Related
I would like to place two plots side by side using the ggplot2 package, i.e. do the equivalent of par(mfrow=c(1,2)).
For example, I would like to have the following two plots show side-by-side with the same scale.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
qplot(x,3*x+eps)
qplot(x,2*x+eps)
Do I need to put them in the same data.frame?
qplot(displ, hwy, data=mpg, facets = . ~ year) + geom_smooth()
Any ggplots side-by-side (or n plots on a grid)
The function grid.arrange() in the gridExtra package will combine multiple plots; this is how you put two side by side.
require(gridExtra)
plot1 <- qplot(1)
plot2 <- qplot(1)
grid.arrange(plot1, plot2, ncol=2)
This is useful when the two plots are not based on the same data, for example if you want to plot different variables without using reshape().
This will plot the output as a side effect. To print the side effect to a file, specify a device driver (such as pdf, png, etc), e.g.
pdf("foo.pdf")
grid.arrange(plot1, plot2)
dev.off()
or, use arrangeGrob() in combination with ggsave(),
ggsave("foo.pdf", arrangeGrob(plot1, plot2))
This is the equivalent of making two distinct plots using par(mfrow = c(1,2)). This not only saves time arranging data, it is necessary when you want two dissimilar plots.
Appendix: Using Facets
Facets are helpful for making similar plots for different groups. This is pointed out below in many answers below, but I want to highlight this approach with examples equivalent to the above plots.
mydata <- data.frame(myGroup = c('a', 'b'), myX = c(1,1))
qplot(data = mydata,
x = myX,
facets = ~myGroup)
ggplot(data = mydata) +
geom_bar(aes(myX)) +
facet_wrap(~myGroup)
Update
the plot_grid function in the cowplot is worth checking out as an alternative to grid.arrange. See the answer by #claus-wilke below and this vignette for an equivalent approach; but the function allows finer controls on plot location and size, based on this vignette.
One downside of the solutions based on grid.arrange is that they make it difficult to label the plots with letters (A, B, etc.), as most journals require.
I wrote the cowplot package to solve this (and a few other) issues, specifically the function plot_grid():
library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot() + theme_bw()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) + theme_bw() +
theme(legend.position = c(0.8, 0.8))
plot_grid(iris1, iris2, labels = "AUTO")
The object that plot_grid() returns is another ggplot2 object, and you can save it with ggsave() as usual:
p <- plot_grid(iris1, iris2, labels = "AUTO")
ggsave("plot.pdf", p)
Alternatively, you can use the cowplot function save_plot(), which is a thin wrapper around ggsave() that makes it easy to get the correct dimensions for combined plots, e.g.:
p <- plot_grid(iris1, iris2, labels = "AUTO")
save_plot("plot.pdf", p, ncol = 2)
(The ncol = 2 argument tells save_plot() that there are two plots side-by-side, and save_plot() makes the saved image twice as wide.)
For a more in-depth description of how to arrange plots in a grid see this vignette. There is also a vignette explaining how to make plots with a shared legend.
One frequent point of confusion is that the cowplot package changes the default ggplot2 theme. The package behaves that way because it was originally written for internal lab uses, and we never use the default theme. If this causes problems, you can use one of the following three approaches to work around them:
1. Set the theme manually for every plot. I think it's good practice to always specify a particular theme for each plot, just like I did with + theme_bw() in the example above. If you specify a particular theme, the default theme doesn't matter.
2. Revert the default theme back to the ggplot2 default. You can do this with one line of code:
theme_set(theme_gray())
3. Call cowplot functions without attaching the package. You can also not call library(cowplot) or require(cowplot) and instead call cowplot functions by prepending cowplot::. E.g., the above example using the ggplot2 default theme would become:
## Commented out, we don't call this
# library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) +
theme(legend.position = c(0.8, 0.8))
cowplot::plot_grid(iris1, iris2, labels = "AUTO")
Updates:
As of cowplot 1.0, the default ggplot2 theme is not changed anymore.
As of ggplot2 3.0.0, plots can be labeled directly, see e.g. here.
Using the patchwork package, you can simply use + operator:
library(ggplot2)
library(patchwork)
p1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
p2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
p1 + p2
Other operators include / to stack plots to place plots side by side, and () to group elements. For example you can configure a top row of 3 plots and a bottom row of one plot with (p1 | p2 | p3) /p. For more examples, see the package documentation.
You can use the following multiplot function from Winston Chang's R cookbook
multiplot(plot1, plot2, cols=2)
multiplot <- function(..., plotlist=NULL, cols) {
require(grid)
# Make a list from the ... arguments and plotlist
plots <- c(list(...), plotlist)
numPlots = length(plots)
# Make the panel
plotCols = cols # Number of columns of plots
plotRows = ceiling(numPlots/plotCols) # Number of rows needed, calculated from # of cols
# Set up the page
grid.newpage()
pushViewport(viewport(layout = grid.layout(plotRows, plotCols)))
vplayout <- function(x, y)
viewport(layout.pos.row = x, layout.pos.col = y)
# Make each plot, in the correct location
for (i in 1:numPlots) {
curRow = ceiling(i/plotCols)
curCol = (i-1) %% plotCols + 1
print(plots[[i]], vp = vplayout(curRow, curCol ))
}
}
Yes, methinks you need to arrange your data appropriately. One way would be this:
X <- data.frame(x=rep(x,2),
y=c(3*x+eps, 2*x+eps),
case=rep(c("first","second"), each=100))
qplot(x, y, data=X, facets = . ~ case) + geom_smooth()
I am sure there are better tricks in plyr or reshape -- I am still not really up to speed
on all these powerful packages by Hadley.
