Stratified splitting train/test - H2O

user_11616033 0
Viewed 440

I use H2O to build and compare models but I wanted to know if there is a feature that allows me to split the data into train and test in a stratified way?

I know that for cross validation the option fold_assignment exist and I use it.

My question is more on the manipulation of the data and when it needs to be split before any model is built.

I had a look a this page but it does not specify if stratified splitting is done for classification.

Maybe for classification problems it is implicit and the stratified splitting is always achieved?

h2o machine-learning
1 Answers

Split the data into Train/Test/Validation with Train having 70% and test and validation 15% each

train,test,valid = prostate_df.split_frame(ratios=(.7, .15))
user_34864600
Related