trnx = read.table("ass1/a1trn2x")
tstx = read.table("ass1/a1tst2x")
trny = scan("ass1/a1trn2y")
tsty = scan("ass1/a1tst2y")

library(rpart)

cat("Fraction of class 1 in test cases:",mean(tsty),"\n")

cat("\nFITTING TREE MODEL USING FIRST 4 VARIABLES\n\n")

t4 = rpart(
  trny~V1+V2+V3+V4,
  data=trnx, maxsurrogate=0)

print(t4)

cat("\nError rate on test cases:", mean((predict(t4,tstx) > 0.5) != tsty),"\n")

cat("\nFITTING TREE MODEL USING ALL 20 VARIABLES\n\n")

t20 = rpart(
  trny~V1+V2+V3+V4+V5+V6+V7+V8+V9+V10+V11+V12+V13+V14+V15+V16+V17+V18+V19+V20, 
  data=trnx, maxsurrogate=0)

print(t20)

cat("\nError rate on test cases:", mean((predict(t20,tstx) > 0.5) != tsty),"\n")