CODEWITHSUNDEEP
r
JR0428
JR0428

Reputation: 1

Error in table: all arguments must have the same length

I am trying to use naiveBayes to fit and classify a data with binary (location=3 or not) outcome and assess its performance on test data by resampling.

I got an error code: Error in table(train$prog, trainPred) : all arguments must have the same length

Here is my reproducible code:

wifiDat <- read.table("wifi_localization.txt",sep="\t")
colnames(wifiDat) 
dim(wifiDat)
summary(wifiDat)
head(wifiDat)
colnames(wifiDat) <- c("WF1", "WF2", "WF3", "WF4", "WF5", "WF6",  "WF7", "RM")
wifiDat$RM <- factor(wifiDat$RM)
library(e1071)
wifiDat$RM <- factor(wifiDat$RM)
set.seed(1)
row.number = sample(1:nrow(wifiDat), 0.25*nrow(wifiDat))
train = wifiDat[row.number,]
test = wifiDat[-row.number,]
NBclassfier=naiveBayes(RM~WF1+WF2+WF3+WF4+WF5+WF6+WF7, data=train)
print(NBclassfier)
printALL=function(model){
trainPred=predict(model, newdata = train, type = "class")
trainTable=table(train$prog, trainPred)
testPred=predict(NBclassfier, newdata=test, type="class")
testTable=table(test$prog, testPred)
trainAcc=(trainTable[1,1]+trainTable[2,2]+trainTable[3,3])/sum(trainTable)
testAcc=(testTable[1,1]+testTable[2,2]+testTable[3,3])/sum(testTable)
message("Contingency Table for Training Data")
print(trainTable)
message("Contingency Table for Test Data")
print(testTable)
message("Accuracy")
print(round(cbind(trainAccuracy=trainAcc, testAccuracy=testAcc),3))
}
printALL(NBclassfier)

Here is my sample data

V1. V2. V3. V4. V5. V6. V7. V8
-64 -56 -61 -66 -71 -82 -81 1
-68 -57 -61 -65 -71 -85 -85 2
-63 -60 -60 -67 -76 -85 -84 3
-61 -60 -68 -62 -77 -90 -80 4
-63 -65 -60 -63 -77 -81 -87 1
-64 -55 -63 -66 -76 -88 -83 1
-65 -61 -65 -67 -69 -87 -84 2
-61 -63 -58 -66 -74 -87 -82 3
-65 -60 -59 -63 -76 -86 -82 4
-62 -60 -66 -68 -80 -86 -91 1
-67 -61 -62 -67 -77 -83 -91 1
-65 -59 -61 -67 -72 -86 -81 1
-63 -57 -61 -65 -73 -84 -84 1

Please how can I fix the error?

Upvotes: 0

Views: 2020

Answers (1)

stragu
stragu

Reputation: 1291

If you look at the traceback of your error, you should see that the issue lies in the table(train$prog, trainPred) part of your function. The base::table() function requires you to provide objects of the same length for the ... argument, as

Stepping through your custom function, you'll see that the length of trainPred is 3, and train$prog doesn't exist (and all columns of train are of length 10 anyway). Hence the error.

Are you missing a step somewhere?

Upvotes: 1

Related Questions