tf.contrib.data.sliding_window_batch

View source on GitHub

A sliding window over a dataset. (deprecated) (deprecated arguments)

This transformation passes a sliding window over this dataset. The window size is window_size, the stride of the input elements is window_stride, and the shift between consecutive windows is window_shift. If the remaining elements cannot fill up the sliding window, this transformation will drop the final smaller element. For example:

# NOTE: The following examples use `{ ... }` to represent the
# contents of a dataset.
a = { [1], [2], [3], [4], [5], [6] }

a.apply(sliding_window_batch(window_size=3)) ==
{ [[1], [2], [3]], [[2], [3], [4]], [[3], [4], [5]], [[4], [5], [6]] }

a.apply(sliding_window_batch(window_size=3, window_shift=2)) ==
{ [[1], [2], [3]], [[3], [4], [5]] }

a.apply(sliding_window_batch(window_size=3, window_stride=2)) ==
{ [[1], [3], [5]], [[2], [4], [6]] }

window_size A tf.int64 scalar tf.Tensor, representing the number of elements in the sliding window. It must be positive.
stride (Optional.) A tf.int64 scalar tf.Tensor, representing the forward shift of the sliding window in each iteration. The default is 1. It must be positive. Deprecated alias for window_shift.
window_shift (Optional.) A tf.int64 scalar tf.Tensor, representing the forward shift of the sliding window in each iteration. The default is 1. It must be positive.
window_stride (Optional.) A tf.int64 scalar tf.Tensor, representing the stride of the input elements in the sliding window. The default is 1. It must be positive.

A Dataset transformation function, which can be passed to tf.data.Dataset.apply.

ValueError if invalid arguments are provided.