Merge branch 'feat_standalone_1.1_local' into 'develop'

Feat standalone 1.1 local See merge request joboog/LRB_shiny_app!2
joboog · Dec 3, 2017 · 1136bb2 · 1136bb2
2 parents fcdce67 + 44e8266
commit 1136bb2
Show file tree

Hide file tree

Showing 11 changed files with 259,585 additions and 105 deletions.
diff --git a/Code/Support.R b/Code/Support.R
@@ -0,0 +1,188 @@
+summary_table2 <- function(df,col_n, ByFactor, ByFactorCol){
+  # df: inout data frame containing numeric data and a factor to summarize by
+  # col_n: a numeric vector specifying the column numbers fo the numeric variables to subset the data frame
+  # ByfactorName: Name of the Factor by which to summarize
+  # ByFactorCol: Column number of the summarizing Factor
+
+  # Own summary table using 
+  #
+  library(dplyr)
+  # initialise function to count non NA values =========================================
+  countValues <- function(x){
+    as.integer(length(!is.na(x)[!is.na(x)==TRUE]))
+  }
+
+  # df <- lRaw_new2
+  # ByFactor <- "SamplePoint"
+  # ByFactorCol <- 1
+  # col_n <- 4:length(colnames(lRaw_new2))
+
+  #  compute summary stats =============================================================
+  df <- df[,c(ByFactorCol,col_n)]
+  # compute means values
+  means <- df %>% group_by_(ByFactor) %>% summarise_all(funs(mean), na.rm=TRUE)
+  means <- cbind(means[,1], round(select(means, -1), digits = 1))
+  means <- format(means, digits = 1, nsmall = 1)
+  colnames(means) <- c(ByFactor, colnames(select(means, -1)))
+
+  # compute standard deviations
+  sds <- df %>% group_by_(ByFactor) %>% summarise_all(funs(sd), na.rm=TRUE)
+  sds <- cbind(sds[,1], round(select(sds, -1), digits = 1))
+  sds <- format(sds, digits = 1, nsmall = 1)
+  colnames(sds) <- c(ByFactor, colnames(select(sds, -1)))
+  # set value of first columnof sds to NA, just to make the table look better in the end
+  sds[,1] <- NA
+
+  # count non NA values
+  counts <- df %>% group_by_(ByFactor) %>% summarise_all(funs(countValues))
+  counts <- format(as.data.frame(counts), digits = 0, nsmall = 0)
+  colnames(counts) <- c(ByFactor, colnames(select(counts, -1)))
+  counts[,1] <- NA
+
+
+
+  # creates a summary table with means values ==========================================
+  # standards deviations and counts
+
+  # create the data frame
+  df1 <- rbind(means[1,], sds[1,], counts[1,])
+
+  # if more than one SamplePoint, combine mean values, standard deviations, counts using a loop
+  if (length(means[,1])>1) {
+    for (i in 2:length(means[,1])){
+      df1 <- rbind(df1, means[i,], sds[i,], counts[i,])
+    }
+  }
+
+  # create a parameter vector which will be implemented in the summary table
+  Parameter <- rep(c("Mean", "StDev", "Count"), times=length(means[,1]))
+  df1 <- cbind(df1,Parameter)
+  # reorder df1, put parameters into 2nd column
+  df1 <- df1[,c(1,length(df1),3:length(df1)-1)]
+  # rename ByFactorColumn
+
+}
+
+
+# ===============================================================================
+summary_table2_old <- function(df,col_n, ByFactor, ByFactorCol){
+  # df: inout data frame containing numeric data and a factor to summarize by
+  # col_n: a numeric vector specifying the column numbers fo the numeric variables to subset the data frame
+  # ByfactorName: Name of the Factor by which to summarize
+  # ByFactorCol: Column number of the summarizing Factor
+
+  # Own summary table using 
+  #
+  library(dplyr)
+  # initialise function to count non NA values =========================================
+  countValues <- function(x){
+    as.integer(length(!is.na(x)[!is.na(x)==TRUE]))
+  }
+
+  # df <- lRaw_new2
+  # ByFactor <- "SamplePoint"
+  # ByFactorCol <- 1
+  # col_n <- 3:length(colnames(lRaw_new2))
+
+  #  compute summary stats =============================================================
+  # compute means values
+  means <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=mean, na.rm=TRUE, simplify = TRUE)
+  means <- cbind(means[,1], round(select(means, -ByFactor), digits = 1))  #means[,2:length(means[1,])],digits=1))
+  means <- format(means, digits = 1, nsmall = 1)