Using the reshape package you can do something like this.
library(ggplot2)
wide <- data.frame(x = rnorm(100), eps = rnorm(100, 0, .2))
wide$first <- with(wide, 3 * x + eps)
wide$second <- with(wide, 2 * x + eps)
long <- melt(wide, id.vars = c("x", "eps"))
ggplot(long, aes(x = x, y = value)) + geom_smooth() + geom_point() + facet_grid(.~ variable)
There is also multipanelfigure package that is worth to mention. See also this answer.
library(ggplot2)
theme_set(theme_bw())
q1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
q2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
q3 <- ggplot(mtcars) + geom_smooth(aes(disp, qsec))
q4 <- ggplot(mtcars) + geom_bar(aes(carb))
library(magrittr)
library(multipanelfigure)
figure1 <- multi_panel_figure(columns = 2, rows = 2, panel_label_type = "none")
# show the layout
figure1
figure1 %<>%
fill_panel(q1, column = 1, row = 1) %<>%
fill_panel(q2, column = 2, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2, row = 2)
figure1
# complex layout
figure2 <- multi_panel_figure(columns = 3, rows = 3, panel_label_type = "upper-roman")
figure2
figure2 %<>%
fill_panel(q1, column = 1:2, row = 1) %<>%
fill_panel(q2, column = 3, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2:3, row = 2:3)
figure2
Created on 2018-07-06 by the reprex package (v0.2.0.9000).
ggplot2 is based on grid graphics, which provide a different system for arranging plots on a page. The par(mfrow...) command doesn't have a direct equivalent, as grid objects (called grobs) aren't necessarily drawn immediately, but can be stored and manipulated as regular R objects before being converted to a graphical output. This enables greater flexibility than the draw this now model of base graphics, but the strategy is necessarily a little different.
I wrote grid.arrange() to provide a simple interface as close as possible to par(mfrow). In its simplest form, the code would look like:
library(ggplot2)
x <- rnorm(100)
eps <- rnorm(100,0,.2)
p1 <- qplot(x,3*x+eps)
p2 <- qplot(x,2*x+eps)
library(gridExtra)
grid.arrange(p1, p2, ncol = 2)
More options are detailed in this vignette.
One common complaint is that plots aren't necessarily aligned e.g. when they have axis labels of different size, but this is by design: grid.arrange makes no attempt to special-case ggplot2 objects, and treats them equally to other grobs (lattice plots, for instance). It merely places grobs in a rectangular layout.
For the special case of ggplot2 objects, I wrote another function, ggarrange, with a similar interface, which attempts to align plot panels (including facetted plots) and tries to respect the aspect ratios when defined by the user.
library(egg)
ggarrange(p1, p2, ncol = 2)
Both functions are compatible with ggsave(). For a general overview of the different options, and some historical context, this vignette offers additional information.
Update: This answer is very old. gridExtra::grid.arrange() is now the recommended approach.
I leave this here in case it might be useful.
Stephen Turner posted the arrange() function on Getting Genetics Done blog (see post for application instructions)
vp.layout <- function(x, y) viewport(layout.pos.row=x, layout.pos.col=y)
arrange <- function(..., nrow=NULL, ncol=NULL, as.table=FALSE) {
dots <- list(...)
n <- length(dots)
if(is.null(nrow) & is.null(ncol)) { nrow = floor(n/2) ; ncol = ceiling(n/nrow)}
if(is.null(nrow)) { nrow = ceiling(n/ncol)}
if(is.null(ncol)) { ncol = ceiling(n/nrow)}
## NOTE see n2mfrow in grDevices for possible alternative
grid.newpage()
pushViewport(viewport(layout=grid.layout(nrow,ncol) ) )
ii.p <- 1
for(ii.row in seq(1, nrow)){
ii.table.row <- ii.row
if(as.table) {ii.table.row <- nrow - ii.table.row + 1}
for(ii.col in seq(1, ncol)){
ii.table <- ii.p
if(ii.p > n) break
print(dots[[ii.table]], vp=vp.layout(ii.table.row, ii.col))
ii.p <- ii.p + 1
}
}
}
Using tidyverse:
x <- rnorm(100)
eps <- rnorm(100,0,.2)
df <- data.frame(x, eps) %>%
mutate(p1 = 3*x+eps, p2 = 2*x+eps) %>%
tidyr::gather("plot", "value", 3:4) %>%
ggplot(aes(x = x , y = value)) +
geom_point() +
geom_smooth() +
facet_wrap(~plot, ncol =2)
df
Consider also ggarrange from the ggpubr package. It has many benefits, including options to align axes between plots and to merge common legends into one.
The above solutions may not be efficient if you want to plot multiple ggplot plots using a loop (e.g. as asked here: Creating multiple plots in ggplot with different Y-axis values using a loop), which is a desired step in analyzing the unknown (or large) data-sets (e.g., when you want to plot Counts of all variables in a data-set).
The code below shows how to do that using the mentioned above 'multiplot()', the source of which is here: http://www.cookbook-r.com/Graphs/Multiple_graphs_on_one_page_(ggplot2):
plotAllCounts <- function (dt){
plots <- list();
for(i in 1:ncol(dt)) {
strX = names(dt)[i]
print(sprintf("%i: strX = %s", i, strX))
plots[[i]] <- ggplot(dt) + xlab(strX) +
geom_point(aes_string(strX),stat="count")
}
columnsToPlot <- floor(sqrt(ncol(dt)))
multiplot(plotlist = plots, cols = columnsToPlot)
}
Now run the function - to get Counts for all variables printed using ggplot on one page
dt = ggplot2::diamonds
plotAllCounts(dt)
One things to note is that:
using aes(get(strX)), which you would normally use in loops when working with ggplot , in the above code instead of aes_string(strX) will NOT draw the desired plots. Instead, it will plot the last plot many times. I have not figured out why - it may have to do the aes and aes_string are called in ggplot.
Otherwise, hope you'll find the function useful.
In my experience gridExtra:grid.arrange works perfectly, if you are trying to generate plots in a loop.
