Generate integers encoded categoricals

mlbenchm-py-int-encoded.R

## generate integer-encoded categoricals
 
for SIZE in 1; do
 
time R --vanilla --quiet << EOF
 
library(data.table)
 
d1 <- as.data.frame(fread("train-${SIZE}m.csv"))
d2 <- as.data.frame(fread("test.csv"))
d <- rbind(d1,d2)
 
for (k in c("Month","DayofMonth","DayOfWeek","UniqueCarrier","Origin","Dest")) {
  d[,k] <- as.numeric(as.factor(d[,k]))-1
}
d[["dep_delayed_15min"]] <- ifelse(d[["dep_delayed_15min"]]=="Y",1,0)
 

dd1 <- d[1:nrow(d1),]
dd2 <- d[(nrow(d1)+1):(nrow(d1)+nrow(d2)),]
 
write.table(dd1, "train-intcateg-${SIZE}m.csv", row.names=FALSE, col.names=FALSE, sep=",")
write.table(dd2, "test-intcateg-${SIZE}m.csv", row.names=FALSE, col.names=FALSE, sep=",")
 
EOF
 
done