+  colnames(means) <- c(ByFactor, colnames(select(means, -1))) #(means)[2:length(means)])
+  # compute standard deviations
+  sds <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=sd, na.rm=TRUE, simplify = TRUE)
+  sds <- cbind(sds[,1], round(select(sds, -ByFactor),digits=1))
+  sds <- format(sds, digits = 1, nsmall = 1)
+  colnames(sds) <- c(ByFactor, colnames(select(sds, -1)))
+  # set value of first columnof sds to NA, just to make the table look better in the end
+  sds[,1] <- NA
+  # count no NA values
+  counts <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=countValues,  simplify = TRUE)
+  counts <- format(counts, digits = 0, nsmall = 0)
+  colnames(counts) <- c(ByFactor, colnames(select(counts, -1)))
+  counts[,1] <- NA
+
+
+
+  # creates a summary table with means values ==========================================
+  # standards deviations and counts
+
+  # create the data frame
+  df1 <- rbind(means[1,], sds[1,], counts[1,])
+
+  # if more than one SamplePoint, combine mean values, standard deviations, counts using a loop
+  if (length(means[,1])>1) {
+    for (i in 2:length(means[,1])){
+      df1 <- rbind(df1, means[i,], sds[i,], counts[i,])
+    }
+  }
+
+  # create a parameter vector which will be implemented in the summary table
+  Parameter <- rep(c("Mean", "StDev", "Count"), times=length(means[,1]))
+  df1 <- cbind(df1,Parameter)
+  # reorder df1, put parameters into 2nd column
+  df1 <- df1[,c(1,length(df1),3:length(df1)-1)]
+  # rename ByFactorColumn
+
+}
+
+
+# =================================================================================
+
+
+summary_table3 <- function(df,col_n, ByFactor, ByFactorCol){
+  # input: data frame containing numeric data and a factor to summarize by
+  # col_n: a vector specifying the column numbers to subset the data frame
+  # ByfactorName: Name of the Factor by which to summarize
+  # ByFactorCol: Column number of the summarizing Factor
+
+  # Own summary table using 
+  #
+  # initialise function to count values =========================================
+  countValues <- function(x){
+    as.integer(length(!is.na(x)[!is.na(x)==TRUE]))
+  }
+
+  # =============================================================================
+  # compute means values
+  means <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=mean, na.rm=TRUE, simplify = TRUE)
+  means <- cbind(means[,1], round(means[,2:length(means),1]))
+  means <- format(means, digits = 1, nsmall = 1)
+  colnames(means) <- c(ByFactor, colnames(means)[2:length(means)])
+  # compute standard deviations
+  sds <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=sd, na.rm=TRUE, simplify = TRUE)
+  sds <- cbind(sds[,1], round(sds[,2:length(sds)],1))
+  sds <- format(sds, digits = 1, nsmall = 1)
+  colnames(sds) <- c(ByFactor, colnames(sds)[2:length(sds)])
+  # set value of first columnof sds to NA, just to make the table look better in the end
+  sds[,1] <- NA
+  # count no NA values
+  counts <- aggregate(df[,col_n], by=list(ByFactor=df[,ByFactorCol]), FUN=countValues,  simplify = TRUE)
+  count <- format(counts, digits = 0, nsmall = 0)
+  colnames(counts) <- c(ByFactor, colnames(counts)[2:length(counts)])
+  counts[,1] <- NA
+  # ===============================================================================
+
+
+  # creates a summary table with means values, 
+  # standards deviations and counts
+  # ==========================================
+  # create the data frame
+  df1 <- rbind(means[1,], sds[1,], counts[1,])
+  # combine mean values, standard deviations, counts using a loop
+  for (i in 2:length(means[,1])){
+    df1 <- rbind(df1, means[i,], sds[i,], counts[i,])
+  }
+  # create a parameter vector which will be implemented in the summary table
+  Parameter <- rep(c("Mean", "StDev", "Count"), times=length(means[,1]))
+  df1 <- cbind(df1,Parameter)
+  # reorder df1, put parameters into 2nd column
+  df1 <- df1[,c(1,length(df1),3:length(df1)-1)]
+  # rename ByFactorColumn
+
+}
+# =================================================================================
+
+
+