c50 code called exit with value 1

[moh-kelani]

hi everybody
I need help, please
I face this problem when I run my code

Call:
C5.0.default(x = trainingDataset[, -5], y = trainingDataset$play)
-------------------------------
Class specified by attribute `outcome'
*** line 1 of `undefined.data': bad value of `c(2, 3, 1, 1, 2, 1, 3, 2, 1)' for attribute `outlook'
Error limit exceeded

you can have a look ...

> Classes ‘tbl_df’, ‘tbl’ and 'data.frame':	14 obs. of  5 variables:
 $ outlook : Factor w/ 3 levels "Overcast","Rain",..: 3 3 1 2 2 2 1 3 3 2 ...
 $ temp    : Factor w/ 3 levels "Cool","Hot","Mild": 2 2 2 3 1 1 1 3 1 3 ...
 $ humidity: Factor w/ 2 levels "High","Normal": 1 1 1 1 2 2 2 1 2 2 ...
 $ wind    : Factor w/ 2 levels "Strong","Weak": 2 1 2 2 2 1 1 2 2 2 ...
 $ play    : Factor w/ 2 levels "No","Yes": 1 1 2 2 2 1 2 1 2 2 ...

and this is my code

library(C50)
library(caret)
library("readxl")
require(class)
library(rpart)
library(data.tree)
library(dplyr)
library(Cubist)
tennis <- read_excel(file.choose())
dataset=tennis
#tennis =read.csv(file.choose(), header = T, stringsAsFactors = TRUE)
dataset = tennis[,-1]
dataset = mutate(dataset,outlook=as.factor(outlook),temp=as.factor(temp),humidity=as.factor(humidity),wind=as.factor(wind),play=as.factor(play) )
View(dataset)
str(dataset)
summary(dataset)
View(dataset)

#dataset$play <- as.factor(dataset$play)
#dataset$wind <- as.factor(dataset$wind)
#dataset$humidity <- as.factor(dataset$humidity)
#dataset$temp <- as.factor(dataset$temp)
#dataset$outlook <- as.factor(dataset$outlook)

str(dataset)
set.seed(123)
shuffledDataset = dataset[order(runif(nrow(dataset))),]
View(shuffledDataset)
trainingDataset = shuffledDataset[1:9,] #floor(nrow(shuffledDataset)*0.7),]
testingDataset =  shuffledDataset[10:14,] #(floor(nrow(shuffledDataset)*0.7)+1): nrow(shuffledDataset),]

View(trainingDataset)
View(testingDataset)
str(trainingDataset)
str(testingDataset)

model = C5.0(trainingDataset[,-5],trainingDataset$play)
summary(model)
plot(model,type = "s",main = "Decision Tree")
results = predict(model, testingDataset[,-5])
View(cbind(testingDataset, results))
#tmp = results == testingDataset
#tmp
#which(tmp)
#length(which(tmp))
#length(tmp)
#accuracy = length(which(tmp)) / length(tmp) * 100.0
#sprintf("The Accuracy = %2f", accuracy)
table(results,testingDataset$play)
confusionMatrix(results,testingDataset$play)

modelRules <- C5.0(trainingDataset[,-5],trainingDataset$play, rules = TRUE)
summary(modelRules) # view the ruleset`

[tennis2.xlsx](https://github.com/topepo/C5.0/files/3985763/tennis2.xlsx)

[moh-kelani]

please .. anyone can help?

[topepo]

Sorry for the delay. I was able to run the code without an error. The data set didn't have any splits or rules. This means that none of the predictors were informative.

Can you try to reproduce (perhaps using reprex::reprex()) with the current version?

[topepo]

I think that this is solved by installing the current devel version of Cubist.