I am trying to loop a ggplot2 plot with a linear regression line over it. It works when I type the y column name manually, but the loop method I am trying does not work. It is definitely not a dataset issue.
I've tried many solutions from various websites on how to loop a ggplot and the one I've attempted is the simplest I could find that almost does the job.
The code that works is the following:
plots <- ggplot(Everything.any, mapping = aes(x = stock_VWRETD, y = stock_10065)) +
geom_point() +
labs(x = 'Market Returns', y = 'Stock Returns', title ='Stock vs Market Returns') +
geom_smooth(method='lm',formula=y~x)
But I do not want to do this another 40 times (and then 5 times more for other reasons). The code that I've found on-line and have tried to modify it for my means is the following:
plotRegression <- function(z,na.rm=TRUE,...){
nm <- colnames(z)
for (i in seq_along(nm)){
plots <- ggplot(z, mapping = aes(x = stock_VWRETD, y = nm[i])) +
geom_point() +
labs(x = 'Market Returns', y = 'Stock Returns', title ='Stock vs Market Returns') +
geom_smooth(method='lm',formula=y~x)
ggsave(plots,filename=paste("regression1",nm[i],".png",sep=" "))
}
}
plotRegression(Everything.any)
I expect it to be the nice graph that I'd expect to get, a Stock returns vs Market returns graph, but instead on the y-axis, I get one value which is the name of the respective column, and the Market value plotted as normally, but as if on a straight number-line across the one y-axis value. Please let me know what I am doing wrong.
Desired Plot:
Actual Plot:
Sample Data is available on Google Drive here:
https://drive.google.com/open?id=1Xa1RQQaDm0pGSf3Y-h5ZR0uTWE-NqHtt
The problem is that when you assign variables to aesthetics in aes, you mix bare names and strings. In this example, both X and Y are supposed to be variables in z:
aes(x = stock_VWRETD, y = nm[i])
You refer to stock_VWRETD using a bare name (as required with aes), however for y=, you provide the name as a character vector produced by colnames. See what happens when we replicate this with the iris dataset:
ggplot(iris, aes(Petal.Length, 'Sepal.Length')) + geom_point()
Since aes expects variable names to be given as bare names, it doesn't interpret 'Sepal.Length' as a variable in iris but as a separate vector (consisting of a single character value) which holds the y-values for each point.
What can you do? Here are 2 options that both give the proper plot
1) Use aes_string and change both variable names to character:
ggplot(iris, aes_string('Petal.Length', 'Sepal.Length')) + geom_point()
2) Use square bracket subsetting to manually extract the appropriate variable:
ggplot(iris, aes(Petal.Length, .data[['Sepal.Length']])) + geom_point()
you need to use aes_string instead of aes, and double-quotes around your x variable, and then you can directly use your i variable. You can also simplify your for loop call. Here is an example using iris.
library(ggplot2)
plotRegression <- function(z,na.rm=TRUE,...){
nm <- colnames(z)
for (i in nm){
plots <- ggplot(z, mapping = aes_string(x = "Sepal.Length", y = i)) +
geom_point()+
geom_smooth(method='lm',formula=y~x)
ggsave(plots,filename=paste("regression1_",i,".png",sep=""))
}
}
myiris<-iris
plotRegression(myiris)
I have a function to generate scatter plots from data, where an argument is provided to select which column to use for coloring the points. Here is a simplified version:
library(ggplot2)
plot_gene <- function (df, gene) {
ggplot(df, aes(x, y)) +
geom_point(aes_string(col = gene)) +
scale_color_gradient()
}
where df is a data.frame with columns x, y, and then a bunch of gene names. This works fine for most gene names; however, some have dashes and these fail:
print(plot_gene(df, "Gapdh")) # great!
print(plot_gene(df, "H2-Aa")) # Error: object "H2" not found
It appears the gene variable is getting parsed ("H2-Aa" becomes H2 - Aa). How can I get around this? Is there a way to indicate that a string should not go through eval in aes_string?
Reproducible Input
If you need some input to play with, this fails like my data:
df <- data.frame(c(1,2), c(2,1), c(1,2), c(2,1))
colnames(df) <- c("x", "y", "Gapdh", "H2-Aa")
For my real data, I am using read.table(..., header=TRUE) and get column names with dashes because the raw data files have them.
