# tf.raw_ops.RandomIndexShuffle

<br />

Outputs the position of `value` in a permutation of \[0, ..., max_index\].

#### View aliases


**Compat aliases for migration**

See
[Migration guide](https://www.tensorflow.org/guide/migrate) for
more details.

[`tf.compat.v1.raw_ops.RandomIndexShuffle`](https://www.tensorflow.org/api_docs/python/tf/raw_ops/RandomIndexShuffle)

<br />

    tf.raw_ops.RandomIndexShuffle(
        index, seed, max_index, rounds=4, name=None
    )

Output values are a bijection of the `index` for any combination and `seed` and `max_index`.

If multiple inputs are vectors (matrix in case of seed) then the size of the
first dimension must match.

The outputs are deterministic.

<br />

<br />

<br />

| ## Args ||
|---|---|
| `index` | A `Tensor`. Must be one of the following types: `int32`, `uint32`, `int64`, `uint64`. A scalar tensor or a vector of dtype `dtype`. The index (or indices) to be shuffled. Must be within \[0, max_index\]. |
| `seed` | A `Tensor`. Must be one of the following types: `int32`, `uint32`, `int64`, `uint64`. A tensor of dtype `Tseed` and shape \[3\] or \[n, 3\]. The random seed. |
| `max_index` | A `Tensor`. Must have the same type as `index`. A scalar tensor or vector of dtype `dtype`. The upper bound(s) of the interval (inclusive). |
| `rounds` | An optional `int`. Defaults to `4`. The number of rounds to use the in block cipher. |
| `name` | A name for the operation (optional). |

<br />

<br />

<br />

<br />

| ## Returns ||
|---|---|
| A `Tensor`. Has the same type as `index`. ||

<br />