Commit | Line | Data |
---|---|---|
ffdf9447 | 1 | #' valse |
086ca318 BA |
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 | |
43d76c49 | 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 | |
aa480ac1 BA |
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 | |
43d76c49 | 19 | #' @param thresh real, threshold to say a variable is relevant, by default = 1e-8 |
aa480ac1 BA |
20 | #' @param size_coll_mod (Maximum) size of a collection of models |
21 | #' @param fast TRUE to use compiled C code, FALSE for R code only | |
22 | #' @param verbose TRUE to show some execution traces | |
086ca318 BA |
23 | #' |
24 | #' @return a list with estimators of parameters | |
25 | #' | |
26 | #' @examples | |
27 | #' #TODO: a few examples | |
28 | #' @export | |
ffdf9447 BA |
29 | valse <- function(X, Y, procedure = "LassoMLE", selecMod = "DDSE", gamma = 1, mini = 10, |
30 | maxi = 50, eps = 1e-04, kmin = 2, kmax = 3, rank.min = 1, rank.max = 5, ncores_outer = 1, | |
31 | ncores_inner = 1, thresh = 1e-08, size_coll_mod = 10, fast = TRUE, verbose = FALSE, | |
32 | plot = TRUE) | |
33 | { | |
34 | p <- dim(X)[2] | |
35 | m <- dim(Y)[2] | |
36 | n <- dim(X)[1] | |
fb6e49cb | 37 | |
ffdf9447 | 38 | if (verbose) |
fb6e49cb | 39 | print("main loop: over all k and all lambda") |
40 | ||
41 | if (ncores_outer > 1) | |
42 | { | |
ffdf9447 BA |
43 | cl <- parallel::makeCluster(ncores_outer, outfile = "") |
44 | parallel::clusterExport(cl = cl, envir = environment(), varlist = c("X", | |
45 | "Y", "procedure", "selecMod", "gamma", "mini", "maxi", "eps", "kmin", | |
46 | "kmax", "rank.min", "rank.max", "ncores_outer", "ncores_inner", "thresh", | |
47 | "size_coll_mod", "verbose", "p", "m")) | |
fb6e49cb | 48 | } |
49 | ||
50 | # Compute models with k components | |
51 | computeModels <- function(k) | |
52 | { | |
ffdf9447 BA |
53 | if (ncores_outer > 1) |
54 | require("valse") #nodes start with an empty environment | |
fb6e49cb | 55 | |
ffdf9447 BA |
56 | if (verbose) |
57 | print(paste("Parameters initialization for k =", k)) | |
58 | # smallEM initializes parameters by k-means and regression model in each | |
59 | # component, doing this 20 times, and keeping the values maximizing the | |
60 | # likelihood after 10 iterations of the EM algorithm. | |
61 | P <- initSmallEM(k, X, Y) | |
62 | grid_lambda <- computeGridLambda(P$phiInit, P$rhoInit, P$piInit, P$gamInit, | |
63 | X, Y, gamma, mini, maxi, eps, fast) | |
64 | if (length(grid_lambda) > size_coll_mod) | |
65 | grid_lambda <- grid_lambda[seq(1, length(grid_lambda), length.out = size_coll_mod)] | |
fb6e49cb | 66 | |
ffdf9447 | 67 | if (verbose) |
fb6e49cb | 68 | print("Compute relevant parameters") |
ffdf9447 BA |
69 | # select variables according to each regularization parameter from the grid: |
70 | # S$selected corresponding to selected variables | |
71 | S <- selectVariables(P$phiInit, P$rhoInit, P$piInit, P$gamInit, mini, maxi, | |
72 | gamma, grid_lambda, X, Y, thresh, eps, ncores_inner, fast) | |
086cf723 | 73 | |
ffdf9447 | 74 | if (procedure == "LassoMLE") |
fb6e49cb | 75 | { |
ffdf9447 BA |
76 | if (verbose) |
77 | print("run the procedure Lasso-MLE") | |
78 | # compute parameter estimations, with the Maximum Likelihood Estimator, | |
79 | # restricted on selected variables. | |
80 | models <- constructionModelesLassoMLE(P$phiInit, P$rhoInit, P$piInit, | |
