Plotting varImp in R

Question

Im using the varImp function of the caret package and Im trying to plot the resulting dataframe that it creates. Here is the code:

RocImp2 <- varImp(svmFit, scale = FALSE)

and Im using a simple plot to plot the chart:

plot(RocImp2)

The results are supposed to look like this:

whereas mine looks like this:

Here is a copy of the dataframe:

disregard the names in the plot as they are from another example.

You should provide a [reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example). Include all data and code necessary to reproducible the plot. Do not include images of data, include actual data. — MrFlick, Mar 25 '16 at 22:01

score 5 · Answer 1 · edited Nov 20 '19 at 12:34

5

My issue was fixed by adding the following.

ggplot2::ggplot(varImp())

edited Nov 20 '19 at 12:34

Adam McMahon

765
3
24

answered Nov 20 '19 at 12:18

Lucas

91
1
5

score 3 · Answer 2 · answered May 31 '20 at 05:10

here is a solution to plot the varImp in the case of a binary classification model.

model = glm(Survived~.,family="binomial", data=Titanic)

V = caret::varImp(model)

ggplot2::ggplot(V, aes(x=reorder(rownames(V),Overall), y=Overall)) +
geom_point( color="blue", size=4, alpha=0.6)+
geom_segment( aes(x=rownames(V), xend=rownames(V), y=0, yend=Overall), 
color='skyblue') +
xlab('Variable')+
ylab('Overall Importance')+
theme_light() +
coord_flip()

score -1 · Answer 3 · edited May 31 '18 at 12:32

-1

I found a workaround this problem. You could try:

write.csv(RocImp2, file = 'RocImp2.csv')

VIMP <- read.csv('RocImp2.csv')

plot(VIMP)

edited May 31 '18 at 12:32

Machavity

30,841
27
92
100

answered May 31 '18 at 12:30

Milad Botros

1
2

Plotting varImp in R

3 Answers3