Short Code Snippet:
gridExtra::grid.arrange(plot1, plot2, ncol = 2)
** Updating this comment to show how to use grid.arrange() within a for loop to generate plots for different factors of a categorical variable.
for (bin_i in levels(athlete_clean$BMI_cat)) {
plot_BMI <- athlete_clean %>% filter(BMI_cat == bin_i) %>% group_by(BMI_cat,Team) %>% summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Participating Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
plot_BMI_Medal <- athlete_clean %>%
filter(!is.na(Medal), BMI_cat == bin_i) %>%
group_by(BMI_cat,Team) %>%
summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Winning Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
gridExtra::grid.arrange(plot_BMI, plot_BMI_Medal, ncol = 2)
}
One of the Sample Plots from the above for loop is included below.
The above loop will produce multiple plots for all levels of BMI category.
Sample Image
If you wish to see a more comprehensive use of grid.arrange() within for loops, check out https://rpubs.com/Mayank7j_2020/olympic_data_2000_2016
The cowplot package gives you a nice way to do this, in a manner that suits publication.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
A = qplot(x,3*x+eps, geom = c("point", "smooth"))+theme_gray()
B = qplot(x,2*x+eps, geom = c("point", "smooth"))+theme_gray()
cowplot::plot_grid(A, B, labels = c("A", "B"), align = "v")
I would like to place two plots side by side using the ggplot2 package, i.e. do the equivalent of par(mfrow=c(1,2)).
For example, I would like to have the following two plots show side-by-side with the same scale.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
qplot(x,3*x+eps)
qplot(x,2*x+eps)
Do I need to put them in the same data.frame?
qplot(displ, hwy, data=mpg, facets = . ~ year) + geom_smooth()
Any ggplots side-by-side (or n plots on a grid)
The function grid.arrange() in the gridExtra package will combine multiple plots; this is how you put two side by side.
require(gridExtra)
plot1 <- qplot(1)
plot2 <- qplot(1)
grid.arrange(plot1, plot2, ncol=2)
This is useful when the two plots are not based on the same data, for example if you want to plot different variables without using reshape().
This will plot the output as a side effect. To print the side effect to a file, specify a device driver (such as pdf, png, etc), e.g.
pdf("foo.pdf")
grid.arrange(plot1, plot2)
dev.off()
or, use arrangeGrob() in combination with ggsave(),
ggsave("foo.pdf", arrangeGrob(plot1, plot2))
This is the equivalent of making two distinct plots using par(mfrow = c(1,2)). This not only saves time arranging data, it is necessary when you want two dissimilar plots.
Appendix: Using Facets
Facets are helpful for making similar plots for different groups. This is pointed out below in many answers below, but I want to highlight this approach with examples equivalent to the above plots.
mydata <- data.frame(myGroup = c('a', 'b'), myX = c(1,1))
qplot(data = mydata,
x = myX,
facets = ~myGroup)
ggplot(data = mydata) +
geom_bar(aes(myX)) +
facet_wrap(~myGroup)
Update
the plot_grid function in the cowplot is worth checking out as an alternative to grid.arrange. See the answer by #claus-wilke below and this vignette for an equivalent approach; but the function allows finer controls on plot location and size, based on this vignette.
One downside of the solutions based on grid.arrange is that they make it difficult to label the plots with letters (A, B, etc.), as most journals require.
I wrote the cowplot package to solve this (and a few other) issues, specifically the function plot_grid():
library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot() + theme_bw()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) + theme_bw() +
theme(legend.position = c(0.8, 0.8))
plot_grid(iris1, iris2, labels = "AUTO")
The object that plot_grid() returns is another ggplot2 object, and you can save it with ggsave() as usual:
p <- plot_grid(iris1, iris2, labels = "AUTO")
ggsave("plot.pdf", p)
Alternatively, you can use the cowplot function save_plot(), which is a thin wrapper around ggsave() that makes it easy to get the correct dimensions for combined plots, e.g.:
p <- plot_grid(iris1, iris2, labels = "AUTO")
save_plot("plot.pdf", p, ncol = 2)
(The ncol = 2 argument tells save_plot() that there are two plots side-by-side, and save_plot() makes the saved image twice as wide.)
For a more in-depth description of how to arrange plots in a grid see this vignette. There is also a vignette explaining how to make plots with a shared legend.
One frequent point of confusion is that the cowplot package changes the default ggplot2 theme. The package behaves that way because it was originally written for internal lab uses, and we never use the default theme. If this causes problems, you can use one of the following three approaches to work around them:
1. Set the theme manually for every plot. I think it's good practice to always specify a particular theme for each plot, just like I did with + theme_bw() in the example above. If you specify a particular theme, the default theme doesn't matter.
2. Revert the default theme back to the ggplot2 default. You can do this with one line of code:
theme_set(theme_gray())
3. Call cowplot functions without attaching the package. You can also not call library(cowplot) or require(cowplot) and instead call cowplot functions by prepending cowplot::. E.g., the above example using the ggplot2 default theme would become:
## Commented out, we don't call this
# library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) +
theme(legend.position = c(0.8, 0.8))
cowplot::plot_grid(iris1, iris2, labels = "AUTO")
Updates:
As of cowplot 1.0, the default ggplot2 theme is not changed anymore.
As of ggplot2 3.0.0, plots can be labeled directly, see e.g. here.
Using the patchwork package, you can simply use + operator:
library(ggplot2)
library(patchwork)
p1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
p2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
p1 + p2
Other operators include / to stack plots to place plots side by side, and () to group elements. For example you can configure a top row of 3 plots and a bottom row of one plot with (p1 | p2 | p3) /p. For more examples, see the package documentation.
You can use the following multiplot function from Winston Chang's R cookbook
multiplot(plot1, plot2, cols=2)
multiplot <- function(..., plotlist=NULL, cols) {
require(grid)
# Make a list from the ... arguments and plotlist
plots <- c(list(...), plotlist)
numPlots = length(plots)
# Make the panel
plotCols = cols # Number of columns of plots
plotRows = ceiling(numPlots/plotCols) # Number of rows needed, calculated from # of cols
# Set up the page
grid.newpage()
pushViewport(viewport(layout = grid.layout(plotRows, plotCols)))
vplayout <- function(x, y)
viewport(layout.pos.row = x, layout.pos.col = y)
# Make each plot, in the correct location
for (i in 1:numPlots) {
curRow = ceiling(i/plotCols)
curCol = (i-1) %% plotCols + 1
print(plots[[i]], vp = vplayout(curRow, curCol ))
}
}
Yes, methinks you need to arrange your data appropriately. One way would be this:
X <- data.frame(x=rep(x,2),
y=c(3*x+eps, 2*x+eps),
case=rep(c("first","second"), each=100))
qplot(x, y, data=X, facets = . ~ case) + geom_smooth()
I am sure there are better tricks in plyr or reshape -- I am still not really up to speed
on all these powerful packages by Hadley.
