Method similar to train_test_split of sklearn about rumale HOT 4 CLOSED

@lalusaud Thank you for your suggestion and the useful notebook. I would like to implement a module function equivalent to train_test_split in the next version.
By the way, I always obtain random split as follows:

train_ids, test_ids = Rumale::ModelSelection::ShuffleSplit.new(
                        n_splits: 1, test_size: 0.3, random_seed: 1
                      ).split(x, y).first
x_train = x[train_ids, true]
y_train = y[train_ids]
x_test = x[test_ids, true]
y_test = y[test_ids]

from rumale.

@lalusaud I have released version 1.8.0 including train_test_split method. Please confirm that.

https://yoshoku.github.io/rumale/doc/Rumale/ModelSelection.html#train_test_split-class_method

x_train, x_test, y_train, y_test = Rumale::ModelSelection.train_test_split(
  x, y, test_size: 0.4, random_seed: 1
)

from rumale.

Works exactly as it is supposed to! Thank you! 🙏

from rumale.

Awesome! I look forward to the module. Thank you for the cleaner suggestion @yoshoku, I will update my notebook code.

from rumale.