-1

I'm trying to analyse some tennis data and I'm hitting a problem with the code:

library(rpart)
library(rpart.plot)
library(ggplot2)
library(wesanderson)

train=read.csv("/ags_test.csv",header=T, na.strings=c("","NA"))

Please note this is a complete set, not one I've cobbled together through the code. All the gaps have NA values in them.

control=rpart.control(cp=0.007)
train$res=as.factor(train$res)
tree=rpart(res~Tournament+Surface+Round+J1Rank+J2Rank+J1Pts+J2Pts+DRank+DPts,data=train,control=control,parms=list(split="gini"))

All good until the last line when it kicks out:

Error in cbind(yval2, yprob, nodeprob) : 
  number of rows of matrices must match (see arg 2)

The data isn't a massive set but comprises of 17 columns and 50 lines.

Any ideas would be much appreciated.

Foothill_trudger
  • 73
  • 7

1 Answers1

0

Turns out that the problem the data is too certain, i.e. the pros are all in the same columns and the cons in a similar structure.

Therefore, there's little to run the decision tree against.

Foothill_trudger
  • 73
  • 7