From 5dd5017cb64105d51a8536d53d1ca9a16239f9f7 Mon Sep 17 00:00:00 2001
From: Mia Croiset <mia.croiset@ens-lyon.fr>
Date: Mon, 24 Jun 2024 17:11:59 +0200
Subject: [PATCH] order process by category and pipeline order
---
benchmark/plots.r | 56 ++++++++++++++++++++++++++++++++++++++++++-----
1 file changed, 50 insertions(+), 6 deletions(-)
diff --git a/benchmark/plots.r b/benchmark/plots.r
index 4e6d6d1..28a0609 100644
--- a/benchmark/plots.r
+++ b/benchmark/plots.r
@@ -3,6 +3,7 @@ install.packages("ggplot2")
library(tidyverse)
library(ggplot2)
library(dplyr)
+library(forcats)
df <- read.csv("/home/mcroiset/HiC/benchmark/recap.txt", sep = "\t", header = TRUE)
@@ -79,7 +80,50 @@ df$file <- filenames
important_processes <- c(15, 16, 17, 18, 7, 1, 1, 1, 1, 1, 1, 1, 1,1,1,1,1,9,1,2,5,6,1,1,1,1,1,15,11,1,1,1,1,12,1,1,1,1,1)
-pdf("process_time.pdf")
+categories <- character(0)
+for (x in df$name) {
+ if (grepl('\\w*_ALIGN\\w*', x) || grepl('\\w*TRIM\\w*', x) || grepl('MERGE_BOWTIE2', x)) {
+ categories <- append(categories, "align")
+ }
+ else if (grepl('\\w*COOLER\\w*', x)) {
+ categories <- append(categories, "cooler")
+ }
+ else if (grepl('\\w*PAIRS\\w*', x)) {
+ categories<- append(categories, "pairs")
+ }
+ else if (grepl('\\w*MATRIX\\w*', x)) {
+ categories<- append(categories, "matrix")
+ }
+ else if (grepl('\\w*FILTER\\w*', x) || grepl('\\w*PICARD\\w*', x) || grepl('\\w*SAMTOOLS\\w*', x)) {
+ categories<- append(categories, "filter")
+ }
+ else if (grepl('\\w*CUTSITE\\w*', x)) {
+ categories<- append(categories, "cutsite")
+ }
+ else if (grepl('\\w*ITERALIGN\\w*', x)) {
+ categories<- append(categories, "iteralign")
+ }
+ else if (grepl('\\w*SAMPLESHEET\\w*', x) || (grepl('BOWTIE2_BUILD', x) || grepl('\\w*GETCHROM\\w*', x) || grepl('\\w*GET_RESTRIC\\w*', x))) {
+ categories<- append(categories, "data_prep")
+ }
+ else {
+ categories <- append(categories, "n")
+ }
+}
+# categories
+df$categorie <- categories
+
+print(levels(as.factor(df$name)))
+ordered_processes <- tibble(levels(as.factor(df$name)))
+order <- c(18, 7, 8, 7, 2, 20, 21, 14, 24, 23, 27, 22, 25, 30, 31, 32, 3, 6, 5, 19, 13, 19, 9, 4, 15, 28, 29, 18, 7, 8, 17, 16, 33, 11, 1, 10, 12, 27, 8)
+ordered_processes <- ordered_processes %>% add_column(order = order)
+ordered_processes <- rename(ordered_processes, name = `levels(as.factor(df$name))`)
+
+df <- left_join(df, ordered_processes, by= c("name" = "name"))
+
+
+
+# pdf("process_time.pdf")
ggplot(df, aes(x = name, y = duration_minutes)) +
geom_point(aes(color = name, shape = name)) +
@@ -91,15 +135,15 @@ ggplot(df, aes(x = name, y = duration_minutes)) +
theme(axis.text.x=element_blank(), axis.ticks.x=element_blank()) +
ggtitle("Duration time of each process")
-ggplot(df, aes(x = name, y = realtime_minutes)) +
- geom_point(aes(color = name, shape = name)) +
+ggplot(df, aes(x = fct_reorder(name, order), y = realtime_minutes)) +
+ geom_point(aes(color = categorie)) +
scale_shape_manual(values = important_processes) +
facet_wrap(~ file, ncol = 8) +
- scale_y_continuous(breaks = seq(0, 490, by = 30)) +
xlab("Processes") +
ylab("Duration in minutes") +
theme(axis.text.x=element_blank(), axis.ticks.x=element_blank())+
- ggtitle("Real execution time of each process")
+ ggtitle("Real execution time of each process") +
+ scale_y_log10()
-dev.off()
+# dev.off()
--
GitLab