Using the reshape package you can do something like this.
library(ggplot2)
wide <- data.frame(x = rnorm(100), eps = rnorm(100, 0, .2))
wide$first <- with(wide, 3 * x + eps)
wide$second <- with(wide, 2 * x + eps)
long <- melt(wide, id.vars = c("x", "eps"))
ggplot(long, aes(x = x, y = value)) + geom_smooth() + geom_point() + facet_grid(.~ variable)
There is also multipanelfigure package that is worth to mention. See also this answer.
library(ggplot2)
theme_set(theme_bw())
q1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
q2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
q3 <- ggplot(mtcars) + geom_smooth(aes(disp, qsec))
q4 <- ggplot(mtcars) + geom_bar(aes(carb))
library(magrittr)
library(multipanelfigure)
figure1 <- multi_panel_figure(columns = 2, rows = 2, panel_label_type = "none")
# show the layout
figure1
figure1 %<>%
fill_panel(q1, column = 1, row = 1) %<>%
fill_panel(q2, column = 2, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2, row = 2)
figure1
# complex layout
figure2 <- multi_panel_figure(columns = 3, rows = 3, panel_label_type = "upper-roman")
figure2
figure2 %<>%
fill_panel(q1, column = 1:2, row = 1) %<>%
fill_panel(q2, column = 3, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2:3, row = 2:3)
figure2
Created on 2018-07-06 by the reprex package (v0.2.0.9000).
ggplot2 is based on grid graphics, which provide a different system for arranging plots on a page. The par(mfrow...) command doesn't have a direct equivalent, as grid objects (called grobs) aren't necessarily drawn immediately, but can be stored and manipulated as regular R objects before being converted to a graphical output. This enables greater flexibility than the draw this now model of base graphics, but the strategy is necessarily a little different.
I wrote grid.arrange() to provide a simple interface as close as possible to par(mfrow). In its simplest form, the code would look like:
library(ggplot2)
x <- rnorm(100)
eps <- rnorm(100,0,.2)
p1 <- qplot(x,3*x+eps)
p2 <- qplot(x,2*x+eps)
library(gridExtra)
grid.arrange(p1, p2, ncol = 2)
More options are detailed in this vignette.
One common complaint is that plots aren't necessarily aligned e.g. when they have axis labels of different size, but this is by design: grid.arrange makes no attempt to special-case ggplot2 objects, and treats them equally to other grobs (lattice plots, for instance). It merely places grobs in a rectangular layout.
For the special case of ggplot2 objects, I wrote another function, ggarrange, with a similar interface, which attempts to align plot panels (including facetted plots) and tries to respect the aspect ratios when defined by the user.
library(egg)
ggarrange(p1, p2, ncol = 2)
Both functions are compatible with ggsave(). For a general overview of the different options, and some historical context, this vignette offers additional information.
Update: This answer is very old. gridExtra::grid.arrange() is now the recommended approach.
I leave this here in case it might be useful.
Stephen Turner posted the arrange() function on Getting Genetics Done blog (see post for application instructions)
vp.layout <- function(x, y) viewport(layout.pos.row=x, layout.pos.col=y)
arrange <- function(..., nrow=NULL, ncol=NULL, as.table=FALSE) {
dots <- list(...)
n <- length(dots)
if(is.null(nrow) & is.null(ncol)) { nrow = floor(n/2) ; ncol = ceiling(n/nrow)}
if(is.null(nrow)) { nrow = ceiling(n/ncol)}
if(is.null(ncol)) { ncol = ceiling(n/nrow)}
## NOTE see n2mfrow in grDevices for possible alternative
grid.newpage()
pushViewport(viewport(layout=grid.layout(nrow,ncol) ) )
ii.p <- 1
for(ii.row in seq(1, nrow)){
ii.table.row <- ii.row
if(as.table) {ii.table.row <- nrow - ii.table.row + 1}
for(ii.col in seq(1, ncol)){
ii.table <- ii.p
if(ii.p > n) break
print(dots[[ii.table]], vp=vp.layout(ii.table.row, ii.col))
ii.p <- ii.p + 1
}
}
}
Using tidyverse:
x <- rnorm(100)
eps <- rnorm(100,0,.2)
df <- data.frame(x, eps) %>%
mutate(p1 = 3*x+eps, p2 = 2*x+eps) %>%
tidyr::gather("plot", "value", 3:4) %>%
ggplot(aes(x = x , y = value)) +
geom_point() +
geom_smooth() +
facet_wrap(~plot, ncol =2)
df
Consider also ggarrange from the ggpubr package. It has many benefits, including options to align axes between plots and to merge common legends into one.
The above solutions may not be efficient if you want to plot multiple ggplot plots using a loop (e.g. as asked here: Creating multiple plots in ggplot with different Y-axis values using a loop), which is a desired step in analyzing the unknown (or large) data-sets (e.g., when you want to plot Counts of all variables in a data-set).
