pkg/R/plot_valse.R

   1 utils::globalVariables(c("Var1","Var2","X1","X2","value")) #, package="valse")
   2 #' Plot
   3 #'
   4 #' It is a function which plots relevant parameters
   5 #'
   6 #' @param X matrix of covariates (of size n*p)
   7 #' @param Y matrix of responses (of size n*m)
   8 #' @param model the model constructed by valse procedure
   9 #' @param comp TRUE to enable pairwise clusters comparison
  10 #' @param k1 index of the first cluster to be compared
  11 #' @param k2 index of the second cluster to be compared
  12 #'
  13 #' @importFrom ggplot2 ggplot aes ggtitle geom_tile geom_line scale_fill_gradient2 geom_boxplot theme
  14 #' @importFrom cowplot background_grid
  15 #' @importFrom reshape2 melt
  16 #'
  17 #' @export
  18 plot_valse <- function(X, Y, model, comp = FALSE, k1 = NA, k2 = NA)
  19 {
  20   n <- nrow(X)
  21   K <- length(model$pi)
  22   ## regression matrices
  23   gReg <- list()
  24   for (r in 1:K) {
  25     Melt <- reshape2::melt(t((model$phi[, , r])))
  26     gReg[[r]] <- ggplot2::ggplot(data = Melt, ggplot2::aes(x = Var1, y = Var2, fill = value))  +
  27       ggplot2::geom_tile() + ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white",
  28       midpoint = 0, space = "Lab") + ggplot2::ggtitle(paste("Regression matrices in cluster", r))
  29   }
  30   print(gReg)
  31
  32   ## Differences between two clusters
  33   if (comp) {
  34     if (is.na(k1) || is.na(k2))
  35       print("k1 and k2 must be integers, representing the clusters you want to compare")
  36     Melt <- reshape2::melt(t(model$phi[, , k1] - model$phi[, , k2]))
  37     gDiff <- ggplot2::ggplot(data = Melt, ggplot2::aes(x = Var1, y = Var2, fill = value)) +
  38       ggplot2::geom_tile() + ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0,
  39         space = "Lab") + ggplot2::ggtitle(paste("Difference between regression matrices in cluster",
  40         k1, "and", k2))
  41     print(gDiff)
  42   }
  43
  44   ### Covariance matrices
  45   matCov <- matrix(NA, nrow = dim(model$rho[, , 1])[1], ncol = K)
  46   for (r in 1:K)
  47     matCov[, r] <- diag(model$rho[, , r])
  48   MeltCov <- reshape2::melt(matCov)
  49   gCov <- ggplot2::ggplot(data = MeltCov, ggplot2::aes(x = Var1, y = Var2, fill = value)) + ggplot2::geom_tile() +
  50     ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0,
  51       space = "Lab") + ggplot2::ggtitle("Covariance matrices (diag., one row per cluster)")
  52   print(gCov)
  53
  54   ### Proportions
  55   gam2 <- matrix(NA, ncol = 2, nrow = n)
  56   for (i in 1:n)
  57     gam2[i, ] <- c(model$proba[i, model$affec[i]], model$affec[i])
  58
  59   bp <- ggplot2::ggplot(data.frame(gam2), ggplot2::aes(x = X2, y = X1, color = X2, group = X2)) + ggplot2::geom_boxplot() +
  60      ggplot2::theme(legend.position = "none") + cowplot::background_grid(major = "xy", minor = "none")  +
  61     ggplot2::ggtitle("Assignment boxplot per cluster")
  62   print(bp)
  63 }