tf.raw_ops.NcclAllReduce
Stay organized with collections
Save and categorize content based on your preferences.
Outputs a tensor containing the reduction across all input tensors.
tf.raw_ops.NcclAllReduce(
input, reduction, num_devices, shared_name, name=None
)
Outputs a tensor containing the reduction across all input tensors passed to ops
within the same `shared_name.
The graph should be constructed so if one op runs with shared_name value c,
then num_devices ops will run with shared_name value c. Failure to do so
will cause the graph execution to fail to complete.
input: the input to the reduction
data: the value of the reduction across all num_devices devices.
reduction: the reduction operation to perform.
num_devices: The number of devices participating in this reduction.
shared_name: Identifier that shared between ops of the same reduction.
Args |
|---|
input
|
A Tensor. Must be one of the following types: half, float32, float64, int32, int64.
|
reduction
|
A string from: "min", "max", "prod", "sum".
|
num_devices
|
An int.
|
shared_name
|
A string.
|
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.NcclAllReduce\n\n\u003cbr /\u003e\n\nOutputs a tensor containing the reduction across all input tensors.\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.NcclAllReduce`](https://www.tensorflow.org/api_docs/python/tf/raw_ops/NcclAllReduce)\n\n\u003cbr /\u003e\n\n tf.raw_ops.NcclAllReduce(\n input, reduction, num_devices, shared_name, name=None\n )\n\nOutputs a tensor containing the reduction across all input tensors passed to ops\nwithin the same \\`shared_name.\n\nThe graph should be constructed so if one op runs with shared_name value `c`,\nthen `num_devices` ops will run with shared_name value `c`. Failure to do so\nwill cause the graph execution to fail to complete.\n\ninput: the input to the reduction\ndata: the value of the reduction across all `num_devices` devices.\nreduction: the reduction operation to perform.\nnum_devices: The number of devices participating in this reduction.\nshared_name: Identifier that shared between ops of the same reduction.\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: `half`, `float32`, `float64`, `int32`, `int64`. |\n| `reduction` | A `string` from: `\"min\", \"max\", \"prod\", \"sum\"`. |\n| `num_devices` | An `int`. |\n| `shared_name` | A `string`. |\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"]]