The code below shows how to do that using the mentioned above 'multiplot()', the source of which is here: http://www.cookbook-r.com/Graphs/Multiple_graphs_on_one_page_(ggplot2):
plotAllCounts <- function (dt){
plots <- list();
for(i in 1:ncol(dt)) {
strX = names(dt)[i]
print(sprintf("%i: strX = %s", i, strX))
plots[[i]] <- ggplot(dt) + xlab(strX) +
geom_point(aes_string(strX),stat="count")
}
columnsToPlot <- floor(sqrt(ncol(dt)))
multiplot(plotlist = plots, cols = columnsToPlot)
}
Now run the function - to get Counts for all variables printed using ggplot on one page
dt = ggplot2::diamonds
plotAllCounts(dt)
One things to note is that:
using aes(get(strX)), which you would normally use in loops when working with ggplot , in the above code instead of aes_string(strX) will NOT draw the desired plots. Instead, it will plot the last plot many times. I have not figured out why - it may have to do the aes and aes_string are called in ggplot.
Otherwise, hope you'll find the function useful.
In my experience gridExtra:grid.arrange works perfectly, if you are trying to generate plots in a loop.
Short Code Snippet:
gridExtra::grid.arrange(plot1, plot2, ncol = 2)
** Updating this comment to show how to use grid.arrange() within a for loop to generate plots for different factors of a categorical variable.
for (bin_i in levels(athlete_clean$BMI_cat)) {
plot_BMI <- athlete_clean %>% filter(BMI_cat == bin_i) %>% group_by(BMI_cat,Team) %>% summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Participating Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
plot_BMI_Medal <- athlete_clean %>%
filter(!is.na(Medal), BMI_cat == bin_i) %>%
group_by(BMI_cat,Team) %>%
summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Winning Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
gridExtra::grid.arrange(plot_BMI, plot_BMI_Medal, ncol = 2)
}
One of the Sample Plots from the above for loop is included below.
The above loop will produce multiple plots for all levels of BMI category.
Sample Image
If you wish to see a more comprehensive use of grid.arrange() within for loops, check out https://rpubs.com/Mayank7j_2020/olympic_data_2000_2016
The cowplot package gives you a nice way to do this, in a manner that suits publication.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
A = qplot(x,3*x+eps, geom = c("point", "smooth"))+theme_gray()
B = qplot(x,2*x+eps, geom = c("point", "smooth"))+theme_gray()
cowplot::plot_grid(A, B, labels = c("A", "B"), align = "v")
I am trying to plot two variables where N=700K. The problem is that there is too much overlap, so that the plot becomes mostly a solid block of black. Is there any way of having a grayscale "cloud" where the darkness of the plot is a function of the number of points in an region? In other words, instead of showing individual points, I want the plot to be a "cloud", with the more the number of points in a region, the darker that region.
One way to deal with this is with alpha blending, which makes each point slightly transparent. So regions appear darker that have more point plotted on them.
This is easy to do in ggplot2:
df <- data.frame(x = rnorm(5000),y=rnorm(5000))
ggplot(df,aes(x=x,y=y)) + geom_point(alpha = 0.3)
Another convenient way to deal with this is (and probably more appropriate for the number of points you have) is hexagonal binning:
ggplot(df,aes(x=x,y=y)) + stat_binhex()
And there is also regular old rectangular binning (image omitted), which is more like your traditional heatmap:
ggplot(df,aes(x=x,y=y)) + geom_bin2d()
An overview of several good options in ggplot2:
library(ggplot2)
x <- rnorm(n = 10000)
y <- rnorm(n = 10000, sd=2) + x
df <- data.frame(x, y)
Option A: transparent points
o1 <- ggplot(df, aes(x, y)) +
geom_point(alpha = 0.05)
Option B: add density contours
o2 <- ggplot(df, aes(x, y)) +
geom_point(alpha = 0.05) +
geom_density_2d()
Option C: add filled density contours
(Note that the points distort the perception of the colors underneath, may be better without points.)
o3 <- ggplot(df, aes(x, y)) +
stat_density_2d(aes(fill = stat(level)), geom = 'polygon') +
scale_fill_viridis_c(name = "density") +
geom_point(shape = '.')
Option D: density heatmap
(Same note as C.)
o4 <- ggplot(df, aes(x, y)) +
stat_density_2d(aes(fill = stat(density)), geom = 'raster', contour = FALSE) +
scale_fill_viridis_c() +
coord_cartesian(expand = FALSE) +
geom_point(shape = '.', col = 'white')
Option E: hexbins
(Same note as C.)
o5 <- ggplot(df, aes(x, y)) +
geom_hex() +
scale_fill_viridis_c() +
geom_point(shape = '.', col = 'white')
Option F: rugs
Possibly my favorite option. Not quite as flashy, but visually simple and simple to understand. Very effective in many cases.
o6 <- ggplot(df, aes(x, y)) +
geom_point(alpha = 0.1) +
geom_rug(alpha = 0.01)
Combine in one figure:
cowplot::plot_grid(
o1, o2, o3, o4, o5, o6,
ncol = 2, labels = 'AUTO', align = 'v', axis = 'lr'
)
You can also have a look at the ggsubplot package. This package implements features which were presented by Hadley Wickham back in 2011 (http://blog.revolutionanalytics.com/2011/10/ggplot2-for-big-data.html).
(In the following, I include the "points"-layer for illustration purposes.)
library(ggplot2)
library(ggsubplot)
# Make up some data
set.seed(955)
dat <- data.frame(cond = rep(c("A", "B"), each=5000),
xvar = c(rep(1:20,250) + rnorm(5000,sd=5),rep(16:35,250) + rnorm(5000,sd=5)),
yvar = c(rep(1:20,250) + rnorm(5000,sd=5),rep(16:35,250) + rnorm(5000,sd=5)))
# Scatterplot with subplots (simple)
ggplot(dat, aes(x=xvar, y=yvar)) +
geom_point(shape=1) +
geom_subplot2d(aes(xvar, yvar,
subplot = geom_bar(aes(rep("dummy", length(xvar)), ..count..))), bins = c(15,15), ref = NULL, width = rel(0.8), ply.aes = FALSE)
However, this features rocks if you have a third variable to control for.