Normally R tries very hard to make sure you have column names in your data.frame that can be valid variable names. Using non-standard column names (those that are not valid variable names) will lead to problems when using functions that use non-standard evaluation type syntax. When focused to use such variable names you often have to wrap them in back ticks. In the normal case
ggplot(df, aes(x, y)) +
geom_point(aes(col = H2-Aa)) +
scale_color_gradient()
# Error in FUN(X[[i]], ...) : object 'H2' not found
would return an error but
ggplot(df, aes(x, y)) +
geom_point(aes(col = `H2-Aa`)) +
scale_color_gradient()
would work.
You can paste in backticks if you really want
geom_point(aes_string(col = paste0("`", gene, "`")))
or you could treat it as a symbol from the get-go and use aes_q instread
geom_point(aes_q(col = as.name(gene)))
The latest release of ggplot support escaping via !! rather than using aes_string or aes_q so you could do
geom_point(aes(col = !!rlang::sym(gene)))
I can not find the documentation for the double dots around density
set.seed(1234)
df <- data.frame(cond = factor(rep(c("A","B"), each=200)), rating = c(rnorm(200),rnorm(200, mean=.8)))
print(head(df))
print(ggplot(df, aes(x=rating)) +
geom_histogram(aes(y=..density..), # Histogram with density instead of count on y-axis
binwidth=.5,
colour="black", fill="white") +
geom_density(alpha=.2, fill="#FF6666") +
geom_vline(aes(xintercept=mean(rating, na.rm=T)), # Ignore NA values for mean
color="red", linetype="dashed", size=1))
Do you know what operator they represent ?
Edit
I know what it does when used in a geom, I would like to know what it is.
For instance, the single dot operator is defined as
> .
function (..., .env = parent.frame())
{
structure(as.list(match.call()[-1]), env = .env, class = "quoted")
}
<environment: namespace:plyr>
If I redefine density, then ..density.. has a different effect, so it seems XX -> ..XX.. is an operator. I would like to find how it is defined.
Unlike many other languages, in R, the dot is perfectly valid in identifiers. In this case, ..count.. is an identifier. However, there is special code in ggplot2 to detect this pattern, and to strip the dots. It feels unlikely that real code would use identifiers formatted like that, and so this is a neat way to distinguish between defined and calculated aesthetics.
The relevant code is at the end of layer.r:
# Determine if aesthetic is calculated
is_calculated_aes <- function(aesthetics) {
match <- "\\.\\.([a-zA-z._]+)\\.\\."
stats <- rep(FALSE, length(aesthetics))
grepl(match, sapply(aesthetics, deparse))
}
# Strip dots from expressions
strip_dots <- function(aesthetics) {
match <- "\\.\\.([a-zA-z._]+)\\.\\."
strings <- lapply(aesthetics, deparse)
strings <- lapply(strings, gsub, pattern = match, replacement = "\\1")
lapply(strings, function(x) parse(text = x)[[1]])
}
It is used further up above in the map_statistic function. If a calculated aesthetic is present, another data frame (one that contains e.g. the count column) is used for the plot.
The single dot . is just another identifier, defined in the plyr package. As you can see, it is a function.
new to ggplot2, I've scoured the web but still couldn't figure this out.
I understand how to plot a boxplot in ggplot2, my problem is that I can't pass along the variable I use for groups when it is inside a function.
so, normally (i.e. NOT inside a function), I would write this:
ggplot(myData, aes(factor(Variable1), Variable2)) +
geom_boxplot(fill="grey", colour="black")+
labs(title = "Variable1 vs. Variable2" ) +
labs (x = "variable1", y = "Variable2")
Where myData is my data frame
Variable 1 is a 2 level factor variable
Variable 2 is a continuous variable
I want to make boxplots of Variable 1 by its 2 levels/groups
and this works fine,
but as soon as I write this inside a function I couldn't get it to work.
my attempt in writing the function:
myfunction = function (data, Variable1) {
ggplot(data=myData, aes_string(factor("Variable1"), "Variable2"))+
geom_boxplot(fill="grey", colour="black")+
labs(title = paste("Variable1 vs. Variable2" )) +
labs (x = "variable1", y = "Variable2")
}
this only gives me a single boxplot(instead of 2), as if it never understood the factor(Variable1) command (and did a single boxplot of the entire Variable 2, rather than separate them by Variable 1 level first, then boxplot them).
Aes_string evaluates the entire string, so if you do sprintf("factor(%s)",Variable1) you get the desired result. As a further remark: your function has a data-argument, but inside the plotting you use myData. I have also edited the x-lab and title, so that you can pass 'Variable3' and get proper labels.
