From 63aa07dad518235894014367e4fdd6115550d749 Mon Sep 17 00:00:00 2001
From: Rodrigo Arias Mallo <rodrigo.arias@bsc.es>
Date: Tue, 6 Apr 2021 18:40:19 +0200
Subject: [PATCH] heat: update granularity plot with modern ggplot

---
 garlic/fig/heat/granul.R | 151 +++++++++++++--------------------------
 1 file changed, 49 insertions(+), 102 deletions(-)

diff --git a/garlic/fig/heat/granul.R b/garlic/fig/heat/granul.R
index d3dc5d7..fb72e00 100644
--- a/garlic/fig/heat/granul.R
+++ b/garlic/fig/heat/granul.R
@@ -1,120 +1,67 @@
 library(ggplot2)
-library(dplyr)
+library(dplyr, warn.conflicts = FALSE)
 library(scales)
 library(jsonlite)
+library(viridis, warn.conflicts = FALSE)
+library(stringr)
 
-args=commandArgs(trailingOnly=TRUE)
+args = commandArgs(trailingOnly=TRUE)
 
-# Read the timetable from args[1]
-input_file = "input.json"
-if (length(args)>0) { input_file = args[1] }
+# Set the input dataset if given in argv[1], or use "input" as default
+if (length(args)>0) { input_file = args[1] } else { input_file = "input" }
 
-# Load the dataset in NDJSON format
-dataset = jsonlite::stream_in(file(input_file)) %>%
-	jsonlite::flatten()
+df = jsonlite::stream_in(file(input_file), verbose=FALSE) %>%
 
+  jsonlite::flatten() %>%
 
-# We only need the nblocks and time
-df = select(dataset, config.cbs, config.rbs, time) %>%
-	rename(cbs=config.cbs, rbs=config.rbs)
+  select(unit,
+    config.cbs,
+    config.rbs,
+    time,
+    total_time) %>%
 
-df$cbs = as.factor(df$cbs)
-df$rbs = as.factor(df$rbs)
+  rename(cbs=config.cbs,
+    rbs=config.rbs) %>%
 
-# Normalize the time by the median
-df=group_by(df, cbs, rbs) %>%
-	mutate(mtime = median(time)) %>%
-	mutate(tnorm = time / mtime - 1) %>%
-	mutate(logmtime = log(mtime)) %>%
-  ungroup() %>%
-  filter(between(mtime, mean(time) - (1 * sd(time)),
-    mean(time) + (1 * sd(time))))
+  # Convert to factors
+  mutate(cbs = as.factor(cbs)) %>%
+  mutate(rbs = as.factor(rbs)) %>%
+  mutate(unit = as.factor(unit)) %>%
 
-ppi=300
-h=5
-w=5
+  # Compute median times
+  group_by(unit) %>%
+  mutate(median.time = median(time)) %>%
+  mutate(normalized.time = time / median.time - 1) %>%
+  mutate(log.median.time = log(median.time)) %>%
+  ungroup()
 
-png("box.png", width=w*ppi, height=h*ppi, res=ppi)
-#
-#
-#
-# Create the plot with the normalized time vs nblocks
-p = ggplot(data=df, aes(x=cbs, y=tnorm)) +
+dpi = 300
+h = 6
+w = 6
 
-	# Labels
-	labs(x="cbs", y="Normalized time",
-              title=sprintf("Heat normalized time"), 
-              subtitle=input_file) +
+# ---------------------------------------------------------------------
 
-	# Center the title
-	#theme(plot.title = element_text(hjust = 0.5)) +
+p = ggplot(df, aes(x=cbs, y=normalized.time)) +
+  geom_boxplot() +
+  geom_hline(yintercept=c(-0.01, 0.01), linetype="dashed", color="red") +
+  theme_bw() +
+  labs(y="Normalized time",
+       title="Heat granularity: normalized time", 
+    subtitle=input_file) + 
+  theme(plot.subtitle=element_text(size=8))
 
-	# Black and white mode (useful for printing)
-	#theme_bw() +
+ggsave("normalized.time.png", plot=p, width=w, height=h, dpi=dpi)
+ggsave("normalized.time.pdf", plot=p, width=w, height=h, dpi=dpi)
 
-	# Add the maximum allowed error lines
-	geom_hline(yintercept=c(-0.01, 0.01),
-		linetype="dashed", color="red") +
+# ---------------------------------------------------------------------
 
-	# Draw boxplots
-	geom_boxplot() +
+p = ggplot(df, aes(x=cbs, y=time)) +
+  geom_point(shape=21, size=3) +
+  geom_line(aes(y=median.time, group=0)) +
+  theme_bw() +
+  labs(y="Time (s)", title="Heat granularity: time", 
+    subtitle=input_file) + 
+  theme(plot.subtitle=element_text(size=8))
 
-	#scale_y_continuous(breaks = scales::pretty_breaks(n = 10)) +
-
-	theme_bw() +
-
-	theme(plot.subtitle=element_text(size=8)) +
-
-	theme(legend.position = c(0.85, 0.85)) #+
-
-
-
-
-# Render the plot
-print(p)
-
-## Save the png image
-dev.off()
-#
-png("scatter.png", width=w*ppi, height=h*ppi, res=ppi)
-#
-## Create the plot with the normalized time vs nblocks
-p = ggplot(df, aes(x=cbs, y=time, linetype=rbs, group=rbs)) +
-
-	labs(x="cbs", y="Time (s)",
-              title=sprintf("Heat granularity"), 
-              subtitle=input_file) +
-	theme_bw() +
-	theme(plot.subtitle=element_text(size=8)) +
-	theme(legend.position = c(0.5, 0.88)) +
-
-	geom_point(shape=21, size=3) +
-	geom_line(aes(y=mtime)) +
-	#scale_x_continuous(trans=log2_trans()) +
-	scale_y_continuous(trans=log2_trans())
-
-# Render the plot
-print(p)
-
-# Save the png image
-dev.off()
-
-
-png("heatmap.png", width=w*ppi, height=h*ppi, res=ppi)
-#
-## Create the plot with the normalized time vs nblocks
-p = ggplot(df, aes(x=cbs, y=rbs, fill=logmtime)) +
-	geom_raster() +
-  scale_fill_gradient(high="black", low="white") +
-  coord_fixed() +
-	theme_bw() +
-	theme(plot.subtitle=element_text(size=8)) +
-	labs(x="cbs", y="rbs",
-    title=sprintf("Heat granularity"), 
-    subtitle=input_file)
-
-# Render the plot
-print(p)
-
-# Save the png image
-dev.off()
+ggsave("time.png", plot=p, width=w, height=h, dpi=dpi)
+ggsave("time.pdf", plot=p, width=w, height=h, dpi=dpi)