# Scatterplot with subplots (including a third variable)
ggplot(dat, aes(x=xvar, y=yvar)) +
geom_point(shape=1, aes(color = factor(cond))) +
geom_subplot2d(aes(xvar, yvar,
subplot = geom_bar(aes(cond, ..count.., fill = cond))),
bins = c(15,15), ref = NULL, width = rel(0.8), ply.aes = FALSE)
Or another approach would be to use smoothScatter():
smoothScatter(dat[2:3])
Alpha blending is easy to do with base graphics as well.
df <- data.frame(x = rnorm(5000),y=rnorm(5000))
with(df, plot(x, y, col="#00000033"))
The first six numbers after the # are the color in RGB hex and the last two are the opacity, again in hex, so 33 ~ 3/16th opaque.
You can also use density contour lines (ggplot2):
df <- data.frame(x = rnorm(15000),y=rnorm(15000))
ggplot(df,aes(x=x,y=y)) + geom_point() + geom_density2d()
Or combine density contours with alpha blending:
ggplot(df,aes(x=x,y=y)) +
geom_point(colour="blue", alpha=0.2) +
geom_density2d(colour="black")
You may find useful the hexbin package. From the help page of hexbinplot:
library(hexbin)
mixdata <- data.frame(x = c(rnorm(5000),rnorm(5000,4,1.5)),
y = c(rnorm(5000),rnorm(5000,2,3)),
a = gl(2, 5000))
hexbinplot(y ~ x | a, mixdata)
geom_pointdenisty from the ggpointdensity package (recently developed by Lukas Kremer and Simon Anders (2019)) allows you visualize density and individual data points at the same time:
library(ggplot2)
# install.packages("ggpointdensity")
library(ggpointdensity)
df <- data.frame(x = rnorm(5000), y = rnorm(5000))
ggplot(df, aes(x=x, y=y)) + geom_pointdensity() + scale_color_viridis_c()
My favorite method for plotting this type of data is the one described in this question - a scatter-density plot. The idea is to do a scatter-plot but to colour the points by their density (roughly speaking, the amount of overlap in that area).
It simultaneously:
clearly shows the location of outliers, and
reveals any structure in the dense area of the plot.
Here is the result from the top answer to the linked question:
I have two graphs with the same x axis - the range of x is 0-5 in both of them.
I would like to combine both of them to one graph and I didn't find a previous example.
Here is what I got:
c <- ggplot(survey, aes(often_post,often_privacy)) + stat_smooth(method="loess")
c <- ggplot(survey, aes(frequent_read,often_privacy)) + stat_smooth(method="loess")
How can I combine them?
The y axis is "often privacy" and in each graph the x axis is "often post" or "frequent read".
I thought I can combine them easily (somehow) because the range is 0-5 in both of them.
Many thanks!
Example code for Ben's solution.
#Sample data
survey <- data.frame(
often_post = runif(10, 0, 5),
frequent_read = 5 * rbeta(10, 1, 1),
often_privacy = sample(10, replace = TRUE)
)
#Reshape the data frame
survey2 <- melt(survey, measure.vars = c("often_post", "frequent_read"))
#Plot using colour as an aesthetic to distinguish lines
(p <- ggplot(survey2, aes(value, often_privacy, colour = variable)) +
geom_point() +
geom_smooth()
)
You can use + to combine other plots on the same ggplot object. For example, to plot points and smoothed lines for both pairs of columns:
ggplot(survey, aes(often_post,often_privacy)) +
geom_point() +
geom_smooth() +
geom_point(aes(frequent_read,often_privacy)) +
geom_smooth(aes(frequent_read,often_privacy))
Try this:
df <- data.frame(x=x_var, y=y1_var, type='y1')
df <- rbind(df, data.frame(x=x_var, y=y2_var, type='y2'))
ggplot(df, aes(x, y, group=type, col=type)) + geom_line()
I would like to place two plots side by side using the ggplot2 package, i.e. do the equivalent of par(mfrow=c(1,2)).
For example, I would like to have the following two plots show side-by-side with the same scale.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
qplot(x,3*x+eps)
qplot(x,2*x+eps)
Do I need to put them in the same data.frame?
qplot(displ, hwy, data=mpg, facets = . ~ year) + geom_smooth()
Any ggplots side-by-side (or n plots on a grid)
The function grid.arrange() in the gridExtra package will combine multiple plots; this is how you put two side by side.
require(gridExtra)
plot1 <- qplot(1)
plot2 <- qplot(1)
grid.arrange(plot1, plot2, ncol=2)
This is useful when the two plots are not based on the same data, for example if you want to plot different variables without using reshape().
This will plot the output as a side effect. To print the side effect to a file, specify a device driver (such as pdf, png, etc), e.g.
pdf("foo.pdf")
grid.arrange(plot1, plot2)
dev.off()
or, use arrangeGrob() in combination with ggsave(),
ggsave("foo.pdf", arrangeGrob(plot1, plot2))
This is the equivalent of making two distinct plots using par(mfrow = c(1,2)). This not only saves time arranging data, it is necessary when you want two dissimilar plots.
Appendix: Using Facets
Facets are helpful for making similar plots for different groups. This is pointed out below in many answers below, but I want to highlight this approach with examples equivalent to the above plots.
mydata <- data.frame(myGroup = c('a', 'b'), myX = c(1,1))
qplot(data = mydata,
x = myX,
facets = ~myGroup)
ggplot(data = mydata) +
geom_bar(aes(myX)) +
facet_wrap(~myGroup)
Update
the plot_grid function in the cowplot is worth checking out as an alternative to grid.arrange. See the answer by #claus-wilke below and this vignette for an equivalent approach; but the function allows finer controls on plot location and size, based on this vignette.
One downside of the solutions based on grid.arrange is that they make it difficult to label the plots with letters (A, B, etc.), as most journals require.
I wrote the cowplot package to solve this (and a few other) issues, specifically the function plot_grid():
library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot() + theme_bw()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) + theme_bw() +
theme(legend.position = c(0.8, 0.8))
plot_grid(iris1, iris2, labels = "AUTO")
The object that plot_grid() returns is another ggplot2 object, and you can save it with ggsave() as usual:
p <- plot_grid(iris1, iris2, labels = "AUTO")
ggsave("plot.pdf", p)
Alternatively, you can use the cowplot function save_plot(), which is a thin wrapper around ggsave() that makes it easy to get the correct dimensions for combined plots, e.g.:
p <- plot_grid(iris1, iris2, labels = "AUTO")
save_plot("plot.pdf", p, ncol = 2)
(The ncol = 2 argument tells save_plot() that there are two plots side-by-side, and save_plot() makes the saved image twice as wide.)
