Ticket ## : Close case instance + test human task

Author: habarthierry-hue

src/CaseManagement.CMMN.Benchmark/BenchmarkDotNet.Artifacts/results/BuildPlots.R

@@ -0,0 +1,184 @@
+BenchmarkDotNetVersion <- "BenchmarkDotNet v0.12.0 "
+dir.create(Sys.getenv("R_LIBS_USER"), recursive = TRUE, showWarnings = FALSE)
+list.of.packages <- c("ggplot2", "dplyr", "gdata", "tidyr", "grid", "gridExtra", "Rcpp")
+new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
+if(length(new.packages)) install.packages(new.packages, lib = Sys.getenv("R_LIBS_USER"), repos = "https://cran.rstudio.com/")
+library(ggplot2)
+library(dplyr)
+library(gdata)
+library(tidyr)
+library(grid)
+library(gridExtra)
+
+isEmpty <- function(val){
+   is.null(val) | val == ""
+}
+
+createPrefix <- function(params){ 
+   separator <- "-"
+   values <- params[!isEmpty(params)]
+   paste(replace(values, TRUE, paste0(separator, values)), collapse = "")
+}
+
+ends_with <- function(vars, match, ignore.case = TRUE) {
+  if (ignore.case)
+    match <- tolower(match)
+  n <- nchar(match)
+
+  if (ignore.case)
+    vars <- tolower(vars)
+  length <- nchar(vars)
+
+  substr(vars, pmax(1, length - n + 1), length) == match
+}
+std.error <- function(x) sqrt(var(x)/length(x))
+cummean <- function(x) cumsum(x)/(1:length(x))
+BenchmarkDotNetVersionGrob <- textGrob(BenchmarkDotNetVersion, gp = gpar(fontface=3, fontsize=10), hjust=1, x=1)
+nicePlot <- function(p) grid.arrange(p, bottom = BenchmarkDotNetVersionGrob)
+printNice <- function(p) print(nicePlot(p))
+ggsaveNice <- function(fileName, p, ...) {
+  cat(paste0("*** Plot: ", fileName, " ***\n"))
+  ggsave(fileName, plot = nicePlot(p), ...)
+  cat("------------------------------\n")
+}
+
+args <- commandArgs(trailingOnly = TRUE)
+files <- if (length(args) > 0) args else list.files()[list.files() %>% ends_with("-measurements.csv")]
+for (file in files) {
+  title <- gsub("-measurements.csv", "", basename(file))
+  measurements <- read.csv(file, sep = ";")
+
+  result <- measurements %>% filter(Measurement_IterationStage == "Result")
+  if (nrow(result[is.na(result$Job_Id),]) > 0)
+    result[is.na(result$Job_Id),]$Job_Id <- ""
+  if (nrow(result[is.na(result$Params),]) > 0) {
+    result[is.na(result$Params),]$Params <- ""
+  } else {
+    result$Job_Id <- trim(paste(result$Job_Id, result$Params))
+  }
+  result$Job_Id <- factor(result$Job_Id, levels = unique(result$Job_Id))
+
+  timeUnit <- "ns"
+  if (min(result$Measurement_Value) > 1000) {
+    result$Measurement_Value <- result$Measurement_Value / 1000
+    timeUnit <- "us"
+  }
+  if (min(result$Measurement_Value) > 1000) {
+    result$Measurement_Value <- result$Measurement_Value / 1000
+    timeUnit <- "ms"
+  }
+  if (min(result$Measurement_Value) > 1000) {
+    result$Measurement_Value <- result$Measurement_Value / 1000
+    timeUnit <- "sec"
+  }
+
+  resultStats <- result %>%
+    group_by_(.dots = c("Target_Method", "Job_Id")) %>%
+    summarise(se = std.error(Measurement_Value), Value = mean(Measurement_Value))
+
+  benchmarkBoxplot <- ggplot(result, aes(x=Target_Method, y=Measurement_Value, fill=Job_Id)) +
+    guides(fill=guide_legend(title="Job")) +
+    xlab("Target") +
+    ylab(paste("Time,", timeUnit)) +
+    ggtitle(title) +
+    geom_boxplot()
+  benchmarkBarplot <- ggplot(resultStats, aes(x=Target_Method, y=Value, fill=Job_Id)) +
+    guides(fill=guide_legend(title="Job")) +
+    xlab("Target") +
+    ylab(paste("Time,", timeUnit)) +
+    ggtitle(title) +
+    geom_bar(position=position_dodge(), stat="identity")
+    #geom_errorbar(aes(ymin=Value-1.96*se, ymax=Value+1.96*se), width=.2, position=position_dodge(.9))
+
+  printNice(benchmarkBoxplot)
+  printNice(benchmarkBarplot)
+  ggsaveNice(gsub("-measurements.csv", "-boxplot.png", file), benchmarkBoxplot)
+  ggsaveNice(gsub("-measurements.csv", "-barplot.png", file), benchmarkBarplot)
+
+  for (target in unique(result$Target_Method)) {
+    df <- result %>% filter(Target_Method == target)
+    df$Launch <- factor(df$Measurement_LaunchIndex)
