Warning: This API is deprecated and will be removed in a future
version of TensorFlow after
the replacement is stable.
XlaSendTPUEmbeddingGradients
Stay organized with collections
Save and categorize content based on your preferences.
An op that performs gradient updates of embedding tables.
The gradients argument is a TensorList having the same length and shapes as the
return value of XlaRecvTPUEmbeddingActivations, but contains gradients of the
model's loss with respect to the embedding activations. The embedding tables are
updated from these gradients via the optimizer specified in the
TPUEmbeddingConfiguration proto given to tpu.initialize_system.
Inherited Methods
From class
java.lang.Object
|
boolean
|
equals(Object arg0)
|
|
final
Class<?>
|
getClass()
|
|
int
|
hashCode()
|
|
final
void
|
notify()
|
|
final
void
|
notifyAll()
|
|
String
|
toString()
|
|
final
void
|
wait(long arg0, int arg1)
|
|
final
void
|
wait(long arg0)
|
|
final
void
|
wait()
|
Public Methods
public
static
XlaSendTPUEmbeddingGradients
create
(Scope scope, Iterable<Operand<Float>> gradients, Iterable<Operand<Float>> learningRates, Operand<?> deduplicationData, String config)
Factory method to create a class wrapping a new XlaSendTPUEmbeddingGradients operation.
Parameters
| scope |
current scope |
| gradients |
A TensorList of gradients with which to update embedding tables. |
| learningRates |
A TensorList of learning rates used for updating the embedding
tables via the optimizer. The length of the TensorList must be equal to the
number of dynamic learning rate tags specified in the
TPUEmbeddingConfiguration proto. |
| deduplicationData |
A Tensor with type=DT_VARIANT containing the deduplication
data. The tensor is an XLA nested tuple containing N elements (where N is
the ratio of the number of embedding to tensor cores per TPU chip). Each
element of the nested tuple is a tuple of rank 1 tensors. Each tensor either
contains indices (DT_UINT32) for embedding lookup on the TensorCore or
weights (DT_FLOAT) to apply to the output of the embedding lookup operation. |
| config |
Serialized TPUEmbeddingConfiguration proto. |
Returns
- a new instance of XlaSendTPUEmbeddingGradients
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 2022-09-07 UTC.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2022-09-07 UTC."],[],[],null,["# XlaSendTPUEmbeddingGradients\n\npublic final class **XlaSendTPUEmbeddingGradients** \nAn op that performs gradient updates of embedding tables.\n\n\nThe gradients argument is a TensorList having the same length and shapes as the\nreturn value of XlaRecvTPUEmbeddingActivations, but contains gradients of the\nmodel's loss with respect to the embedding activations. The embedding tables are\nupdated from these gradients via the optimizer specified in the\nTPUEmbeddingConfiguration proto given to tpu.initialize_system.\n\n\u003cbr /\u003e\n\n### Public Methods\n\n|-----------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| static [XlaSendTPUEmbeddingGradients](/api_docs/java/org/tensorflow/op/core/XlaSendTPUEmbeddingGradients) | [create](/api_docs/java/org/tensorflow/op/core/XlaSendTPUEmbeddingGradients#create(org.tensorflow.op.Scope,%20java.lang.Iterable\u003corg.tensorflow.Operand\u003cjava.lang.Float\u003e\u003e,%20java.lang.Iterable\u003corg.tensorflow.Operand\u003cjava.lang.Float\u003e\u003e,%20org.tensorflow.Operand\u003c?\u003e,%20java.lang.String))([Scope](/api_docs/java/org/tensorflow/op/Scope) scope, Iterable\\\u003c[Operand](/api_docs/java/org/tensorflow/Operand)\\\u003cFloat\\\u003e\\\u003e gradients, Iterable\\\u003c[Operand](/api_docs/java/org/tensorflow/Operand)\\\u003cFloat\\\u003e\\\u003e learningRates, [Operand](/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e deduplicationData, String config) Factory method to create a class wrapping a new XlaSendTPUEmbeddingGradients operation. |\n\n### Inherited Methods\n\nFrom class [org.tensorflow.op.PrimitiveOp](/api_docs/java/org/tensorflow/op/PrimitiveOp) \n\n|------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------|\n| final boolean | [equals](/api_docs/java/org/tensorflow/op/PrimitiveOp#equals(java.lang.Object))(Object obj) |\n| final int | [hashCode](/api_docs/java/org/tensorflow/op/PrimitiveOp#hashCode())() |\n| [Operation](/api_docs/java/org/tensorflow/Operation) | [op](/api_docs/java/org/tensorflow/op/PrimitiveOp#op())() Returns the underlying [Operation](/api_docs/java/org/tensorflow/Operation) |\n| final String | [toString](/api_docs/java/org/tensorflow/op/PrimitiveOp#toString())() |\n\nFrom class java.lang.Object \n\n|------------------|---------------------------|\n| boolean | equals(Object arg0) |\n| final Class\\\u003c?\\\u003e | getClass() |\n| int | hashCode() |\n| final void | notify() |\n| final void | notifyAll() |\n| String | toString() |\n| final void | wait(long arg0, int arg1) |\n| final void | wait(long arg0) |\n| final void | wait() |\n\nPublic Methods\n--------------\n\n#### public static [XlaSendTPUEmbeddingGradients](/api_docs/java/org/tensorflow/op/core/XlaSendTPUEmbeddingGradients)\n**create**\n([Scope](/api_docs/java/org/tensorflow/op/Scope) scope, Iterable\\\u003c[Operand](/api_docs/java/org/tensorflow/Operand)\\\u003cFloat\\\u003e\\\u003e gradients, Iterable\\\u003c[Operand](/api_docs/java/org/tensorflow/Operand)\\\u003cFloat\\\u003e\\\u003e learningRates, [Operand](/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e deduplicationData, String config)\n\nFactory method to create a class wrapping a new XlaSendTPUEmbeddingGradients operation. \n\n##### Parameters\n\n| scope | current scope |\n| gradients | A TensorList of gradients with which to update embedding tables. |\n| learningRates | A TensorList of learning rates used for updating the embedding tables via the optimizer. The length of the TensorList must be equal to the number of dynamic learning rate tags specified in the TPUEmbeddingConfiguration proto. |\n| deduplicationData | A Tensor with type=DT_VARIANT containing the deduplication data. The tensor is an XLA nested tuple containing N elements (where N is the ratio of the number of embedding to tensor cores per TPU chip). Each element of the nested tuple is a tuple of rank 1 tensors. Each tensor either contains indices (DT_UINT32) for embedding lookup on the TensorCore or weights (DT_FLOAT) to apply to the output of the embedding lookup operation. |\n| config | Serialized TPUEmbeddingConfiguration proto. |\n|-------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n\n##### Returns\n\n- a new instance of XlaSendTPUEmbeddingGradients"]]
