Error while building decision tree using C5.0 in R language

Question

Here, I am predicting the type of glass based on its chemical content.chemical contents are input which are of numeric type(Ri,Na etc) and type of glass is output(Typec) which is factor.After building model using C5.0 I am getting error like

"Error: Tree models only"

This is my code:

train=datax[sample(nrow(datax),80,replace=F),]

test=datax[!(1:nrow(datax))%in%as.numeric(row.names(train)),]

model=C5.0(Typec~.,data=train,rules=T, ntree= 50)

plot(model)

Here after running the plot instruction Im getting error mentioned.

What is the error? Are you sure the model was built successfully? Any errors or warning before the `plot()` command? — AntoniosK, Aug 24 '18 at 09:57
Can you please post a sample of your data? Like `dput(datax)`. — RLave, Aug 24 '18 at 10:00
@AntoniosK ya there was no error while building model and there were no warning before plot instruction.And i got 0.83 of accuracy in confusion matrix — Sree Lakshmi K M, Aug 24 '18 at 11:22
Is every value in `Typec` "y"? Can you check with `table(train$Typec)`? — RLave, Aug 24 '18 at 11:34
FYI that's not very useful data for us, paste here the output from `dput(head(train))` — RLave, Aug 24 '18 at 11:35
Try to build and plot the model without `rules=T`. Check more info here: https://cran.r-project.org/web/packages/C50/vignettes/C5.0.html where it mentions that rules based models have no plot methods. — AntoniosK, Aug 24 '18 at 11:35

score 0 · Answer 1 · edited Mar 24 '20 at 21:32

0

model=C5.0(Typec~.,data=train,rules=T, ntree= 50)

Remove "rules = T" argument and then you'll be able to plot the graph

edited Mar 24 '20 at 21:32

Paul Roub

36,322
27
84
93

answered Mar 24 '20 at 21:14

Amanpreet Singh

1

But will it be the desired model? – skan May 18 '21 at 19:05
Anyway I'm getting the error "'list' object cannot be coerced to type 'double'" – skan May 18 '21 at 19:48

Error while building decision tree using C5.0 in R language

1 Answers1