Data Visualization — Bivariate Analysis in R
--
I am working on R Programming as part of my course module. I was trying to visualize Boxplots for all feature variables against target variable in one plot. Initially I tried to find out an article or video on how to do that but I didn’t get complete information anywhere. Hence I started to figure it out from very basics.
Let us take the example of Iris dataset
#Loading the datasets package
library(datasets)
#Loading Iris Dataset
data(iris)
#to check summary of dataset
summary(iris)
Feature variables are Sepal.Length, Sepal.Width, Petal.Length, Petal.Width and target variable is Species.
I know how to plot boxplot for a feature variable against target and here is how to do that.
Species_boxplot = ggplot(iris, aes(x = Species, y = Sepal.Length)) + geom_boxplot(aes(fill=Species))
Species_boxplot
Now to implement it for all the other features, I just added all the features to Species_boxplot variable
Species_boxplot = ggplot(iris, aes(x = Species, y = Sepal.Length)) + geom_boxplot(aes(fill=Species)) +
ggplot(iris, aes(x=Species, y = Sepal.Width)) + geom_boxplot(aes(fill=Species)) +
ggplot(iris, aes(x=Species, y = Petal.Length)) + geom_boxplot(aes(fill=Species))+
ggplot(iris, aes(x=Species, y = Petal.Width)) + geom_boxplot(aes(fill=Species))
#After running above code in RStudio, run command below to view the plots.
Species_boxplot
Oops! I got the error below —
Error: Can’t add `ggplot(iris, aes(x = Species, y = Sepal.Width))` to a ggplot object.
Run `rlang::last_error()` to see where the error occurred.
> rlang::last_error()
<error/rlang_error>
Can’t add `ggplot(iris, aes(x = Species, y = Sepal.Width))` to a ggplot object.
Backtrace:
1. ggplot2:::`+.gg`(…)
2. ggplot2:::add_ggplot(e1, e2, e2name)
4. ggplot2:::ggplot_add.default(object, p, objectname)
Run `rlang::last_trace()` to see the full context.
To solve this error, just load the library patchwork
library(patchwork)
Run the code again, you will be able to see boxplots for all feature variables against target.
I am pretty sure that there will be an alternative to do this, but I as tried this on my own, I thought of sharing.
Thank you and have a good day :)
