clean R code

Author: zhizuio

R/plotBrier.R

@@ -218,17 +218,17 @@ plotBrier <- function(dat, datMCMC,
   }
 
   list.models <- list(
-    "Cox.clin" = pred.fitCox.clin,
-    "Cox.X.mean" = pred.fitCox.X.mean,
-    "Cox.X.median" = pred.fitCox.X.median,
-    # "Cox.clin.X.median"=fitCox.clin.X.median,
-    "Cox.clin.X.mean" = pred.fitCox.clin.X.mean,
-    # "PTCM.clin" = predPTCM.prob,
+    "Cox-X0" = pred.fitCox.clin,
+    "Cox-Xmean" = pred.fitCox.X.mean,
+    "Cox-Xmedian" = pred.fitCox.X.median,
+    # "Cox-X0-Xmedian"=fitCox.clin.X.median,
+    "Cox-X0Xmean" = pred.fitCox.clin.X.mean,
+    # "PTCM-X0" = predPTCM.prob,
     # "GPTCM-BetaBin" = pred.prob2,
     "GPTCM" = pred.prob
   )
   if (PTCM) {
-    list.models <- c(list.models, list("GPTCM-PTCM.clin" = predPTCM.prob))
+    list.models <- c(list.models, list("PTCM-X0" = predPTCM.prob))
   }
   g <- riskRegression::Score(
     list.models,
@@ -242,10 +242,7 @@ plotBrier <- function(dat, datMCMC,
     data <- data[data$times <= time.star, ]
   }
   levels(data$model)[1] <- "Kaplan-Meier"
-  # utils::globalVariables(c("times", "Brier", "model"))
-  # NOTE: `aes_string()` was deprecated in ggplot2 3.0.0.
   g2 <- ggplot2::ggplot(data, aes(
-    # x = "times", y = "Brier", group = "model", color = "model"
     x = .data$times, y = .data$Brier, group = .data$model, color = .data$model
   )) +
     xlab(xlab) +

R/plotCoeff.R

@@ -59,11 +59,10 @@ plotCoeff <- function(dat,
                       label.y = NULL,
                       first.coef = NULL,
                       y.axis.size = 8, ...) {
-  # n <- dim(dat$X)[1]
+
   p <- dim(dat$X)[2]
   L <- dim(dat$X)[3]
   thin <- datMCMC$input$thin
-  # nIter <- datMCMC$input$nIter / thin
   burnin <- datMCMC$input$burnin / thin + 1
 
   xlab0 <- xlab
@@ -78,8 +77,6 @@ plotCoeff <- function(dat,
     xlim.upper <- xlim[2]
   }
 
-  # legend0 <- ifelse(is.null(legend.labs), 0, 1)
-
   if (estimator == "beta") {
     betas.mcmc <- datMCMC$output$betas[-c(1:burnin), ]
     betas.true <- dat$betas

R/plotMCMC.R

@@ -41,19 +41,14 @@ plotMCMC <- function(dat, datMCMC, estimator = "xi") {
   oldpar <- par(no.readonly = TRUE)
   on.exit(par(oldpar))
 
-  # n <- dim(dat$X)[1]
   p <- dim(dat$X)[2]
   L <- dim(dat$X)[3]
-  # nIter <- datMCMC$input$nIter
-  # burnin <- datMCMC$input$burnin
-  # p.idx <- 1:(p * L)
+  
+  # set covariates' indexes for plotting
   p.idx <- 1:((p + 1) * L)
-  # p.DirTrue.idx <- 1:((p + 1) * L)
   p.DirFalse.idx <- 1:((p + 1) * (L - 1))
   if (p > 10) {
-    # p.idx <- rep(1:10, L) + p * rep(0:(L - 1), each = 10)
     p.idx <- rep(1:11, L) + (p + 1) * rep(0:(L - 1), each = 11)
-    # p.DirTrue.idx <- rep(1:11, L) + (p + 1) * rep(0:(L - 1), each = 11)
     p.DirFalse.idx <- rep(1:11, L - 1) + (p + 1) * rep(0:(L - 2), each = 11)
     p <- 10
   }

R/predict.GPTCM.R

@@ -39,7 +39,6 @@ predict.GPTCM <- function(object, dat,
                           type = "survival",
                           times = NULL, ...) {
   n <- dim(dat$X)[1]
-  # p <- dim(dat$X)[2]
   L <- dim(dat$X)[3]
 
   if (is.null(newdata)) {
@@ -48,7 +47,6 @@ predict.GPTCM <- function(object, dat,
     survObj.new <- newdata$survObj
   }
 
-  # nIter <- object$input$nIter
   burnin <- object$input$burnin / object$input$thin
 
   # survival predictions based on posterior mean