With some example data:
set.seed(123)
dat <- data.frame(Variable2=rnorm(100),Variable1=c(0,1),Variable3=sample(0:1,100,T))
myfunction = function (data, Variable1) {
ggplot(data=data, aes_string(sprintf("factor(%s)",Variable1), "Variable2"))+
geom_boxplot(fill="grey", colour="black")+
labs(title = sprintf("%s and Variable2", Variable1)) +
labs (x = Variable1, y = "Variable2")
}
p1 <- myfunction(dat,"Variable1")
p2 <- myfunction(dat,"Variable3")
I'm trying to write a simple plot function, using the ggplot2 library. But the call to ggplot doesn't find the function argument.
Consider a data.frame called means that stores two conditions and two mean values that I want to plot (condition will appear on the X axis, means on the Y).
library(ggplot2)
m <- c(13.8, 14.8)
cond <- c(1, 2)
means <- data.frame(means=m, condition=cond)
means
# The output should be:
# means condition
# 1 13.8 1
# 2 14.8 2
testplot <- function(meansdf)
{
p <- ggplot(meansdf, aes(fill=meansdf$condition, y=meansdf$means, x = meansdf$condition))
p + geom_bar(position="dodge", stat="identity")
}
testplot(means)
# This will output the following error:
# Error in eval(expr, envir, enclos) : object 'meansdf' not found
So it seems that ggplot is calling eval, which can't find the argument meansdf. Does anyone know how I can successfully pass the function argument to ggplot?
(Note: Yes I could just call the ggplot function directly, but in the end I hope to make my plot function do more complicated stuff! :) )
The "proper" way to use ggplot programmatically is to use aes_string() instead of aes() and use the names of the columns as characters rather than as objects:
For more programmatic uses, for example if you wanted users to be able to specify column names for various aesthetics as arguments, or if this function is going in a package that needs to pass R CMD CHECK without warnings about variable names without definitions, you can use aes_string(), with the columns needed as characters.
testplot <- function(meansdf, xvar = "condition", yvar = "means",
fillvar = "condition") {
p <- ggplot(meansdf,
aes_string(x = xvar, y= yvar, fill = fillvar)) +
geom_bar(position="dodge", stat="identity")
}
As Joris and Chase have already correctly answered, standard best practice is to simply omit the meansdf$ part and directly refer to the data frame columns.
testplot <- function(meansdf)
{
p <- ggplot(meansdf,
aes(fill = condition,
y = means,
x = condition))
p + geom_bar(position = "dodge", stat = "identity")
}
This works, because the variables referred to in aes are looked for either in the global environment or in the data frame passed to ggplot. That is also the reason why your example code - using meansdf$condition etc. - did not work: meansdf is neither available in the global environment, nor is it available inside the data frame passed to ggplot, which is meansdf itself.
The fact that the variables are looked for in the global environment instead of in the calling environment is actually a known bug in ggplot2 that Hadley does not consider fixable at the moment.
This leads to problems, if one wishes to use a local variable, say, scale, to influence the data used for the plot:
testplot <- function(meansdf)
{
scale <- 0.5
p <- ggplot(meansdf,
aes(fill = condition,
y = means * scale, # does not work, since scale is not found
x = condition))
p + geom_bar(position = "dodge", stat = "identity")
}
A very nice workaround for this case is provided by Winston Chang in the referenced GitHub issue: Explicitly setting the environment parameter to the current environment during the call to ggplot.
Here's what that would look like for the above example:
testplot <- function(meansdf)
{
scale <- 0.5
p <- ggplot(meansdf,
aes(fill = condition,
y = means * scale,
x = condition),
environment = environment()) # This is the only line changed / added
p + geom_bar(position = "dodge", stat = "identity")
}
## Now, the following works
testplot(means)
Here is a simple trick I use a lot to define my variables in my functions environment (second line):
FUN <- function(fun.data, fun.y) {
fun.data$fun.y <- fun.data[, fun.y]
ggplot(fun.data, aes(x, fun.y)) +
geom_point() +
scale_y_continuous(fun.y)
}
datas <- data.frame(x = rnorm(100, 0, 1),
y = x + rnorm(100, 2, 2),
z = x + rnorm(100, 5, 10))
FUN(datas, "y")
FUN(datas, "z")
Note how the y-axis label also changes when different variables or data-sets are used.
