tf.raw_ops.CollectiveReduceV2
Stay organized with collections
Save and categorize content based on your preferences.
Mutually reduces multiple tensors of identical type and shape.
tf.raw_ops.CollectiveReduceV2(
input,
group_size,
group_key,
instance_key,
ordering_token,
merge_op,
final_op,
communication_hint='auto',
timeout_seconds=0,
max_subdivs_per_device=-1,
name=None
)
Args |
|---|
input
|
A Tensor. Must be one of the following types: bfloat16, float32, half, float64, int32, int64.
|
group_size
|
A Tensor of type int32.
|
group_key
|
A Tensor of type int32.
|
instance_key
|
A Tensor of type int32.
|
ordering_token
|
A list of Tensor objects with type resource.
|
merge_op
|
A string from: "Min", "Max", "Mul", "Add".
|
final_op
|
A string from: "Id", "Div".
|
communication_hint
|
An optional string. Defaults to "auto".
|
timeout_seconds
|
An optional float. Defaults to 0.
|
max_subdivs_per_device
|
An optional int. Defaults to -1.
|
name
|
A name for the operation (optional).
|
Returns |
|---|
A Tensor. Has the same type as input.
|
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. Some content is licensed under the numpy license.
Last updated 2023-10-06 UTC.
[null,null,["Last updated 2023-10-06 UTC."],[],[],null,["# tf.raw_ops.CollectiveReduceV2\n\n\u003cbr /\u003e\n\nMutually reduces multiple tensors of identical type and shape.\n\n#### View aliases\n\n\n**Compat aliases for migration**\n\nSee\n[Migration guide](https://www.tensorflow.org/guide/migrate) for\nmore details.\n\n[`tf.compat.v1.raw_ops.CollectiveReduceV2`](https://www.tensorflow.org/api_docs/python/tf/raw_ops/CollectiveReduceV2)\n\n\u003cbr /\u003e\n\n tf.raw_ops.CollectiveReduceV2(\n input,\n group_size,\n group_key,\n instance_key,\n ordering_token,\n merge_op,\n final_op,\n communication_hint='auto',\n timeout_seconds=0,\n max_subdivs_per_device=-1,\n name=None\n )\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ---- ||\n|--------------------------|-------------------------------------------------------------------------------------------------------------|\n| `input` | A `Tensor`. Must be one of the following types: `bfloat16`, `float32`, `half`, `float64`, `int32`, `int64`. |\n| `group_size` | A `Tensor` of type `int32`. |\n| `group_key` | A `Tensor` of type `int32`. |\n| `instance_key` | A `Tensor` of type `int32`. |\n| `ordering_token` | A list of `Tensor` objects with type `resource`. |\n| `merge_op` | A `string` from: `\"Min\", \"Max\", \"Mul\", \"Add\"`. |\n| `final_op` | A `string` from: `\"Id\", \"Div\"`. |\n| `communication_hint` | An optional `string`. Defaults to `\"auto\"`. |\n| `timeout_seconds` | An optional `float`. Defaults to `0`. |\n| `max_subdivs_per_device` | An optional `int`. Defaults to `-1`. |\n| `name` | A name for the operation (optional). |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ------- ||\n|---|---|\n| A `Tensor`. Has the same type as `input`. ||\n\n\u003cbr /\u003e"]]