For a more in-depth description of how to arrange plots in a grid see this vignette. There is also a vignette explaining how to make plots with a shared legend.
One frequent point of confusion is that the cowplot package changes the default ggplot2 theme. The package behaves that way because it was originally written for internal lab uses, and we never use the default theme. If this causes problems, you can use one of the following three approaches to work around them:
1. Set the theme manually for every plot. I think it's good practice to always specify a particular theme for each plot, just like I did with + theme_bw() in the example above. If you specify a particular theme, the default theme doesn't matter.
2. Revert the default theme back to the ggplot2 default. You can do this with one line of code:
theme_set(theme_gray())
3. Call cowplot functions without attaching the package. You can also not call library(cowplot) or require(cowplot) and instead call cowplot functions by prepending cowplot::. E.g., the above example using the ggplot2 default theme would become:
## Commented out, we don't call this
# library(cowplot)
iris1 <- ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot()
iris2 <- ggplot(iris, aes(x = Sepal.Length, fill = Species)) +
geom_density(alpha = 0.7) +
theme(legend.position = c(0.8, 0.8))
cowplot::plot_grid(iris1, iris2, labels = "AUTO")
Updates:
As of cowplot 1.0, the default ggplot2 theme is not changed anymore.
As of ggplot2 3.0.0, plots can be labeled directly, see e.g. here.
Using the patchwork package, you can simply use + operator:
library(ggplot2)
library(patchwork)
p1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
p2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
p1 + p2
Other operators include / to stack plots to place plots side by side, and () to group elements. For example you can configure a top row of 3 plots and a bottom row of one plot with (p1 | p2 | p3) /p. For more examples, see the package documentation.
You can use the following multiplot function from Winston Chang's R cookbook
multiplot(plot1, plot2, cols=2)
multiplot <- function(..., plotlist=NULL, cols) {
require(grid)
# Make a list from the ... arguments and plotlist
plots <- c(list(...), plotlist)
numPlots = length(plots)
# Make the panel
plotCols = cols # Number of columns of plots
plotRows = ceiling(numPlots/plotCols) # Number of rows needed, calculated from # of cols
# Set up the page
grid.newpage()
pushViewport(viewport(layout = grid.layout(plotRows, plotCols)))
vplayout <- function(x, y)
viewport(layout.pos.row = x, layout.pos.col = y)
# Make each plot, in the correct location
for (i in 1:numPlots) {
curRow = ceiling(i/plotCols)
curCol = (i-1) %% plotCols + 1
print(plots[[i]], vp = vplayout(curRow, curCol ))
}
}
Yes, methinks you need to arrange your data appropriately. One way would be this:
X <- data.frame(x=rep(x,2),
y=c(3*x+eps, 2*x+eps),
case=rep(c("first","second"), each=100))
qplot(x, y, data=X, facets = . ~ case) + geom_smooth()
I am sure there are better tricks in plyr or reshape -- I am still not really up to speed
on all these powerful packages by Hadley.
Using the reshape package you can do something like this.
library(ggplot2)
wide <- data.frame(x = rnorm(100), eps = rnorm(100, 0, .2))
wide$first <- with(wide, 3 * x + eps)
wide$second <- with(wide, 2 * x + eps)
long <- melt(wide, id.vars = c("x", "eps"))
ggplot(long, aes(x = x, y = value)) + geom_smooth() + geom_point() + facet_grid(.~ variable)
There is also multipanelfigure package that is worth to mention. See also this answer.
library(ggplot2)
theme_set(theme_bw())
q1 <- ggplot(mtcars) + geom_point(aes(mpg, disp))
q2 <- ggplot(mtcars) + geom_boxplot(aes(gear, disp, group = gear))
q3 <- ggplot(mtcars) + geom_smooth(aes(disp, qsec))
q4 <- ggplot(mtcars) + geom_bar(aes(carb))
library(magrittr)
library(multipanelfigure)
figure1 <- multi_panel_figure(columns = 2, rows = 2, panel_label_type = "none")
# show the layout
figure1
figure1 %<>%
fill_panel(q1, column = 1, row = 1) %<>%
fill_panel(q2, column = 2, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2, row = 2)
figure1
# complex layout
figure2 <- multi_panel_figure(columns = 3, rows = 3, panel_label_type = "upper-roman")
figure2
figure2 %<>%
fill_panel(q1, column = 1:2, row = 1) %<>%
fill_panel(q2, column = 3, row = 1) %<>%
fill_panel(q3, column = 1, row = 2) %<>%
fill_panel(q4, column = 2:3, row = 2:3)
figure2
Created on 2018-07-06 by the reprex package (v0.2.0.9000).
ggplot2 is based on grid graphics, which provide a different system for arranging plots on a page. The par(mfrow...) command doesn't have a direct equivalent, as grid objects (called grobs) aren't necessarily drawn immediately, but can be stored and manipulated as regular R objects before being converted to a graphical output. This enables greater flexibility than the draw this now model of base graphics, but the strategy is necessarily a little different.
I wrote grid.arrange() to provide a simple interface as close as possible to par(mfrow). In its simplest form, the code would look like:
library(ggplot2)
x <- rnorm(100)
eps <- rnorm(100,0,.2)
p1 <- qplot(x,3*x+eps)
p2 <- qplot(x,2*x+eps)
library(gridExtra)
grid.arrange(p1, p2, ncol = 2)
More options are detailed in this vignette.
One common complaint is that plots aren't necessarily aligned e.g. when they have axis labels of different size, but this is by design: grid.arrange makes no attempt to special-case ggplot2 objects, and treats them equally to other grobs (lattice plots, for instance). It merely places grobs in a rectangular layout.
