GitHub

tfp.substrates.numpy.stats.iterative_mergesort

Non-recusive mergesort that counts exchanges.

Main aliases

tfp.substrates.numpy.stats.iterative_mergesort(
    y, permutation, name=None
)

Args
`y`	a `Tensor` of shape `[n]` containing values to be sorted.
`permutation`	`Tensor` of shape `[n]` with original ordering.
`name`	Optional Python `str` name for ops created by this method. Default value: `None` (i.e., 'iterative_mergesort').

Returns
`exchanges`	`int32` scalar that counts the number of exchanges required to produce a sorted permutation
`permutation`	and a `tf.int32` Tensor that contains the ordering of y values that are sorted.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2021-04-20 UTC.

Args