dataset_window: Combines input elements into a dataset of windows.

In tfdatasets: Interface to 'TensorFlow' Datasets

Combines input elements into a dataset of windows.

Description

Combines input elements into a dataset of windows.

Usage

dataset_window(dataset, size, shift = NULL, stride = 1, drop_remainder = FALSE)

Arguments

dataset	A dataset
size	representing the number of elements of the input dataset to combine into a window.
shift	epresenting the forward shift of the sliding window in each iteration. Defaults to size.
stride	representing the stride of the input elements in the sliding window.
drop_remainder	representing whether a window should be dropped in case its size is smaller ⁠than window_size⁠.

See Also

Other dataset methods: dataset_batch(), dataset_cache(), dataset_collect(), dataset_concatenate(), dataset_decode_delim(), dataset_filter(), dataset_interleave(), dataset_map(), dataset_map_and_batch(), dataset_padded_batch(), dataset_prefetch(), dataset_prefetch_to_device(), dataset_reduce(), dataset_repeat(), dataset_shuffle(), dataset_shuffle_and_repeat(), dataset_skip(), dataset_take(), dataset_take_while()

tfdatasets documentation built on Sept. 11, 2024, 8:53 p.m.