For the special case of ggplot2 objects, I wrote another function, ggarrange, with a similar interface, which attempts to align plot panels (including facetted plots) and tries to respect the aspect ratios when defined by the user.
library(egg)
ggarrange(p1, p2, ncol = 2)
Both functions are compatible with ggsave(). For a general overview of the different options, and some historical context, this vignette offers additional information.
Update: This answer is very old. gridExtra::grid.arrange() is now the recommended approach.
I leave this here in case it might be useful.
Stephen Turner posted the arrange() function on Getting Genetics Done blog (see post for application instructions)
vp.layout <- function(x, y) viewport(layout.pos.row=x, layout.pos.col=y)
arrange <- function(..., nrow=NULL, ncol=NULL, as.table=FALSE) {
dots <- list(...)
n <- length(dots)
if(is.null(nrow) & is.null(ncol)) { nrow = floor(n/2) ; ncol = ceiling(n/nrow)}
if(is.null(nrow)) { nrow = ceiling(n/ncol)}
if(is.null(ncol)) { ncol = ceiling(n/nrow)}
## NOTE see n2mfrow in grDevices for possible alternative
grid.newpage()
pushViewport(viewport(layout=grid.layout(nrow,ncol) ) )
ii.p <- 1
for(ii.row in seq(1, nrow)){
ii.table.row <- ii.row
if(as.table) {ii.table.row <- nrow - ii.table.row + 1}
for(ii.col in seq(1, ncol)){
ii.table <- ii.p
if(ii.p > n) break
print(dots[[ii.table]], vp=vp.layout(ii.table.row, ii.col))
ii.p <- ii.p + 1
}
}
}
Using tidyverse:
x <- rnorm(100)
eps <- rnorm(100,0,.2)
df <- data.frame(x, eps) %>%
mutate(p1 = 3*x+eps, p2 = 2*x+eps) %>%
tidyr::gather("plot", "value", 3:4) %>%
ggplot(aes(x = x , y = value)) +
geom_point() +
geom_smooth() +
facet_wrap(~plot, ncol =2)
df
Consider also ggarrange from the ggpubr package. It has many benefits, including options to align axes between plots and to merge common legends into one.
The above solutions may not be efficient if you want to plot multiple ggplot plots using a loop (e.g. as asked here: Creating multiple plots in ggplot with different Y-axis values using a loop), which is a desired step in analyzing the unknown (or large) data-sets (e.g., when you want to plot Counts of all variables in a data-set).
The code below shows how to do that using the mentioned above 'multiplot()', the source of which is here: http://www.cookbook-r.com/Graphs/Multiple_graphs_on_one_page_(ggplot2):
plotAllCounts <- function (dt){
plots <- list();
for(i in 1:ncol(dt)) {
strX = names(dt)[i]
print(sprintf("%i: strX = %s", i, strX))
plots[[i]] <- ggplot(dt) + xlab(strX) +
geom_point(aes_string(strX),stat="count")
}
columnsToPlot <- floor(sqrt(ncol(dt)))
multiplot(plotlist = plots, cols = columnsToPlot)
}
Now run the function - to get Counts for all variables printed using ggplot on one page
dt = ggplot2::diamonds
plotAllCounts(dt)
One things to note is that:
using aes(get(strX)), which you would normally use in loops when working with ggplot , in the above code instead of aes_string(strX) will NOT draw the desired plots. Instead, it will plot the last plot many times. I have not figured out why - it may have to do the aes and aes_string are called in ggplot.
Otherwise, hope you'll find the function useful.
In my experience gridExtra:grid.arrange works perfectly, if you are trying to generate plots in a loop.
Short Code Snippet:
gridExtra::grid.arrange(plot1, plot2, ncol = 2)
** Updating this comment to show how to use grid.arrange() within a for loop to generate plots for different factors of a categorical variable.
for (bin_i in levels(athlete_clean$BMI_cat)) {
plot_BMI <- athlete_clean %>% filter(BMI_cat == bin_i) %>% group_by(BMI_cat,Team) %>% summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Participating Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
plot_BMI_Medal <- athlete_clean %>%
filter(!is.na(Medal), BMI_cat == bin_i) %>%
group_by(BMI_cat,Team) %>%
summarize(count_BMI_team = n()) %>%
mutate(percentage_cbmiT = round(count_BMI_team/sum(count_BMI_team) * 100,2)) %>%
arrange(-count_BMI_team) %>% top_n(10,count_BMI_team) %>%
ggplot(aes(x = reorder(Team,count_BMI_team), y = count_BMI_team, fill = Team)) +
geom_bar(stat = "identity") +
theme_bw() +
# facet_wrap(~Medal) +
labs(title = paste("Top 10 Winning Teams with \n",bin_i," BMI",sep=""), y = "Number of Athletes",
x = paste("Teams - ",bin_i," BMI Category", sep="")) +
geom_text(aes(label = paste(percentage_cbmiT,"%",sep = "")),
size = 3, check_overlap = T, position = position_stack(vjust = 0.7) ) +
theme(axis.text.x = element_text(angle = 00, vjust = 0.5), plot.title = element_text(hjust = 0.5), legend.position = "none") +
coord_flip()
gridExtra::grid.arrange(plot_BMI, plot_BMI_Medal, ncol = 2)
}
One of the Sample Plots from the above for loop is included below.
The above loop will produce multiple plots for all levels of BMI category.
Sample Image
If you wish to see a more comprehensive use of grid.arrange() within for loops, check out https://rpubs.com/Mayank7j_2020/olympic_data_2000_2016
The cowplot package gives you a nice way to do this, in a manner that suits publication.
x <- rnorm(100)
eps <- rnorm(100,0,.2)
A = qplot(x,3*x+eps, geom = c("point", "smooth"))+theme_gray()
B = qplot(x,2*x+eps, geom = c("point", "smooth"))+theme_gray()
cowplot::plot_grid(A, B, labels = c("A", "B"), align = "v")