pkg/R/plot_valse.R

   1 utils::globalVariables(c("Var1","Var2","X1","X2","value")) #, package="valse")
   2
   3 #' Plot
   4 #'
   5 #' A function which plots relevant parameters.
   6 #'
   7 #' @param X matrix of covariates (of size n*p)
   8 #' @param Y matrix of responses (of size n*m)
   9 #' @param model the model constructed by valse procedure
  10 #' @param comp TRUE to enable pairwise clusters comparison
  11 #' @param k1 index of the first cluster to be compared
  12 #' @param k2 index of the second cluster to be compared
  13 #'
  14 #' @importFrom ggplot2 ggplot aes ggtitle geom_tile geom_line scale_fill_gradient2 geom_boxplot theme
  15 #' @importFrom cowplot background_grid
  16 #' @importFrom reshape2 melt
  17 #'
  18 #' @return No return value (only plotting).
  19 #'
  20 #' @export
  21 plot_valse <- function(X, Y, model, comp = FALSE, k1 = NA, k2 = NA)
  22 {
  23   n <- nrow(X)
  24   K <- length(model$pi)
  25   ## regression matrices
  26   gReg <- list()
  27   for (r in 1:K) {
  28     Melt <- reshape2::melt(t((model$phi[, , r])))
  29     gReg[[r]] <- ggplot2::ggplot(data = Melt, ggplot2::aes(x = Var1, y = Var2, fill = value))  +
  30       ggplot2::geom_tile() + ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white",
  31       midpoint = 0, space = "Lab") + ggplot2::ggtitle(paste("Regression matrices in cluster", r))
  32   }
  33   print(gReg)
  34
  35   ## Differences between two clusters
  36   if (comp) {
  37     if (is.na(k1) || is.na(k2))
  38       print("k1 and k2 must be integers, representing the clusters you want to compare")
  39     Melt <- reshape2::melt(t(model$phi[, , k1] - model$phi[, , k2]))
  40     gDiff <- ggplot2::ggplot(data = Melt, ggplot2::aes(x = Var1, y = Var2, fill = value)) +
  41       ggplot2::geom_tile() + ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0,
  42         space = "Lab") + ggplot2::ggtitle(paste("Difference between regression matrices in cluster",
  43         k1, "and", k2))
  44     print(gDiff)
  45   }
  46
  47   ### Covariance matrices
  48   matCov <- matrix(NA, nrow = dim(model$rho[, , 1])[1], ncol = K)
  49   for (r in 1:K)
  50     matCov[, r] <- diag(model$rho[, , r])
  51   MeltCov <- reshape2::melt(matCov)
  52   gCov <- ggplot2::ggplot(data = MeltCov, ggplot2::aes(x = Var1, y = Var2, fill = value)) + ggplot2::geom_tile() +
  53     ggplot2::scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0,
  54       space = "Lab") + ggplot2::ggtitle("Covariance matrices (diag., one row per cluster)")
  55   print(gCov)
  56
  57   ### Proportions
  58   gam2 <- matrix(NA, ncol = 2, nrow = n)
  59   for (i in 1:n)
  60     gam2[i, ] <- c(model$proba[i, model$affec[i]], model$affec[i])
  61
  62   bp <- ggplot2::ggplot(data.frame(gam2), ggplot2::aes(x = X2, y = X1, color = X2, group = X2)) + ggplot2::geom_boxplot() +
  63      ggplot2::theme(legend.position = "none") + cowplot::background_grid(major = "xy", minor = "none")  +
  64     ggplot2::ggtitle("Assignment boxplot per cluster")
  65   print(bp)
  66 }