Splits a dataset into a left half and a right half (e.g. train / test).

Usage

split_dataset(
  dataset,
  left_size = NULL,
  right_size = NULL,
  shuffle = FALSE,
  seed = NULL,
  preferred_backend = NULL
)

Arguments

dataset: A tf$data$Dataset, a torch$utils$data$Dataset object, or a list of arrays with the same length.
left_size: If float (in the range [0, 1]), it signifies the fraction of the data to pack in the left dataset. If integer, it signifies the number of samples to pack in the left dataset. If NULL, defaults to the complement to right_size. Defaults to NULL.
right_size: If float (in the range [0, 1]), it signifies the fraction of the data to pack in the right dataset. If integer, it signifies the number of samples to pack in the right dataset. If NULL, defaults to the complement to left_size. Defaults to NULL.
shuffle: Boolean, whether to shuffle the data before splitting it.
seed: A random seed for shuffling.
preferred_backend: Optional string specifying which backend to use, such as "tensorflow" or "torch". If NULL, the backend is inferred from dataset: TensorFlow is used for a tf.data.Dataset, Torch is used for a torch.utils.data.Dataset, and the current Keras backend is used for an R list or array. Defaults to NULL.

Value

A list of two dataset objects, the left and right splits. The exact type depends on preferred_backend and on the input dataset type. For example, the TensorFlow backend returns tf.data.Dataset objects and the Torch backend returns torch.utils.data.Dataset objects.

Examples

data <- random_uniform(c(1000, 4))
c(left_ds, right_ds) %<-% split_dataset(list(data$numpy()), left_size = 0.8)
left_ds$cardinality()

## tf.Tensor(800, shape=(), dtype=int64)

right_ds$cardinality()

## tf.Tensor(200, shape=(), dtype=int64)

Splits a dataset into a left half and a right half (e.g. train / test).

Usage

Arguments

Value

Examples

See also