chainer.datasets.TupleDataset¶
-
class
chainer.datasets.
TupleDataset
(*datasets)[source]¶ Dataset of tuples from multiple equal-length datasets.
A
TupleDataset
combines multiple equal-length datasets into a single dataset of tuples. Thei
-th tuple contains thei
-th example from each of the argument datasets, in the same order that they were supplied.Recall that in Chainer, a dataset is defined as an iterable that supports both
__getitem__
and__len__
. The__getitem__
method should support indexing by both an integer and a slice.As an example, consider creating a
TupleDataset
from two argument datasetsd1 = [8, 0, 5, 1]
andd2 = [3, 1, 7, 4]
astuple_dataset = TupleDataset(d1, d2)
. Thetuple_dataset
will then contain the examples(8, 3), (0, 1), (5, 7), (1, 4)
. Note that this behavior is similar to that of the built-inzip()
function.Parameters: datasets – Underlying datasets that will be aggregated. Each dataset must be an iterable that implements __getitem__
and__len__
. Thej
-th dataset will be used for thej
-th item of each example tuple. All datasets must have the same length.Methods