     if (length(grid_lambda) == 0)
  77     {
  78       grid_lambda <- computeGridLambda(P$phiInit, P$rhoInit, P$piInit, P$gamInit,
  79                                        X, Y, gamma, mini, maxi, eps, fast)
  80     }
  81     if (length(grid_lambda) > size_coll_mod)
  82       grid_lambda <- grid_lambda[seq(1, length(grid_lambda), length.out = size_coll_mod)]
  83
  84     if (verbose)
  85       print("Compute relevant parameters")
  86     # select variables according to each regularization parameter from the grid:
  87     # S$selected corresponding to selected variables
  88     S <- selectVariables(P$phiInit, P$rhoInit, P$piInit, P$gamInit, mini, maxi,
  89       gamma, grid_lambda, X, Y, thresh, eps, ncores_inner, fast)
  90
  91     if (procedure == "LassoMLE") {
  92       if (verbose)
  93         print("run the procedure Lasso-MLE")
  94       # compute parameter estimations, with the Maximum Likelihood Estimator,
  95       # restricted on selected variables.
  96       models <- constructionModelesLassoMLE(P$phiInit, P$rhoInit, P$piInit,
  97         P$gamInit, mini, maxi, gamma, X, Y, eps, S, ncores_inner, fast, verbose)
  98     } else {
  99       if (verbose)
 100         print("run the procedure Lasso-Rank")
 101       # compute parameter estimations, with the Low Rank Estimator, restricted on
 102       # selected variables.
 103       models <- constructionModelesLassoRank(S, k, mini, maxi, X, Y, eps, rank.min,
 104         rank.max, ncores_inner, fast, verbose)
 105     }
 106     # warning! Some models are NULL after running selectVariables
 107     models <- models[sapply(models, function(cell) !is.null(cell))]
 108     models
 109   }
 110
 111   # List (index k) of lists (index lambda) of models
 112   models_list <-
 113     if (ncores_outer > 1) {
 114       parLapply(cl, kmin:kmax, computeModels)
 115     } else {
 116       lapply(kmin:kmax, computeModels)
 117     }
 118   if (ncores_outer > 1)
 119     parallel::stopCluster(cl)
 120
 121   if (!requireNamespace("capushe", quietly = TRUE))
 122   {
 123     warning("'capushe' not available: returning all models")
 124     return(models_list)
 125   }
 126
 127   # Get summary 'tableauRecap' from models
 128   tableauRecap <- do.call(rbind, lapply(seq_along(models_list), function(i)
 129   {
 130     models <- models_list[[i]]
 131     # For a collection of models (same k, several lambda):
 132     LLH <- sapply(models, function(model) model$llh[1])
 133     k <- length(models[[1]]$pi)
 134     sumPen <- sapply(models, function(model) k * (dim(model$rho)[1] + sum(model$phi[,
 135       , 1] != 0) + 1) - 1)
 136     data.frame(model = paste(i, ".", seq_along(models), sep = ""), pen = sumPen/n,
 137       complexity = sumPen, contrast = -LLH)
 138   }))
 139   tableauRecap <- tableauRecap[which(tableauRecap[, 4] != Inf), ]
 140   if (verbose)
 141     print(tableauRecap)
 142
 143   if (nrow(tableauRecap) > 10) {
 144     modSel <- capushe::capushe(tableauRecap, n)
 145     indModSel <- if (selecMod == "DDSE")
 146     {
 147       as.numeric(modSel@DDSE@model)
 148     } else if (selecMod == "Djump")
 149     {
 150       as.numeric(modSel@Djump@model)
 151     } else if (selecMod == "BIC")
 152     {
 153       modSel@BIC_capushe$model
 154     } else if (selecMod == "AIC")
 155     {
 156       modSel@AIC_capushe$model
 157     }
 158     listMod <- as.integer(unlist(strsplit(as.character(indModSel), "[.]")))
 159     modelSel <- models_list[[listMod[1]]][[listMod[2]]]
 160     modelSel$models <- tableauRecap
 161
 162     if (plot)
 163       print(plot_valse(X, Y, modelSel))
 164     return(modelSel)
 165   }
 166   tableauRecap
 167 }