+    df <- df %>% group_by(Job_Id, Launch) %>% mutate(cm = cummean(Measurement_Value))
+
+    densityPlot <- ggplot(df, aes(x=Measurement_Value, fill=Job_Id)) +
+      ggtitle(paste(title, "/", target)) +
+      xlab(paste("Time,", timeUnit)) +
+      geom_density(alpha=.5, bw="SJ")
+    printNice(densityPlot)
+    ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-density.png"), file), densityPlot)
+
+    facetDensityPlot <- densityPlot + facet_wrap(~Job_Id)
+    printNice(facetDensityPlot)
+    ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-facetDensity.png"), file), facetDensityPlot)
+
+    for (params in unique(df$Params)) {
+      paramsDf <- df %>% filter(Params == params)
+      paramsDensityPlot <- ggplot(paramsDf, aes(x=Measurement_Value, fill=Job_Id)) +
+        ggtitle(paste(title, "/", target, "/", params)) +
+        xlab(paste("Time,", timeUnit)) +
+        geom_density(alpha=.5, bw="SJ")
+      printNice(paramsDensityPlot)
+      prefix <- createPrefix(c(target,params))
+      ggsaveNice(gsub("-measurements.csv", paste0(prefix, "-density.png"), file), paramsDensityPlot)
+
+      paramsFacetDensityPlot <- paramsDensityPlot + facet_wrap(~Job_Id)
+      printNice(paramsFacetDensityPlot)
+      ggsaveNice(gsub("-measurements.csv", paste0(prefix, "-facetDensity.png"), file), paramsFacetDensityPlot)
+    }
+
+    for (job in unique(df$Job_Id)) {
+      jobDf <- df %>% filter(Job_Id == job)
+      timelinePlot <- ggplot(jobDf, aes(x = Measurement_IterationIndex, y=Measurement_Value, group=Launch, color=Launch)) +
+        ggtitle(paste(title, "/", target, "/", job)) +
+        xlab("IterationIndex") +
+        ylab(paste("Time,", timeUnit)) +
+        geom_line() +
+        geom_point()
+      printNice(timelinePlot)
+      ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-", job, "-timeline.png"), file), timelinePlot)
+      timelinePlotSmooth <- timelinePlot + geom_smooth()
+      printNice(timelinePlotSmooth)
+      ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-", job, "-timelineSmooth.png"), file), timelinePlotSmooth)
+
+      cummeanPlot <- ggplot(jobDf, aes(x = Measurement_IterationIndex, y=cm, group=Launch, color=Launch)) +
+        ggtitle(paste(title, "/", target, "/", job)) +
+        xlab("IterationIndex") +
+        ylab(paste("Cumulative mean time,", timeUnit)) +
+        geom_line() +
+        geom_point()
+      printNice(cummeanPlot)
+      ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-", job, "-cummean.png"), file), cummeanPlot)
+
+
+      densityPlotJob <- ggplot(jobDf, aes(x=Measurement_Value, fill="red")) +
+        ggtitle(paste(title, "/", target, "/", job)) +
+        xlab(paste("Time,", timeUnit)) +
+        geom_density(alpha=.5, bw="SJ")
+      printNice(densityPlotJob)
+      ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-", job, "-density.png"), file), densityPlotJob)
+    }
+
+    timelinePlot <- ggplot(df, aes(x = Measurement_IterationIndex, y=Measurement_Value, group=Launch, color=Launch)) +
+      ggtitle(paste(title, "/", target)) +
+      xlab("IterationIndex") +
+      ylab(paste("Time,", timeUnit)) +
+      geom_line() +
+      geom_point() +
+      facet_wrap(~Job_Id)
+    printNice(timelinePlot)
+    ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-facetTimeline.png"), file), timelinePlot)
+    timelinePlotSmooth <- timelinePlot + geom_smooth()
+    printNice(timelinePlotSmooth)
+    ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-facetTimelineSmooth.png"), file), timelinePlotSmooth)
+
+    facetCummeanPlot <- ggplot(df, aes(x = Measurement_IterationIndex, y=cm, group=Launch, color=Launch)) +
+      ggtitle(paste(title, "/", target)) +
+      xlab("IterationIndex") +
+      ylab(paste("Cumulative mean time,", timeUnit)) +
+      geom_line() +
+      geom_point() +
+      facet_wrap(~Job_Id)
+    printNice(facetCummeanPlot)
+    ggsaveNice(gsub("-measurements.csv", paste0("-", target, "-cummean.png"), file), facetCummeanPlot)
+  }
+}