First I am a newbie to R (less than two months) and first time asking a question here so I apologize if I do not include all is needed.
I am trying to create a confusion matrix from a random forest using:
library (randomForest)
library (caret)
rfSeg1 <- randomForest(desc ~ Seg1, data= roi_dataSeg1, importance= TRUE, mtry = 4, ntree= 500, trControl = rfSeg1Control, tuneGrid = rfSeg1Grid,metric = "Kappa", maximize = true, na.action=na.exclude)
conftbl <- confusionMatrix(rfSeg1$predicted,roi_dataSeg1$desc)
I was successful using this with a different input data set however fails with new data inputs. I have to create a table with Sensitivity, Specificity, PPV, NPV etc. by class for each study boundary. I get the following error:
Error in table(data, reference, dnn = dnn, ...) : all arguments must have the same length
I ran:
> identical(levels(rfSeg1$predicted),levels(roi_dataSeg1$desc))
[1] TRUE
> identical(length(rfSeg1$predicted),length(roi_dataSeg1$desc))
[1] FALSE
> length(roi_dataSeg1$desc)
[1] 152242
> length(rfSeg1$predicted)
[1] 151567
I have looked online and tried several things but to no avail. Any help will be greatly appreciated to fix this error.