81 | P$gamInit, mini, maxi, gamma, X, Y, eps, S, ncores_inner, fast, verbose) | |
43d76c49 | 82 | |
ffdf9447 | 83 | } else |
fb6e49cb | 84 | { |
ffdf9447 BA |
85 | if (verbose) |
86 | print("run the procedure Lasso-Rank") | |
87 | # compute parameter estimations, with the Low Rank Estimator, restricted on | |
88 | # selected variables. | |
89 | models <- constructionModelesLassoRank(S, k, mini, maxi, X, Y, eps, rank.min, | |
90 | rank.max, ncores_inner, fast, verbose) | |
086ca318 | 91 | } |
ffdf9447 BA |
92 | # warning! Some models are NULL after running selectVariables |
93 | models <- models[sapply(models, function(cell) !is.null(cell))] | |
0eb161e3 | 94 | models |
086ca318 | 95 | } |
fb6e49cb | 96 | |
97 | # List (index k) of lists (index lambda) of models | |
ffdf9447 BA |
98 | models_list <- if (ncores_outer > 1) |
99 | parLapply(cl, kmin:kmax, computeModels) else lapply(kmin:kmax, computeModels) | |
100 | if (ncores_outer > 1) | |
fb6e49cb | 101 | parallel::stopCluster(cl) |
102 | ||
ffdf9447 | 103 | if (!requireNamespace("capushe", quietly = TRUE)) |
fb6e49cb | 104 | { |
105 | warning("'capushe' not available: returning all models") | |
ffdf9447 | 106 | return(models_list) |
fb6e49cb | 107 | } |
108 | ||
ffdf9447 BA |
109 | # Get summary 'tableauRecap' from models |
110 | tableauRecap <- do.call(rbind, lapply(seq_along(models_list), function(i) | |
111 | { | |
fb6e49cb | 112 | models <- models_list[[i]] |
ffdf9447 BA |
113 | # For a collection of models (same k, several lambda): |
114 | LLH <- sapply(models, function(model) model$llh[1]) | |
115 | k <- length(models[[1]]$pi) | |
116 | sumPen <- sapply(models, function(model) k * (dim(model$rho)[1] + sum(model$phi[, | |
117 | , 1] != 0) + 1) - 1) | |
118 | data.frame(model = paste(i, ".", seq_along(models), sep = ""), pen = sumPen/n, | |
119 | complexity = sumPen, contrast = -LLH) | |
120 | })) | |
fb6e49cb | 121 | |
122 | print(tableauRecap) | |
ffdf9447 | 123 | tableauRecap <- tableauRecap[which(tableauRecap[, 4] != Inf), ] |
fb6e49cb | 124 | |
ffdf9447 BA |
125 | modSel <- capushe::capushe(tableauRecap, n) |
126 | indModSel <- if (selecMod == "DDSE") | |
127 | as.numeric(modSel@DDSE@model) else if (selecMod == "Djump") | |
128 | as.numeric(modSel@Djump@model) else if (selecMod == "BIC") | |
129 | modSel@BIC_capushe$model else if (selecMod == "AIC") | |
7ac88d64 | 130 | modSel@AIC_capushe$model |
ffdf9447 BA |
131 | |
132 | mod <- as.character(tableauRecap[indModSel, 1]) | |
133 | listMod <- as.integer(unlist(strsplit(mod, "[.]"))) | |
134 | modelSel <- models_list[[listMod[1]]][[listMod[2]]] | |
135 | ||
136 | ## Affectations | |
137 | Gam <- matrix(0, ncol = length(modelSel$pi), nrow = n) | |
138 | for (i in 1:n) | |
139 | { | |
140 | for (r in 1:length(modelSel$pi)) | |
141 | { | |
142 | sqNorm2 <- sum((Y[i, ] %*% modelSel$rho[, , r] - X[i, ] %*% modelSel$phi[, | |
143 | , r])^2) | |
144 | Gam[i, r] <- modelSel$pi[r] * exp(-0.5 * sqNorm2) * det(modelSel$rho[, | |
145 | , r]) | |
7ac88d64 | 146 | } |
147 | } | |
ffdf9447 BA |
148 | Gam <- Gam/rowSums(Gam) |
149 | modelSel$affec <- apply(Gam, 1, which.max) | |
150 | modelSel$proba <- Gam | |
151 | ||
152 | if (plot) | |
153 | { | |
154 | print(plot_valse(X, Y, modelSel, n)) | |
7ac88d64 | 155 | } |
ffdf9447 | 156 | |
7ac88d64 | 157 | return(modelSel) |
086ca318 | 158 | } |