plot per condition, process in x time in y

benchmark/plots.r

@@ -54,11 +54,42 @@ df$duration_minutes <- as.numeric(df$duration_minutes)
 
 val <- df %>% group_by(name) %>% summarise(moy = mean(duration_minutes), ect = sd(duration_minutes))
 
-ggplot(df, aes(x=name, y=duration_minutes))+
-  geom_violin()+
-  ylim(0,500)+
-  facet_wrap(~ name, ncol = 8)
+workflow <- character(0)
+for (x in df$file) {
+  full <- sapply(strsplit(x, split = "/"), "[", 2)
+  w <- sapply(strsplit(full, split = "_"), "[", 1)
+  workflow <- append(workflow,w)
+}
+df$workflow <- workflow
+
+processes <- character(0)
+for (x in df$name) {
+  process <- lapply(strsplit(x, split = ":"), tail, n = 1L)  
+  processes <- append(processes, as.character(process))
+}
+df$name <- processes
+
+filenames <- character(0)
+for (x in df$file) {
+  file_short_name <- sapply(strsplit(x, split = "/"), "[", 2)
+  filenames <- append(filenames,file_short_name)
+}
+df$file <- filenames
+
+
+
+
+# ggplot(df, aes(x=name, y=duration_minutes))+
+#   geom_violin()+
+#   ylim(0,500)+
+#   facet_wrap(~ name, ncol = 8)
+
+important_processes <- c(15, 16, 17, 18, 7, 1, 1, 1, 1, 1, 1, 1, 1,1,1,1,1,9,1,2,5,6,1,1,1,1,1,1,11,1,1,1,1,12,1,1,1,1,1)
 
+ggplot(df, aes(x = name, y = duration_minutes))+
+  geom_point(aes(color = name, shape = name))+
+  scale_shape_manual(values = important_processes)+
+  facet_wrap(~ file, ncol = 8)
 
 # for (x in df$duration_minutes) {
 #   print(class(x))