Mastering Machine Learning with R - Second Edition

Create train and test datasets

You can use H2O's functionality to partition the data into train and test sets. The first thing to do is create a vector of random and uniform numbers for the full data:

    > rand <- h2o.runif(bank, seed = 123)

You can then build your partitioned data and assign it with a desired key name, as follows:

    > train <- bank[rand <= 0.7, ]    > train <- h2o.assign(train, key = "train")    > test <- bank[rand > 0.7, ]    > test <- h2o.assign(test, key = "test")

With these created, it is probably a good idea that we have a balanced response variable between the train and test sets. To do this, you can use the h2o.table() function and, in our case, it would be column 64:

    > h2o.table(train[, 64])        y Count    1  no  2783    2 yes   396