I don't think you need to include the meansdf$ part in your function call itself. This seems to work on my machine:
meansdf <- data.frame(means = c(13.8, 14.8), condition = 1:2)
testplot <- function(meansdf)
{
p <- ggplot(meansdf, aes(fill=condition, y=means, x = condition))
p + geom_bar(position="dodge", stat="identity")
}
testplot(meansdf)
to produce:
This is an example of a problem that is discussed earlier. Basically, it comes down to ggplot2 being coded for use in the global environment mainly. In the aes() call, the variables are looked for either in the global environment or within the specified dataframe.
library(ggplot2)
means <- data.frame(means=c(13.8,14.8),condition=1:2)
testplot <- function(meansdf)
{
p <- ggplot(meansdf, aes(fill=condition,
y=means, x = condition))
p + geom_bar(position="dodge", stat="identity")
}
EDIT:
update: After seeing the other answer and updating the ggplot2 package, the code above works. Reason is, as explained in the comments, that ggplot will look for the variables in aes in either the global environment (when the dataframe is specifically added as meandf$... ) or within the mentioned environment.
For this, be sure you work with the latest version of ggplot2.
If is important to pass the variables (column names) to the custom plotting function unquoted, while different variable names are used within the function, then another workaround that I tried, was to make use of match.call() and eval (like here as well):
library(ggplot2)
meansdf <- data.frame(means = c(13.8, 14.8), condition = 1:2)
testplot <- function(df, x, y) {
arg <- match.call()
scale <- 0.5
p <- ggplot(df, aes(x = eval(arg$x),
y = eval(arg$y) * scale,
fill = eval(arg$x)))
p + geom_bar(position = "dodge", stat = "identity")
}
testplot(meansdf, condition, means)
Created on 2019-01-10 by the reprex package (v0.2.1)
Another workaround, but with passing quoted variables to the custom plotting function is using get():
meansdf <- data.frame(means = c(13.8, 14.8), condition = 1:2)
testplot <- function(df, x, y) {
scale <- 0.5
p <- ggplot(df, aes(x = get(x),
y = get(y) * scale,
fill = get(x)))
p + geom_bar(position = "dodge", stat = "identity")
}
testplot(meansdf, "condition", "means")
Created on 2019-01-10 by the reprex package (v0.2.1)
This frustrated me for some time. I wanted to send different data frames with different variable names and I wanted the ability to plot different columns from the data frame. I finally got a work around by creating some dummy (global) variables to handle plotting and forcing assignment inside the function
plotgraph function(df,df.x,df.y) {
dummy.df <<- df
dummy.x <<- df.x
dummy.y <<- df.y
p = ggplot(dummy.df,aes(x=dummy.x,y=dummy.y,.....)
print(p)
}
then in the main code I can just call the function
plotgraph(data,data$time,data$Y1)
plotgraph(data,data$time,data$Y2)
Short answer: Use qplot
Long answer:
In essence you want something like this:
my.barplot <- function(x=this.is.a.data.frame.typically) {
# R code doing the magic comes here
...
}
But that lacks flexibility because you must stick to consistent column naming to avoid the annoying R scope idiosyncrasies. Of course the next logic step is:
my.barplot <- function(data=data.frame(), x=..., y....) {
# R code doing something really really magical here
...
}
But then that starts looking suspiciously like a call to qplot(), right?
qplot(data=my.data.frame, x=some.column, y=some.other column,
geom="bar", stat="identity",...)
Of course now you'd like to change things like scale titles but for that a function comes handy... the good news is that scoping issues are mostly gone.
my.plot <- qplot(data=my.data.frame, x=some.column, y=some.other column,...)
set.scales(p, xscale=scale_X_continuous, xtitle=NULL,
yscale=scale_y_continuous(), title=NULL) {
return(p + xscale(title=xtitle) + yscale(title=ytitle))
}
my.plot.prettier <- set.scale(my.plot, scale_x_discrete, 'Days',
scale_y_discrete, 'Count')
Another workaround is to define the aes(...) as a variable of your function :
func<-function(meansdf, aes(...)){}
This just worked fine for me on a similar topic
You don't need anything fancy. Not even dummy variables. You only need to add a print() inside your function, is like using cat() when you want something to show in the console.
myplot <- ggplot(......) + Whatever you want here
print(myplot)
It worked for me more than one time inside the same function
I just generate new data frame variables with the desired names inside the function:
testplot <- function(df, xVar, yVar, fillVar) {
df$xVar = df[,which(names(df)==xVar)]
df$yVar = df[,which(names(df)==yVar)]
df$fillVar = df[,which(names(df)==fillVar)]
p <- ggplot(df,
aes(x=xvar, y=yvar, fill=fillvar)) +
geom_bar(position="dodge", stat="identity")
}